53results about How to "Not easy to fall into local optimal solution" patented technology

The invention discloses an inner and outer layer nesting ECMS (equivalent fuel consumption minimization strategy) multi-objective double-layer optimization method. The inner and outer layer nesting ECMS multi-objective double-layer optimization method includes steps of building multi-objective optimization models of plug-in hybrid electric vehicles; solving the multi-objective optimization modelsby the aid of inner and outer layer nesting multi-objective particle swarm algorithms to obtain multi-objective optimized Pareto solution set front edges; weighting equivalent fuel consumption per hundred kilometers and variation ranges of deviation of SOC (state of charge) final values and target values, building total evaluation functions related to the equivalent fuel consumption per hundred kilometers and SOC deviation and selecting the optimal charge and discharge equivalent factors and engine and motor power distribution modes corresponding to the optimal charge and discharge equivalentfactors. The inner and outer layer nesting ECMS multi-objective double-layer optimization method has the advantages that output power of engines and motors of the plug-in hybrid electric vehicles canbe reasonably distributed at CS (charge sustaining) stages, so that fuel consumption can be reduced as much as possible, battery SOC balance still can be effectively kept, and the fuel economy of theintegral vehicles can be improved.

The invention discloses a KNN text classifying method for optimizing a training sample set, which belongs to the fields of text mining, natural language processing and the like, and solves the problems of the traditional KNN text classifying method that the efficiency and accuracy are low. The KNN text classifying method is characterized by comprising the following steps: preprocessing text data for training and to-be-classified text data; carrying out the text representation for the preprocessed text data for the training and the to-be-classified text data; carrying out the characteristic extraction for the text data and the to-be-classified text data after the text representation by utilizing a genetic algorithm; carrying out the classification training for the extracted text data characteristics for the training, carrying out the training classification by utilizing a KNN algorithm of the optimized sample set, and configuring a text classifier; utilizing the text classifier to act on the to-be-classified text data after the characteristic extraction to obtain a classification result of the to-be-classified text data. The KNN text classifying method can be better applied to a text information mining system.

The invention relates to a load power consumption mode identification method. The load power consumption mode identification method includes the steps: acquiring the electrical load at a sampling time interval T, and obtaining L daily load curves corresponding to L days of time; performing spatial clustering based on density on the obtained daily load curves, and obtaining a classical load power consumption mode; extracting characteristics describing the power consumption behavior of a user in different time scale; and utilizing a gravitation search algorithm to cluster the obtained power consumption characteristics of the user; repeating clustering, utilizing a cluster evaluation index to evaluate the clustering result, and selecting the optimal clustering result, that is, the identification result of the load power consumption mode. The gravitation search algorithm used by the load power consumption mode identification method has high searching capability and high convergence speed, and is not easy to fall into local optimal solution, and is better than a traditional clustering algorithm on the identification effect, so that identification of the load power consumption mode can be effectively realized and powerful guidance for design of the demand side response scheme, analysis of load characteristics and high-accuracy prediction can be provided.

The invention belongs to the field of computer vision mode recognition and image understanding and relates to an automatic image segmentation method of continuous quantum goose group algorithm evolution pulse coupling neural network system parameters. The method comprises the steps that a minimum combination weighting entropy model of automatic image segmentation of the evolution pulse coupling neural network system parameters is established; a continuous quantum goose group population space is initialized; a simulation quantum rotating door is used for updating the position of each wild goose; the position of each wild goose corresponds to a pulse coupling neural network system parameter, a pulse coupling neural network system is activated for image segmentation, and a fitness value of a new position of an i wild goose is computed; the history optimal quantum positions and the history optimal positions of all wild geese are updated; whether the maximum iteration algebra is reached is checked; and a pulse coupling neural network model is substituted to carry out segmentation on images and output the images after segmentation. The method has the advantages of being small in computing amount, high in convergence rate and high in optimizing capacity.

The invention discloses a production line workshop scheduling method which comprises the steps of setting scheduling parameters, wherein the scheduling parameters comprise the number of workpieces, the number of machines and unit machining time; according to the scheduling parameter, establishing a production line workshop scheduling optimization object function; solving the optimization target function according to a discrete firefly optimization algorithm based on position updating, and obtaining an optimal scheduling sequence. Furthermore the invention discloses a production line workshop scheduling device.

The invention discloses a database query optimization method. The database query optimization method comprises a connection sequence selector and a self-adaptive decision network, wherein the connection sequence selector is used for selecting an optimal connection sequence in the query plan and comprises a new database query plan coding scheme, and codes are in one-to-one correspondence with the connection sequence; a value network which is used for predicting the execution time of the query plan, is trained by the query plan and the corresponding real execution time, and is used for reward feedback in Monte Carlo tree search; a Monte Carlo tree search method which is used for simulating and generating multiple different connection sequences, evaluating the quality of the connection sequences through a connection sequence value network, and returning a recommended connection sequence after preset exploration times are reached. And the adaptive decision network is used for distinguishing whether the query statement uses the connection sequence selector or not, so that the overall performance of the optimization system is improved. According to the method and the system, the limitation of a traditional query optimizer can be effectively avoided, and the database query efficiency is improved.

The invention discloses an intelligent vehicle lane changing track collaborative planning method based on instantaneous risk assessment. The method comprises the following steps: 1) obtaining trafficbasic data: vehicle motion data, vehicle relative position information, vehicle geometric parameters, road geometric conditions and road surface friction coefficients; 2) collecting motion trails of surrounding vehicles to construct a lane changing sample, and extracting motion characteristics of the surrounding vehicles by using a fuzzy C clustering algorithm; 3) based on the motion characteristics of the surrounding vehicles, predicting the vehicle motion state and the vehicle spacing of the surrounding vehicles at the moment t; 4) establishing a lane change condition safety discrimination model, and deriving a vehicle operation instantaneous risk index by using the parking distance index; 5) taking instantaneous risk minimization as an objective function, and planning transverse and longitudinal movement tracks of the vehicle by utilizing a trapezoidal acceleration method and a genetic algorithm respectively; and 6) realizing a vehicle lane changing track planning function through computer programming. According to the invention, lane changing safety can be recognized in advance, the vehicle lane changing track is planned, and theoretical reference and technical support are provided for a vehicle lane changing track planning system.

The invention discloses an image bit depth enhancement method based on a deep learning network, and the method comprises the following steps: carrying out the preprocessing of a high-bit lossless-image-quality image, and carrying out the quantification of the high-bit lossless-image-quality image into a low-bit image; designing a convolutional neural network based on deep learning, taking the quantized low-bit image as input, and taking perception loss between an output result and the original high-bit image as a loss function; training parameters of each convolution layer and each batch normalization layer in the model through an optimizer gradient descent loss function, and storing the corresponding model and parameters of all layers in the model after the descent amplitude of the modelloss function does not exceed a threshold value; and after the high-bit image is quantized into low bits, recovering the high-bit image through the stored convolutional neural network. According to the method, a convolutional neural network framework in deep learning is utilized, so that a high-quality high-bit image can be accurately recovered.

The invention discloses a power transmission line icing thickness prediction method based on CEEMDAN-QFOA-LSTM, and relates to the field of combination of power transmission line state evaluation anddeep learning. The method comprises the following steps o: (1) carrying out data acquisition and preprocessing; (2) carrying out CEEMDAN decomposition on an icing thickness historical data sequence (12); (3) optimizing hyper-parameters of the LSTM by a quantum drosophila melanogaster algorithm; (4) carrying out LSTM model training (14); and (5) predicting the icing thickness of a power transmission line and analyzing a result (15). According to the method, the CEEMDAN decomposition algorithm is used, a sequence which is difficult to directly predict is converted into a plurality of predictablecomponent sequences; a neural network can more accurately grasp the law of the sequence according to multi-dimensional feature information obtained through decomposition; a QFOA optimization algorithm is used for obtaining the hyper-parameters, a complex manual parameter adjustment process is avoided, and a network model is trained more effectively; the used LSTM neural network does not have theproblem of gradient disappearance of a general network, so that optimal convergence of the model is ensured, and the problem of short-term and long-term time sequence prediction is effectively solved.

The invention discloses a global optimization method for multi-thread operation of AGVs used for scheduling of cigarette ingredients. The method comprises the following steps: constructing a time sequence constraint matrix and a resource constraint matrix; initializing particle swarm by adopting a random generation task priority and an execution method; encoding particle individuals according to arandom rule and a random execution method; decoding the particle individuals; calculating a local optimal solution of the particle individuals and a global optimal solution of initial particle swarm;and updating the particle individuals based on the local optimal solution and the global optimal solution, and randomly changing the value of a certain coding position to generate new particle individuals. According to the global optimization method for multi-thread operation of AGVs used for scheduling of cigarette ingredients, updating iteration is conducted on all the particles by means of local solutions and global optimal solutions of the individuals in all the stages, and mutation operators are added to increase population diversity, so the advantages that a particle swarm algorithm iseasy to achieve and fast in convergence are reserved, proneness to falling into the local optimal solutions is avoided, and a solution can be provided for task scheduling of multiple AGVs in distribution operation.

The invention relates to the field of self-adaptive control, in particular to a mechanical arm kinematics self-calibration method based on binocular vision. Firstly, data are collected, and the tail end position information of a mechanical arm and the actual rotation angle and the theoretical rotation angle of each joint of the mechanical arm are obtained through a binocular camera. Then, the DH theory and the hand-eye calibration theory are fused, and a mechanical arm kinematics hybrid model is established. The model is trained by using a multi-population self-adaptive difference algorithm, and parameters of the hybrid model are solved. Finally, each servo motor model is established through a polynomial fitting method, and polynomial parameter solving and compensation prediction are conducted by using a least square method. According to the hybrid model and the servo motor models provided by the invention, the influence of geometric errors on the mechanical arm can be greatly reduced, and more practical model parameters can be calculated. A mechanical arm base coordinate system needed in the hand-eye calibration process does not need to be additionally established through a demonstrator, and automation of the whole calibration process can be achieved on the premise that absolute positioning precision is guaranteed.

Provided is a KNN algorithm based article translation method. The method comprises the steps of: firstly, splitting a relatively large article, extracting a keyword from an article to be translated and classifying the article; and obtaining an optimal allocation result by using an algorithm to match a K value. The invention is mainly about preprocessing articles that are used for training and mail data to be classified; performing textual representation on the preprocessed articles that are used for training and mail data to be classified; using a genetic algorithm to perform feature extraction on the textually represented articles that are used for training and on mail data to be classified; performing classification training on the extracted features of the articles that are used for training; using the optimized sample set KNN algorithm to perform classification training, so as to construct a text classifier; and applying the text classifier to the articles to be classified after the feature extraction, so as to obtain a classification result of the articles to be classified. The method provided by the invention can be better applied to an article text information mining system.

The invention provides a PID permanent magnet synchronous motor control method based on an optimization algorithm, which adopts BP neural network PID control optimized by a GA algorithm and consists of two parts: 1, implanting a genetic algorithm into a BP neural network for learning and training, and finding out a weight and a threshold value suitable for a control system; and further combining an algorithm with a PID controller, so that real-time adjustment of PID parameters is achieved. Compared with other traditional optimization algorithms, the BP neural network optimized by the genetic algorithm has excellent global optimization capability and good convergence property, effectively avoids falling into a local optimal solution, and also has adaptive and self-learning capabilities.

The invention relates to a data intelligent management method for a new energy electric vehicle. The method includes the steps of acquiring remote transmission data of a vehicle-mounted terminal of the new energy electric vehicle; conducting data processing on the remote transmission data of the new energy electric vehicle by using an improved data processing combination algorithm. The vehicle-mounted terminal comprises an OBU, an OBU information acquisition module, a display module, a management module, a chip and a battery, wherein the OBU information acquiring module is used for sending a query request to the OBU and acquiring vehicle information and charging card information which are stored in an electronic tag from the OBU; the display module is used for displaying the vehicle information and charging card information which are acquired by the OBU information acquisition module; the chip is used for controlling the OBU information acquisition module to send the query request to the OBU and acquiring the vehicle information and charging card information which are stored in the electronic tag from the OBU; the management module is used for direction and drive management of thechip and the information acquisition module. According to the method, the data processing capability is high, management is fast and convenient, and decision making is accurate and convenient.

The invention relates to the technical field of computers, in particular to an order management method, an order management device, a computer readable medium and electronic equipment. The method comprises the steps of obtaining a current order set; determining target order allocation information, and replacing one or more to-be-allocated orders in the current order set with candidate orders according to the target order allocation information to obtain a candidate order set corresponding to the current order set; obtaining a current evaluation score of the current order set, and obtaining a candidate evaluation score of the candidate order set; when the candidate evaluation score is greater than or equal to the current evaluation score, taking the candidate order set as a new current order set; when the candidate evaluation score is smaller than the current evaluation score, taking the current order set as a new current order set; and repeatedly executing the above process until the number of repeated execution times exceeds a preset value, and taking the latest current order set as a target order set. The method can improve the utilization rate of order processing resources.

The invention provides a modeling method for a power transmission line icing thickness prediction model based on PR-KELM, and the method comprises the steps: a first stage: converting image data intoLBP image data, carrying out the dimension reduction through employing a PCA algorithm, calculating the gray histogram cascade, and obtaining the extracted image data features; performing feature screening on the meteorological data and the mechanical data by adopting a ReliefF algorithm, and removing highly related redundant features to obtain extracted meteorological and mechanical feature data;and a second stage: forming sample data by using the feature data obtained in the first stage and the icing level in the original image data, training a PR-KELM model by using the training data, testing the trained PR-KELM model by using the test data, and finally obtaining a power transmission line icing thickness prediction model. The method has the advantages that the PR-KELM model is adoptedto predict the icing thickness, selection of the learning rate is not very sensitive, the method is not prone to falling into a local optimal solution, and therefore the accuracy of the prediction model is improved.