162 results about "Random search" patented technology

The invention discloses a charging station loss prediction method based on an LSTM deep learning method, and relates to the technical field of charging stations, and the method comprises the following steps: 1, collecting data including real-time power load data, a power grid topological structure and time sequence data; 2, feature engineering: carrying out preprocessing and feature extraction on the data, and extracting expert knowledge features; step 3, establishing an LSTM model in combination with a PPO algorithm; 4, performing hyper-parameter tuning on the LSTM model through a random search method; and step 5, the optimized miniaturized model is then deployed to each charging station, and local real-time energy loss prediction is realized. In order to further improve the prediction accuracy, expert knowledge is creatively integrated in a feature engineering part, multi-dimensional loss feature recognition is carried out, and the performance of the model in processing a complex nonlinear relation is improved by combining the input of an electric power system expert, recognizing factors such as equipment aging and the like.

The invention discloses a production and manufacturing management method and system based on digital twinning, which are used in the field of production and manufacturing management, and the method comprises the following steps: collecting logistics management data in logistics storage and operation data of a production line, and constructing a logistics digital twinning processing model; simulating a logistics management process based on a logistics digital twin processing model, and tracking the positions of goods and transportation tools in real time by using a positioning and identification technology; utilizing a demand prediction model to predict a material demand and a product demand based on the data in the logistics digital twinborn model updated in real time, and adjusting the stock of logistics storage according to a prediction result; and calculating an optimal transportation path by using a path optimization algorithm, making a production and manufacturing plan of logistics storage, and monitoring the production and manufacturing process in real time. According to the bidirectional fast random search tree, a new path can be quickly recalculated, and the accuracy and adaptability of decision making are improved by comprehensively considering various transportation and operation constraints.

The invention discloses a method for predicting the heavy metal adsorption capacity of micro-plastics, which comprises the following steps of: screening heavy metal and micro-plastic adsorption data from literature and professional databases, and arranging the data into a sample data set; performing preprocessing including data cleaning, missing value processing and data standardization on the sample data set; constructing an adsorption capacity prediction model, and performing training and evaluation to determine an optimal prediction model; using random search, Bayesian optimization or grid search to adjust hyper-parameters of the model; and carrying out adsorption capacity prediction on microplastics and heavy metals with unknown adsorption capacities by using the adsorption prediction model. The method can be practically applied to research on adsorption of the heavy metal ions to the micro-plastics, interaction between the heavy metal ions and the micro-plastics can be visually understood, and ecological risks caused by coexistence of the heavy metal ions and the micro-plastics can be evaluated.

The invention relates to the technical field of unmanned aerial vehicle trajectory planning, in particular to a bidirectional tree random search method based on a complex environment node cost function, which comprises the steps of surveying obstacle data in a flight area and modeling, then introducing a target deviation strategy and a bidirectional search tree mechanism, and combining with an improved minimum cost method to obtain a random search result. Alternately expanding the two trees to generate an initial path, and integrating the path length, the obstacle distance and the flight height energy consumption by a cost function; and finally, pruning the initial path, removing redundant nodes, smoothing the trajectory by adopting a cubic spline interpolation method, ensuring curvature continuity, and generating an optimal path meeting the flight performance of the unmanned aerial vehicle. According to the invention, through an improved bidirectional minimum cost fast expansion random tree algorithm, dynamic association of obstacle constraint, target deviation and path cost in unmanned aerial vehicle path planning in a complex environment is represented, and a more efficient search convergence speed and better path quality are obtained. And rapid convergence and global optimization of the track of the unmanned aerial vehicle in a high-dynamic environment are realized.

The invention relates to the technical field of path planning, and provides an unmanned aerial vehicle three-dimensional path planning method and navigation equipment. The method comprises the following steps: S100, acquiring a flight path planning model comprising an environment model and a total cost function; s200, solving the flight path planning model by adopting a honeypot algorithm, and initializing population parameters; s300, when the coefficient vector A meets the condition that A is larger than or equal to 1, a search strategy with random individual disturbance is executed, and otherwise, the positions of the badgers are still updated with Xprey as the reference; s400, when the number of iterations t reaches the maximum number of iterations tmax, outputting Xprey and a fitness value thereof; otherwise, increasing 1 to the number t of iterations, and returning to the step S300. Wherein A is linearly reduced along with the number of iterations, and in the earlier stage of iteration, the algorithm executes a random search strategy for multiple times, so that premature population aggregation is avoided, the ergodicity of badger individuals in a search space is enhanced, and the global search ability is improved.

The invention discloses a robot redundant attitude planning method and system based on position and obstacle constraints, and the method comprises the following steps: generating a planning path of a robot based on a redundant attitude space and a known robot path position point, the planning path being represented by a position space and an attitude space; constructing an approximate key collision plane; calculating a feasible attitude area based on the obstacle information and the redundant attitude space on the approximate key collision plane; estimating a preferential sampling attitude area; continuously sampling in the position space and the attitude space, acquiring a pose node to expand the random search tree until a target node is reached, and acquiring a target path; and iterating the random search tree extension operation to obtain more target paths, and selecting a target path with the lowest total cost from all the obtained target paths. According to the method, the exploration efficiency of redundant attitude planning can be improved, the integrity and safety of the process are ensured, and the method has relatively high applicability.

The invention discloses an 8-bit S box construction method based on an improved genetic algorithm, and the method comprises the following steps: (1) forming a composite chaotic system through Logistic chaotic mapping and Tent chaotic mapping, and generating an initialized population; (2) designing a fitness function by combining the nonlinearity, the difference uniformity and the strict avalanche criterion of the S box; (3) designing crossover operation for ensuring bijection of the S box; (4) designing a reverse search method to search variation points, combining a hill climbing algorithm and a random search method, designing variation operation based on 2N rounds of search, carrying out two-point variation point search in the first N rounds, adopting three-point variation point search in the last N rounds, and designing a table look-up statistical method to further evaluate and optimize the property of the S box; and (5) constructing an 8-bit password S box based on an improved genetic algorithm. The method is based on the improved genetic algorithm, and the 8-bit S box with excellent cryptographic properties can be generated.

The invention discloses an LDPC code coding method. The method comprises the following steps: acquiring an information bit sequence; performing low density parity check (LDPC) coding on the information bit sequence according to a check matrix to obtain a coded bit sequence; wherein the check matrix is determined by a basis matrix, and the basis matrix comprises a basic sub-matrix; the basic sub-matrix is formed by splicing at least two basic unit blocks with circulation or translation. According to the invention, compared with the condition that the shift value Shifting Value of the existing 5G new air interface NR matrix is a random search result and lacks fixed algebraic structure guidance, the basis matrix with the cyclic block or translation block structure designed by the invention can have fixed algebraic structure guidance, the problem that the LDPC lacks practical code design is solved, and the LDPC performance is improved. And the index matrix of the free is difficult to find through random search.

The invention discloses a black box confrontation attack method based on local region disturbance, which can be applied to digital image attack and physical attack, and comprises the following steps: (1) positioning a salient region in an image by adopting a CAM type activation graph, and determining a local region highly sensitive to disturbance in an original image; (2) further grouping the salient regions to obtain sub-regions; (3) designing an objective function of local confrontation disturbance; (4) adding disturbance to the sub-regions; (5) calculating the size of a loss function of the sub-regions after the disturbance is added, and sorting the sub-regions; and (6) selecting an optimal sub-region, and carrying out disturbance of random search according to a target function. And (7) when an iteration stopping condition is satisfied, outputting the generated confrontation sample.

The invention discloses a cancer recurrence probability deep learning prediction system based on multi-modal time sequence characteristics, and relates to the technical field of medical data processing. The method comprises the following steps: constructing a multi-modal time sequence data set by collecting clinical static data, dynamic treatment data and time sequence monitoring data of a patient; missing value filling and abnormal value elimination based on sliding Z-score are carried out on the data to form an effective data set; screening key clinical static features by using a random forest model; constructing a radiotherapy cumulative toxicity model to quantify residual toxicity, and generating dynamic characteristics by combining the dose change rate three-state characteristics; converting the features into a four-dimensional time sequence tensor through a dynamic sliding window; optimizing LSTM hyper-parameters by adopting random search, and training a high-precision prediction model to output a recurrence probability; finally, differentiated clinical intervention is triggered based on risk grading, a'prediction-decision 'closed-loop system is formed, and the initiative and accuracy of cancer recurrence diagnosis and treatment are remarkably improved.

The invention discloses a micro-grid optimization scheduling method based on a distributed bee foraging algorithm. The method comprises the following steps: establishing a micro-grid optimization scheduling objective function; determining micro-grid optimization scheduling related constraint conditions; performing initialization setting on parameters of the distributed bee foraging algorithm; according to the optimization scheduling related constraint conditions, randomly generating an optimization scheduling scheme; evaluating and sorting nectar sources searched by the investigation bees; carrying out local search by the bees and updating the selected nectar source; carrying out local search by worker bees and updating the selected nectar source; the reconnaissance bees generate optional optimal scheduling schemes in a random search mode; the nectar sources searched by the investigation bees and the selected nectar sources are sorted together, and the selected nectar sources used for next iteration are screened out; and outputting a globally optimal solution and a fitness function value of the target function. According to the method, the bee foraging algorithm is improved to adapt to the optimization requirements of the distributed micro-grid, and the efficient solution of the distributed micro-grid optimization scheduling scheme is realized.

The invention relates to the technical field of intelligent vehicles, in particular to an intelligent vehicle automatic driving formation obstacle avoidance control method and device, and the method comprises the steps: constructing an intelligent vehicle formation controller according to the destination data and obstacle data of a target intelligent vehicle formation, so as to control the target intelligent vehicle formation to move towards the destination; establishing an intelligent vehicle formation space-time transformation strategy based on the queue expansion coefficient according to the queue expansion coefficient of the target intelligent vehicle formation and the driving safety boundary; performing dynamic optimization on the intelligent vehicle formation controller by adopting a Levy flight random search logic based on a cuckoo algorithm to obtain an optimized intelligent vehicle formation controller; and controlling the target intelligent vehicle formation to avoid obstacles by using the optimized intelligent vehicle formation controller and the intelligent vehicle formation space-time transformation strategy based on the queue expansion coefficient. Therefore, the problems that when intelligent vehicle formation obstacle avoidance is carried out through a traditional artificial potential field method, local minimum value and obstacle avoidance track jittering are likely to happen, and complex dynamic environments cannot be rapidly adapted are solved.

The invention discloses a vehicle transverse and longitudinal trajectory tracking collaborative optimization control method and system. The method comprises the following steps: establishing a vehicle monorail model and an error calculation module; respectively designing a transverse MPC controller, a longitudinal MPC upper controller and a lower controller based on an acceleration-driving force inverse dynamics model according to the vehicle monorail model; carrying out normalization processing on the transverse error and the yaw angle error which are calculated by the error calculation module, and taking a normalization processing result as a fitness function of a pelican algorithm; establishing a Latin-pelican optimization algorithm fusing population initialization of sine-Latin hypercube sampling, prey position updating of a grade system and a dynamic random search strategy; a fitness function and a Latin-pelican algorithm are utilized to optimize a transverse MPC controller and a longitudinal MPC upper controller respectively, physical mapping of control quantity is achieved in combination with a lower controller based on an acceleration-driving force inverse dynamics model, optimal weight matrix parameters are obtained, and transverse and longitudinal joint trajectory tracking control of the vehicle is completed.

The invention belongs to the technical field of software testing, and discloses a test case generation method and platform for aircraft performance testing. The method comprises the following steps: determining a first input parameter set, and generating a first test case by utilizing global random search; using the first test case to drive the flight management module to perform simulation test, and receiving first flight state data output after the flight management module performs the simulation test; determining a second input parameter set based on the first flight state information, and generating a second test case by using local random search; driving the flight management module to perform a simulation test by using the second test case, and receiving second flight state data output by the flight management module after the simulation test; an out-of-control input parameter is determined based on the second flight state data. Generation and execution of the test cases are optimized through an intelligent algorithm, the generation efficiency of the test cases is improved, the test cost is reduced, meanwhile, the more comprehensive test coverage rate is ensured, and the test reliability is improved.

The invention discloses an unmanned aerial vehicle flight path planning method, and relates to the technical field of unmanned aerial vehicle navigation and control, and the method comprises the steps: guiding a random search tree to rapidly expand towards a target through a target deviation and artificial potential field gravitational potential field negative gradient deviation sampling strategy, achieving the heuristic intelligent sampling of a free space, achieving the rapid and efficient convergence of an RRT * algorithm, and achieving the rapid and efficient detection of the flight path of an unmanned aerial vehicle. And the optimal track is quickly and efficiently converged. Redundant nodes of a feasible track can be deleted, the number of track nodes is reduced, and the voyage is shortened.

The invention discloses an image multi-level threshold segmentation method based on an improved whale optimization algorithm, and belongs to the technical field of image processing. The method comprises the following steps: initializing a population; calculating a target function value, and finding out the optimal whale position Xbest of the population; updating the position of the whale in the population; performing combined mutation operation based on individual fitness values; similarity removal operation is carried out, and similar individuals are judged according to the difference between the target function values and the distances corresponding to the two individuals; the fitness value of the whale is recalculated, the optimal individual Xbest of the population is updated, and an optimal threshold value is obtained; outputting a global optimal threshold, and segmenting the image by using the global optimal threshold; and outputting the segmented image. The method has the beneficial effects that the development capability of the algorithm is improved by adaptively adjusting the parameters # imgabs0 # and # imgabs1 # in the contraction surrounding stage; the problem that the whale in the population in the later iteration stage is not subjected to position updating in the random search stage is solved, and the algorithm is prevented from falling into local optimum.

The present application discloses an encoding method for low-density parity-check (LDPC) codes. The method comprises: acquiring an information bit sequence; and on the basis of a check matrix, performing LDPC encoding on the information bit sequence to obtain an encoded bit sequence, wherein the check matrix is determined by a base matrix, the base matrix comprises a fundamental sub-matrix, and the fundamental sub-matrix is formed by concatenating at least two basic unit blocks with cyclic or translational properties. Compared with existing 5G new radio (NR) matrices of which shifting values are all obtained by random search without guidance from a fixed algebraic structure, the base matrix having cyclic block structures or translational block structures designed in the present application can be guided by a fixed algebraic structure, thereby solving the technical problems of the lack of practical code design for LDPC and the difficulty in finding a cycle-free exponent matrix by means of random search.

The invention provides a new energy automobile cabin abnormal noise detection method and device based on sound image fusion, and belongs to the technical field of noise detection. The method comprises the following steps: synchronously acquiring multi-channel audio data and video data through a multi-channel microphone array and a camera respectively; determining an abnormal noise type and extracting an abnormal noise signal; an abnormal noise positioning position is obtained through an improved space contraction fast method based on random search, namely an improved SRC method; calibrating the extracted abnormal noise positioning position and an image in the video data, and displaying a noise source in the image in the form of a sound field thermodynamic diagram to obtain a sound field diagram after sound-image fusion; according to the method, more efficient and more visual noise anomaly detection of the new energy automobile cabin can be realized, the accuracy of noise source positioning can be remarkably improved, the efficiency of abnormal noise detection of the new energy automobile cabin can be greatly improved, and the method can be widely applied to noise monitoring and fault diagnosis of the new energy automobile cabin.

The invention provides a pulse eddy current pipeline wall thickness detection method based on an RSBO optimized LSTM network, and the method comprises the steps: estimating the thickness of a pipeline through employing a pulse eddy current signal, determining an interval where a measurement signal is reserved through employing a data sample interception length evaluation criterion based on a least square fitting algorithm, inputting the processed signal into an LSTM model, and carrying out the detection of the thickness of the pipeline. And carrying out hyper-parameter setting on the LSTM model by adopting combined optimization based on random search and Bayesian optimization.

The invention belongs to the technical field of force-controlled polishing of robot optical elements, and discloses a pseudo-random polishing track corner fairing and force velocity planning method and system considering surface shape precision and track dynamic characteristic constraints, and all resident points are connected in modes of random search, dead point removal, residual point supplementation and the like to generate a pseudo-random track. Based on the measured surface shape data points and the given residence points, the residence time of each residence point and the product of the residence time and the force are calculated through a residence time algorithm, and the residence time and the product of the residence time and the force are correspondingly converted into the speed and the relation between the speed and the force. And constructing a corner fairing optimization model considering the surface shape precision, establishing the relationship between the surface shape precision and the trajectory offset, and solving the maximum corner error tolerance within the allowable range of the surface shape precision loss by taking the corner error tolerance as an optimization variable and taking the particle swarm optimization as an optimization method. According to the method, the surface shape precision of the optical element can be reduced, the intermediate frequency error can be removed, and the high-quality optical element can be obtained.

The invention relates to a rock CT image mineral identification method based on Latin hypercube random search, and the method comprises the steps: carrying out the CT scanning of a to-be-identified rock, and obtaining a slice gray-scale map of the to-be-identified rock; according to the slice grey-scale map of the to-be-identified rock, obtaining a two-dimensional distribution map of each mineral in a target area of the to-be-identified rock; according to the two-dimensional distribution diagram of each mineral, obtaining the area proportion and the area proportion range of each mineral, and the dispersion index and the dispersion index range of each mineral; determining constraint conditions according to an area proportion and an area proportion range, and constructing a mathematical model based on a loss function of a dispersion index; the mathematical model is solved based on a Latin hypercube random search method, and the gray threshold segmentation interval of each mineral is obtained; and recognizing the mineral components of the rock to be recognized according to the gray threshold segmentation interval of each mineral. According to the invention, the mineral particles in the rock can be accurately identified.

A control method includes encircling a patient appendage with a pressurizable cuff pressurized via a valve, and directing emitted light through a patient appendage to a light sensor. A sensed plethysmogram signal is generated based on the received light level, and the valve is controlled via a closed feedback loop responsive to an error value computed as the difference between the sensed plethysmogram signal and a calibrated setpoint value. An unperturbed error value characteristic is recorded through a first time window, and a random vector perturbation in the closed-loop control space is generated. A perturbed error value characteristic is recorded through a second time window while temporarily adjusting closed-loop control parameters based on the random vector perturbation. Baseline closed-loop control parameters are updated in response to the unperturbed error value characteristic exceeding the perturbed error value characteristic.

The invention discloses a communication sensing time slot allocation method and system based on a potential game. According to the invention, frequency spectrum hardware sharing of radar sensing and wireless communication is realized through system initialization, a time allocation mode is dynamically selected based on a 5G NR frame structure, and the time allocation mode comprises a segmentation mode and an aggregation mode, so that time slot resources are flexibly divided; constructing a non-cooperative potential game model, taking a vehicle as a participant, defining a utility function which takes mutual information maximization of a radar system as a target and is fused into communication capacity constraint, and ensuring collaborative optimization of perception and communication performance; and in combination with greedy initialization and a random search strategy, distributed algorithm iteration is executed, so that the calculation complexity is remarkably reduced, and the real-time response capability is improved.

The present invention belongs to the technical field of engineering structure analysis and optimization, and discloses a method for predicting vehicle body structural performance based on heterogeneous graph neural networks. The technical points include: sampling the structural samples required by the data set in the design domain to be predicted, using heterogeneous graph representation methods and numerical analysis methods to establish a heterogeneous graph data set; defining the HGNN design space with the HGT layer as the core; using part of the heterogeneous graph data set, controlled random search method and grid search method to construct the optimal HGNN model; using the entire heterogeneous graph data set and the optimal training configuration parameters to complete the training of the optimal HGNN model; using the trained HGNN model to predict the performance of any structural sample in the design domain. The present invention is not only applicable to homogeneous structures corresponding to homogeneous graphs, but also to more widely distributed structures corresponding to heterogeneous graphs, which greatly improves the versatility of the method and enhances the scalability of the method without losing the scope of application.

The present invention discloses a flexible job scheduling method for cloud-based resources based on an improved chimp optimization algorithm, and the method comprises the following steps: (1) adjusting the convergence factor of the chimp algorithm and setting its position update strategy according to the chimp algorithm; (2) designing a cloud-based resource scheduling scheme according to the chimp algorithm obtained in step (1) to provide data reference. The present invention can solve the flexible job shop scheduling problem. The present invention provides a new initialization method combining GS, LS and random search, which improves the quality of the initial solution of the population and speeds up the convergence rate of the genetic algorithm. By comparing with the test results of other genetic algorithms in the prior art, the calculation results are further improved, and at the same time, the calculation time is shortened to a certain extent, verifying the feasibility and effectiveness of the proposed initialization method.

Embodiments of the present disclosure disclose a vehicle escape method and device, electronic equipment and storage medium, the method comprising: in the case where the vehicle is in a blocked state, determining a virtual end point according to the starting point, target lane, each obstacle and preset search distance of the vehicle; generating a sampling map based on the Frenet coordinate system according to the starting point, virtual end point, current lane width, other lane width and each lane line; determining an initial path based on the fast random search tree according to the starting point in the sampling map, virtual end point and each obstacle; determining a plurality of target escape points according to the starting point in the initial path, each initial escape point in the initial path and each obstacle; and performing escape processing on the vehicle according to the plurality of target escape points. The present disclosure can flexibly determine a plurality of escape points when the vehicle is in a blocked state, improve the rationality of the escape points, and improve the success rate of vehicle escape.

The invention provides an operation path planning method and system integrating environmental perception and occupational health, and the method specifically comprises the steps: obtaining a three-dimensional map of a farm, and building a harmful gas prediction model related to an exhaust state; taking a personnel starting and ending point as a root node, performing bias sampling on a low-concentration region according to the model, and respectively constructing a forward and backward random search tree, a path cost comprehensive geometric distance and a predicted average concentration; performing reconnection optimization on the double trees, and updating father nodes of lower-cost adjacent nodes; connecting the optimized dual trees to generate an initial path; accumulated harmful gas exposure of the path is calculated, the gas exposure is the integral of the product of the retention time and the concentration, and target operation parameters of all exhaust equipment along the line are determined accordingly; and updating the prediction model by using the new parameters, repeatedly executing path search and exhaust parameter optimization until the total cost change of adjacent iteration paths is smaller than a convergence threshold value, and outputting an operation path and exhaust cooperation scheme.

The invention discloses a construction industry warehouse order time consumption prediction method and system based on machine learning. The method comprises the steps of collecting order data and cleaning the order data; constructing a multivariable feature set and carrying out one-hot coding; an XGBoost model is adopted for training, and parameters are optimized through grid search and random search; a prediction interface is deployed to a warehouse management system, the system comprises a data acquisition module, a data preprocessing module, a feature engineering module, a model training module, a prediction interface module and a prediction module, and tests show that the prediction precision Rreaches 0.85, the MAE is 10.6 minutes, the scheduling conflict rate is reduced to 2%, and the operating cost is saved by 75% annually. The problems that a traditional method depends on manpower and is low in precision are solved.

The invention discloses a short-term irradiance prediction method and device based on multi-source data fusion, and the method comprises the following steps: (1) data collection: obtaining meteorological data, actual measurement data of a radiation sensor and astronomical calculation data, and extracting time features; (2) data preprocessing: carrying out ADF unit root test on a global level irradiance sequence; (3) feature selection: screening features having a causal relationship with GHI through Granger causal test; based on XGBoost feature importance analysis, selecting an irradiance historical value and a solar azimuth angle as core input features; (4) model construction and training: processing a multivariable nonlinear relationship by adopting an XGBoost model, and dividing a training set and a test set according to a time sequence; time sequence cross validation is carried out to divide a validation set, and hyper-parameter tuning is carried out through grid search and random search; and (5) evaluation and optimization: using MAE, RMSE and MAPE index evaluation models to visualize a predicted value and actual value time sequence curve. According to the method, a night invalid data screening mechanism, ADF test and Granger causality analysis are introduced to perform feature selection, and time sequence cross validation is combined to perform hyper-parameter adjustment and optimization, so that the irradiance prediction model with high precision and strong generalization ability is realized, and the method is suitable for various photovoltaic application scenes and has wide application prospects. And reliable support is provided for power prediction and power station intelligent scheduling.

The invention relates to a step search type artificial neural electrode positioning method based on machine learning, and belongs to the field of neuroscience. Specifically, in a micro-driver control and execution module, a central control chip converts a digital feedback signal into an analog signal, the analog signal is amplified by a stack driving and coil current chip to generate a driving signal, and a micro-driver drives and controls an electrode to execute corresponding independent target actions under excitation of the micro-driver; in the signal acquisition and processing module, an electrode is implanted into a target brain region based on a stepping search type positioning method to perform searching and positioning of neurons and preprocessing after signal acquisition; in the machine learning model analysis and discrimination module, the quality of the neuron signal is discriminated in real time by using an integrated learning model, and a discrimination result is combined with the electrode state parameters to generate a digital feedback signal, and the digital feedback signal is transmitted to the central control chip for the next control cycle. Compared with a traditional random search type artificial neural electrode positioning method, the method can achieve a more accurate, more efficient and safer positioning effect.