50 results about "Sorting problem" patented technology

The embodiment of the invention provides order sorting method and system. The method comprises the steps of: generating a storehouse plane layout according to the putting condition of storehouse shelves; then computing the shortest path between nodes in a storehouse; finding out the positions, sizes and the weight of goods required for an order from pre-stored data; computing the shortest path and distance between any two pieces of required goods by using the shortest path between the nodes; and computing the optimal sorting path according to the capacity of sorting equipment and the required quantity of the required goods in the order so as to sort orders by the sorting equipment according to the optimal sorting path. The embodiment of the invention can effectively compute the sorting path, reduce the working cost, enhance the working efficiency, and solve the problems of inconsistent shelf arrangement, relatively small shelf height, and small expense difference caused by storing and taking the goods on different layers of shelves of the same goods position; and sorting workers mainly need to optimize order sorting problem in the horizontal movement distance for sorting work.

The invention discloses a multi-module clamping combined system. The multi-module clamping combined system comprises modules, a combined box, a management module and a handle. One or multiple modules can be arranged in the combined box in an inserting mode, and the combined box can be combined with the management module to form the multi-module clamping combined system. The multi-module clamping combined system can be a multi-module infusion combined system. The management module and the handle can be arranged in an inserting mode, and therefore moving and transporting of the multi-module infusion combined system are convenient. During use of the multi-module clamping combined system, the modules can be inserted or pulled out freely. The modules can be independent of the combined box and work, after the modules are inserted in the combined box, an internal power supply is used for supplying power for the modules, and the combined box can supply power for the modules through a direct current (DC) power input/output device. Thus, users can allocate infusion modules or modules with other functions conveniently and efficiently according to the demands, and the multi-module clamping combined system solves the disordered problem of power lines and the sorting problem of infusion pipelines.

The invention discloses a defective goods eliminating and sorting device based on a medical infusion visible impurity detecting system. The device comprises a cutter, a swing plate, a driving device, a clamping cylinder and clamping fingers, wherein one end of the swing plate is mounted on a main shaft of the cutter; the driving device is arranged on the swing plate; the driving device comprises a motor, a motor mounting base, a cylinder mounting base and a lead screw; the clamping cylinder is mounted below the cylinder mounting base; the clamping fingers are arranged on the clamping cylinder; and the cutter and the motor are both controlled by a PLC control unit of a medical infusion visible impurity classifying, identifying and detecting system. The device is combined with an IDS-ELM algorithm to realize the identification and the classification of visual impurities and bubbles, can precisely classify and identify various defects, and can discard the defects with different types to different defective goods areas. The cutter is skillfully applied to solve the sorting problem which causes trouble during infusion for a long time in the prior art; and the device is simple in structure and easy to realize.

The invention provides a hybrid traffic flow collaborative optimization control method based on double-layer planning, which belongs to the field of traffic engineering. The hybrid traffic flow collaborative optimization control method adopts a double-layer optimization model based on dynamic planning recursion to carry out hybrid traffic flow collaborative decision-making control, and is suitablefor different traffic scenes in a hybrid traffic flow, including highway ramp vehicle confluence, intersection vehicle confluence and intersection vehicle passing; the double-layer optimization modelcomprises an upper-layer model and a lower-layer model, the upper-layer model is a vehicle sorting problem solved by dynamic planning recursion, and the lower-layer model is a single vehicle trajectory optimization problem in different traffic scenes solved by dynamic planning recursion; and the upper-layer model and the lower-layer model are established, and the models are solved. According to the hybrid traffic flow collaborative optimization control method, the optimal vehicle operation of the system is guaranteed by means of the upper-layer model and the lower-layer model, so that the vehicle conflict of the vehicles in the process of convergence, confluence or intersection passing in the hybrid traffic flow environment, and effectively improves the traffic efficiency and comfort of the vehicles.

The invention provides a radar pulse repetition frequency sorting method and equipment, the main purpose of which is to solve the problem of radar pulse sorting in complex environments in the prior art. The technical solution of the present invention is: obtain the pulse description word PDW of the received pulse, perform pulse repetition interval PRI conversion according to the arrival time TOA in the PDW, and obtain the PRI spectrogram; search for the peak in the PRI spectrogram, and identify the The PRI value corresponding to the peak value; the TOA corresponding to the first pulse received by all pulses is subtracted to obtain the time difference of arrival DTOA of the pulse train, and a two-dimensional plane is generated according to the DTOA and the PRI value, Marking the pixel points corresponding to the pulse amplitude PA, carrier frequency RF, and pulse width PW of the PDW; determining a plurality of straight lines corresponding to the pixel points; fitting the straight lines, and fitting the straight lines corresponding to the fitting results A pulse is determined as the sorting result.

An optimized MMC sub-module capacitor voltage dynamic equalization control method comprises the following steps of: firstly determining a maximum allowable fluctuation range of a sub-module capacitorvoltage; performing equalization control on current sub-module capacitor voltages on an MMC bridge arm; if a certain sub-module capacitor voltage exceeds the specified maximum allowable fluctuation range, using a sub-module capacitor voltage direct equalization method based on improved sorting; and if all sub-module capacitor voltage on the bridge arm are within the specified range, using a sub-module capacitor voltage equalization control method without the need for sorting. The method can greatly reduce the problem that a control system has a large calculated amount due to the sorting problem.

The invention relates to a flexible workshop personnel configuring and work sorting method. An upper-layer ant colony algorithm is used for solving a resource configuration problem. A lower-layer antcolony algorithm is used for solving a work sorting problem on the condition of a fixed resource configuration solution. In the upper-layer ant colony algorithm, for facilitating configuring machiningequipment and operators of the step in an ant traversal process, a new ant traversal map and an ant traversal mode are designed, and furthermore local inspiration regulation information is designed.In the lower-layer ant colony algorithm, a process-based coding and real-number decoding mode utilized; a parent-child ranking elitist preservation strategy which promotes algorithm convergence is performed; and furthermore a calculated adaption value is fed back to the upper-layer ant colony algorithm, thereby guiding a subsequent ant traversal process.

The invention discloses a unique constraint based Deep Web entity identification method. The unique constraint based Deep Web entity identification method includes steps of firstly sorting problems into K-map cluster problems from the aspect of rigid constraint, and providing a clustering algorithm; and expanding the k-map cluster problems to the flexible constraint, sorting entity identification problems into optimization problems, and providing matching algorithm. In the unique constraint based Deep Web entity identification method, recording connection and data integration are integrated to be applied in overall situation, and the k-map cluster problems under rigid constraint are provided and are expanded to the flexible constraint. In the meantime, overall policy is determined on the basis of similarity of attribute values and relevancy among attributes in a same record, incorrect values can be identified and are differentiated from correct values from the beginning, and better identification effect is achieved. Clustering the attribute values can show clustering effect with finer grid.

The invention discloses a table sorting method based on cell adjacency relation and depth-first traversal. Cells in a mobile phone scanning or photographing table file are sorted so as to bring convenience for extracting appointed cell contents, and therefore, the position of each cell needs to be subjected to uniform rule sorting before processing is carried out. The table sorting method comprises the following steps: firstly, processing the binary skeleton map information of a table image, wherein the skeleton is the marginal information of all cells in the table image; then, according to the marginal information of each cell, judging the positions of four corners, and establishing an adjacency relation graph of each cell in the table image according to the outline of each cell; and finally, carrying out the depth-first traversal on a connected graph to realize the sorting of the cells. The invention provides an effective sorting method and can favorably solve the sorting problem of the cells in the table image which is distorted due to scanning or photographing.

The invention discloses a hierarchical bounding box tree construction method on the basis of a linked list sorting balanced binary tree, and belongs to the technical field of software collision detection in virtual reality. An object which needs to establish a bounding box tree is divided into basic geometric primitives according to reasonable accuracy, a corresopnding AABB hierarchical bounding box is generated, a linked list sorting balanced binary tree is created, the created binary tree is traversed, an attribute value is added for each binary tree object in the linked list for storing length from the binary tree to a reference site, and the linked list is subjected to ascending sorting according to the attribute value in the binary tree object in the linked list; the linked list is traversed, the binary trees in the linked list are combined pairwise, and a combination result is stored into the linked list. The height of the tree is lowered by a linked list sorting balanced binarytree method, the repeated sorting problem in a binary tree combination process is solved, time complexity during traversal is lowered, the time consumption of tree construction is saved, and the realtime and the collision detection efficiency of a collision detection process can be improved.

The invention relates to a block chain consensus system based on an asynchronous sorting technology, which is characterized in that an asynchronous sorting algorithm is creatively created, and a consensus problem is converted into a problem of processing large-scale concurrent requests of an asynchronous system and a problem of sorting data; an asynchronous sorting algorithm is creatively created,a consensus problem is converted into a problem of processing large-scale concurrent requests of an asynchronous system and a data sorting problem, and the practicability is high.

The invention particularly relates to a test case sorting method based on test resource awareness. According to the method, regression test is considered as a sequence of test behaviors that are partially ordered, the test behaviors in each phase are subject to those in the previous phase, and considering test resources and time restraints is also required. The method of the invention has the advantages that the prior sorting problem of test cases is solved by awareness of test resources, as many as defects present in software under test are detected by designing high-quality test cases, all test cases can be executed as far as possible with resources limited, product quality is guaranteed, and software needs are better met.

The invention relates to a sphere material roundness sorting machine which comprises a base frame body, a storage hopper, a discharging trough, a conical sorting plate, a retention ring, a drive motor, a rotary bearing, a shaft sleeve, a shaft sleeve installation plate, a waste barrel and a finished product barrel. The storage hopper, the discharging trough, the conical sorting plate and the wastebarrel are sequentially arranged on the base frame body from top to bottom; the conical sorting plate is in a funnel shape; the rotary bearing is fixed under the conical sorting plate and is locatedoutside the waste barrel; the shaft sleeve is arranged outside the rotary bearing in a sleeve mode and is fixed onto the shaft sleeve installation plate; the shaft sleeve installation plate is arranged on the base frame body; the drive motor is arranged on the shaft sleeve installation plate and is connected with the rotary shaft; the finished product barrel is arranged in the waste barrel; a discharge hole of the finished product barrel and an outlet of the waste barrel are not shared; and a notch groove is formed in the retention ring and is located on an opening of the upper edge of the finished product barrel. According to the sphere material roundness sorting machine disclosed by the invention, the funnel-shaped conical sorting plate is arranged, so that the sorting problem of the sphere material roundness can be effectively solved.

The embodiment of the invention provides a multi-target optimization result sorting method and a related device, and the method comprises the steps: dividing a plurality of target parameters into a plurality of optimization sets according to a first priority level, and then determining a reference optimization index of each optimization set, and determining a comprehensive optimization index corresponding to one-time multi-objective optimization according to the reference optimization index of each set, and sorting results of multiple times of multi-objective optimization according to the comprehensive optimization index of each-time multi-objective optimization to obtain a first sorting result. In this way, the multi-target sorting problem is converted into the single-target sorting problem, the complexity of the multi-target optimization sorting problem is reduced, the sorting efficiency is improved, sorting is conducted according to the requirements, quality and the like of actual industrial products, and the sorting efficiency is improved. And the user can directly determine the production parameters and other contents corresponding to one-time multi-objective optimization learning which can be used for actual production according to the sequence, so that the use experience of the user is improved.

The invention discloses an intention recognition algorithm based on an embedding method. The algorithm comprises the steps of based on the data in the specific field, mapping the input text and intention of the user to the same vector space, enabling the input text and intention of the user to be subjected to similarity comparison in the same space, and converting a classification problem into a sorting problem; converting the intention classification problem into an intention sorting problem by mapping statements and intentions to the same vector space; modifying the loss function in the paper, so that the new loss function has better stability and robustness. The invention provides an intention recognition algorithm based on an embedding method. The word vectors pre-trained on the general data set need to occupy the memory space of more than 1 GB, and the model provided by the scheme of the invention only needs about 100 MB memory space due to the fact that only the concerned words and intentions are embedded, so that the occupied memory space is smaller.

The invention provides an element sequence automatic arrangement method of a control strategy configuration. The method is achieved based on a following topological sorting rule: firstly, a priority level of an input side element is superior to a priority level of an output side element; secondly, two elements which are inputted by each other are compared with the number of elements without sorting of input sides of two elements, and elements with less number are firstly sorted; thirdly, input elements or transitional elements just conduct once sorting in the same return circuit, the input elements or the transitional elements are arranged on the input side, and the input elements and the transitional elements have the same subscript numbers and properties; fourthly, the transitional elements with the same subscript numbers and properties repeatedly occur in the same return circuit, and if some transitional elements are arranged on the output side, the transitional elements which are arranged on the output side are taken out to firstly conduct sorting. The invention further provides a logic link method of the control strategy configuration. The logic link method of the control strategy configuration comprises a return circuit link and a system link. The element sequence automatic arrangement method of the control strategy configuration and the logic link method thereof achieve automatic arrangement of element sequences, not only improve efficiency of industrial control configuration, but also solve unscientific sorting problem of a round-robin algorithm, and strongly guarantee stability and safety of an industrial control process.

The invention discloses a height detecting and sorting apparatus of workpieces, wherein the apparatus includes a machine base (11), a machine bench (4), a support, a workpiece press-in apparatus, a place-base fixing apparatus and a work sorting apparatus. The work sorting apparatus includes a qualified product pushing apparatus and a rejected product pushing apparatus which are respectively disposed at the two sides of a workpiece (14) in parallel. A free end of the qualified product pushing apparatus is as same as the top of the workpiece (14) in height. The rejected product pushing apparatus is arranged under the qualified product pushing apparatus. The qualified product pushing apparatus is connected to the workpiece press-in apparatus through a first link mechanism and is connected to the rejected product pushing apparatus through a second link mechanism. The first link mechanism and the second link mechanism are both arranged on one end, which is far away from the place-base fixing apparatus, of the machine bench (4). By means of the structure, detection and sorting efficiency can be greatly improved and detection and sorting operation is more stable, thereby avoiding mis-detection and mis-sorting problems.

The invention discloses a single machine batch sorting method for minimizing total delay time considering learning effect. The single machine batch sorting method comprises the following steps of establishing a mathematical model for minimizing total delay time of a single machine batch sorting problem considering learning effect; according to the number of batches in the model, dividing into small-scale and large-scale cases; according to different scales, determining the corresponding model solving algorithm to obtain the minimum batch total delay time and batch processing order. Combined with the practical problems, the method of the application proposes a single machine batch scheduling problem that minimizes the total delay time considering the learning effects, through modeling analysis, two methods of dynamic programming and SA are proposed to solve this problem, and the problem is divided into the large-scale and small-scale problems, and the experimental analysis is made.

The invention belongs to the related technical field of flexible flow shop scheduling, and discloses a multi-objective mixed spider monkey optimization method for a PCB electroplating task sorting problem, and the method comprises the following steps: S1, parameter definition and population initialization; s2, performing neighborhood movement by adopting insertion operation; s3, performing dominated sorting on the Nei neighborhood solutions of each spider monkey and the spider monkey itself; s4, the fitness F (i) of each non-dominated solution is calculated, and the current position of each non-dominated solution is updated according to the fitness; s5, carrying out interlace operation on the solution F1 and the random solution F2 to obtain a filial generation; s6, forming a joint population by the solution in the S5 and the solution in the global file set; s7, combining the combined population in the step S6 and the generated filial generation, and updating the solution in the current group k; s8, updating the local file set and the global file set of each group; if the termination condition is met, terminating and outputting a solution set; otherwise, continuing iteration until the end. The performance is improved, and the applicability is good.

The invention discloses an evolutionary ensemble learning method for classification problems, and belongs to the field of machine learning and data mining. The method comprises the following steps: firstly, generating a group of base learner groups based on tree coding and a corresponding integrator group based on binary coding for a classification problem; then, performing multi-objective evolutionary search of a base learner group and combined optimization search of an ensemble learner group; two levels of evolutionary search processes are combined and mutually fed back in information, so that a high-quality base learner group is constructed, and a high-accuracy calculation target of integrated learner individuals is generated. According to the method, two levels of search optimization processes can be dynamically combined through an evolutionary iteration mechanism, and then collaborative optimization is realized through information exchange between the search optimization processes, so that a base learner group is promoted to evolve continuously, and balance between high quality of individuals and diversity of the group is kept; Meanwhile, global optimization of the integrated learner structure is realized.

The invention discloses a system for realizing optimization operation by using block chain consensus computing power. The core innovation point of the invention is to provide a new proof-of-work algorithm. Consensus of a block chain system is realized by solving a standardized sorting problem, and meanwhile, a system for converting an actual optimization problem into a series of standardized sorting problems, distributing the standardized sorting problems to each accounting node of the block chain system for solving and finally combining solutions of each sub-problem into an original problem solution is provided. By using the system provided by the invention, the consensus computing power of the block chain system which is completely useless originally can be fully utilized, and strong computing power support is provided for solving various optimization problems, so that huge economic and social benefits are brought into play.

The invention discloses a multi-class multi-scale target sorting method and device based on continuous learning, and the method comprises the steps: achieving the hidden space representation of new and old task features through a generative adversarial network, building a knowledge memory storage warehouse, achieving the quick switching of multi-task sorting by a self-adaptive knowledge migration scheduling module. The training process of a new task is accelerated, the importance degree of model parameters is quantified in combination with a regularization technology, and key parameters are prevented from being covered in the training process. The data replay framework is used for the intelligent part sorting system, the problem of function disastrous forgetting occurring in the continuous learning process of a multi-type multi-scale target sorting model can be solved, the integrity of the function of the sorting model is ensured, continuous learning with dynamically changed targets and scenes is achieved, and the learning efficiency is improved. The system is assisted coordinately and efficiently solve the multi-task sorting problem.

The invention discloses an adaptive multi-target test case sorting method based on a decomposed weight vector. Aiming at the problems of single target, high time cost, non-ideal effect and the like in the traditional test case sorting problem, the MOEA/D-VW algorithm is creatively introduced, and the average branch coverage rate and the effective execution time are taken as optimization targets. According to the method, multi-target test case sorting can be effectively realized in a shorter time. According to the invention, the early detection rate of defects is improved, and the regression test cost is effectively reduced. In the aspect of the method, a traditional multi-objective optimization algorithm is improved, and a self-adaptive weight vector change strategy is added, so the result has better distribution, and the obtained test case sorting sequence has more selectivity.