45 results about "Complex problems" patented technology

The embodiment of the invention discloses a 4K splicing processor and a 4K splicing display system. The 4K splicing processor comprises a driving circuit used for processing a 4K resolution signal, the driving circuit comprises an FPGA chip U5, TVS array chips U10, U11, U12 and U13 and DP signal line connectors CN9 and CN10; the model of the U5 is Spartan-6, the models of the U10, the U11, the U12 and the U13 are all RClamp0524P, the CN9 is connected with the U5 through the U10 and the U11, and the CN10 is connected with the U5 through the U12 and the U13; the U5 receives the input 4K resolution signal through the CN9 to process the 4K resolution signal and outputs the processed 4K resolution signal through the CN10 to achieve 4K resolution signal looping. According to the 4K splicing processor and the 4K splicing display system, the problem that 4K splicing display system equipment is complex is solved.

This invention belongs to the field of artificial intelligence technology and relates to a method for solving complex problems based on dynamic updating of subject graphs and multi-agent collaboration. The method includes: S1: Problem standardization processing; S2: Semantic encoding; S3: Initial subject set generation; S4: Candidate subject graph construction; S5: Initial formal subject graph construction; S6: Node and agent matching; S7: Output result acquisition; S8: State evaluation; S9: Subject graph dynamic update trigger judgment; S10: Subject graph dynamic update; S11: Termination condition satisfaction judgment; S12: Solution result generation. By introducing a dynamic updating of subject graphs and an agent collaboration mechanism, the method can adjust the subject graph in real time based on intermediate results during the reasoning process, ensuring that the subject graph always maintains a high degree of matching with the reasoning requirements, thereby significantly improving the reasoning stability, collaborative efficiency, and result accuracy in the process of solving complex problems.

The application relates to the technical field of pose adjustment, and provides a pose adjustment device, which comprises a base, a positioning seat and a holder. The base is used for being fixed on a workbench. A plurality of groups of clamping grooves are arranged on the base in a first direction. The positioning seat can be clamped into the plurality of groups of clamping grooves respectively to adjust the angle of the positioning seat. The holder is arranged on the positioning seat and is used for clamping a product to be measured. In use, the base is fixed on the workbench, and then the product to be measured is clamped and fixed on the positioning seat by the holder, so that the product to be measured, the holder and the positioning seat form an integral whole. The angle of the positioning seat is adjusted by clamping the positioning seat into the plurality of groups of clamping grooves respectively, so that the installation angle of the product to be measured is simulated. Compared with the prior art, the pose adjustment device does not need to use a driving mechanism, the problem that the structure of the existing pose adjustment device is relatively complex is solved, and the production cost is reduced.

A hierarchical controller synthesis method based on dynamic quantization under a linear sequential logic specification belongs to the field of nonlinear system control technology. First, a model checking algorithm is used to obtain an acceptable path for the linear sequential logic specification, providing a high-dimensional solution for the controller synthesis problem. Second, a quantization interval is generated based on a dynamic quantizer to verify the implementation of the acceptable path. Finally, a local abstract controller is obtained by solving the auxiliary control problem. This local abstract controller is refined to obtain a local hybrid controller, which is then combined into a global hybrid controller. Applying this controller to the original system enables it to satisfy the given linear sequential logic specification. The abstract model construction and controller synthesis adopted in this invention are both local and dynamic, thus effectively reducing computational complexity. Since each quantization space can be considered independently, the proposed hierarchical mechanism is more effective than existing methods in solving larger and more complex problems.

This specification provides a problem-solving method and apparatus, a computing device, and a computer-readable storage medium. The method includes: retrieving related nodes of a target problem from a semantic index graph; decomposing the target problem into multiple atomic search instructions based on the target problem, the graph structure of the semantic index graph, the related nodes, and a preset maintenance experience template; retrieving the target node corresponding to each atomic search instruction from the semantic index graph; and determining the target answer based on the target node corresponding to each atomic search instruction and the target problem. This method, through problem decomposition and graph retrieval based on semantic index graphs and expert experience, effectively improves the accuracy, completeness, and operability of solving complex problems in vertical domains.

The application relates to the technical field of industrial automation and intelligent manufacturing, and discloses a kind of light mixed optimization systems driven by edge computing, including edge computing node, cloud computing center and data interface;The edge computing node is used to execute real-time data acquisition, preliminary scheduling scheme generation and event response operation, to realize the rapid processing of production site dynamic event;The cloud computing center is used to execute task priority prediction and global optimization solving operation, to solve the complex global problem in production planning scheduling, and the data interface is used to realize the data interaction of edge computing node, cloud computing center and external production management system. Through the mixed architecture of edge computing node and cloud computing center cooperation, combined with the light heuristic algorithm, the preliminary scheduling scheme is quickly generated, the cloud global optimization engine solves the complex problem, the response speed of production planning scheduling system is significantly improved, and the real-time demand for dynamic adjustment in production process is effectively met.

The application relates to the technical field of artificial intelligence and industrial simulation, in particular to an evolvable industrial simulation question and answer intelligent agent system and a simulation question and answer method thereof. The evolvable industrial simulation question and answer intelligent agent system comprises a knowledge base module, a tool calling module, an intelligent agent core module, a user interaction module and a log and audit module; the knowledge base module comprises a multi-element knowledge collection unit, a document loading and cutting unit, a vectorization embedding and storage unit, a semantic retrieval unit, a knowledge evolution unit, a rule base and a case base; the application has the beneficial effects that (1) knowledge is dynamically evolved and updated autonomously and timely; (2) the system has the capability of autonomously refining knowledge from simulation results; (3) tool calling is intelligentized, and the system realizes "question and answer as calculation"; (4) the system has strong capabilities of complex problem disassembly and iterative solution; and (5) the system has excellent adaptability and expandability.

The application discloses a multi-aircraft multi-site delivery scheduling planning method, a storage medium and equipment, and belongs to the technical field of aircraft delivery scheduling planning.The application solves the problem that the existing method cannot obtain an optimal scheduling scheme for a large-scale aircraft to visit multiple sites.The application considers a scheduling planning problem of multiple aircrafts successively arriving at multiple sites, considers processes such as unloading, loading, refueling and maintenance, optimizes the visiting sequence of the aircrafts to each site, establishes a multi-aircraft multi-site delivery scheduling model, proposes a discrete data iterative sorting algorithm based on a neural network for a large number of aircraft sorting complex problems, solves the sorting problem of high-dimensional input through layer-by-layer decomposition iteration, realizes the target of the shortest total time consumption of material delivery, and obtains a better delivery scheduling scheme.The method can be applied to multi-aircraft multi-site delivery scheduling planning.

The application relates to the field of intelligent management analysis, and proposes an electric locomotive maintenance practical training intelligent analysis method and system based on an AI model, which constructs an electric locomotive virtual practical training scene, avoids the problem that a large amount of manpower and material resources needs to be consumed during actual scene training, constructs an electric locomotive maintenance knowledge graph, generates dynamic operation risks, makes the practical training operation results and fault problems more consistent with the actual scene, avoids the problem that a single fault case cannot restore the complex problems and complex operations of the actual maintenance scene, is based on a double-base interaction mechanism, ensures the accuracy and reliability of the guidance information, further improves the accuracy of the guidance information through deep expansion, gives deeper information, avoids the problem that the depth of the guidance information in the field knowledge is insufficient, and objectively authenticates the practical training effect through automatic quantitative evaluation, and the application improves the practical training effect and the practical training quality of operation and maintenance personnel.

The present application relates to the technical field of firmware booting, and particularly provides an automatic multi-path booting method based on a configuration file, comprising the following steps: S1, when a UEFI firmware is executed to a preset callback point of a booting device selection stage, a callback function is executed, all accessed storage devices are enumerated in the callback function, whether an external storage medium conforming to a preset file system format exists is detected, and whether a predefined configuration file exists is found by traversing a root directory or a specified path of the medium; S2, if the configuration file is found, contents thereof are loaded to a memory buffer by a UEFI file system protocol, and are terminated by a null character; subsequently, a text structure of the configuration file is parsed according to a keyword, at least one booting strategy identifier and corresponding target booting parameters are extracted, and the present application can effectively solve the problems of backwardness, low efficiency, high operation cost and complex process of an existing UEFI pre-booting interaction mode.

This utility model provides a combined expander and tightener tool, belonging to the field of hydraulic tools. It solves the problem of complex structure in existing integrated expander and tightener methods. This combined expander and tightener tool includes a cylinder body, inside which are housed an expander, a sliding sleeve tightener, and a rotating shaft. The cylinder body can rotate around the rotating shaft. A reversing valve body is fixedly connected to the lower end of the cylinder body. The reversing valve body integrates a reversing valve core and a valve seat for mounting the reversing valve core. One end of the rotating shaft is fixed to the cylinder body, and the other end extends into the reversing valve body and is drively connected to the reversing valve core. The rotating shaft has independent first and second oil passages, respectively connecting the expander and the sliding sleeve tightener. A cylinder connecting shaft is provided on the valve seat, and the cylinder connecting shaft has an internal oil passage communicating with the reversing valve core. The reversing valve body has a reversing positioning mechanism for locking the conduction position of the reversing valve core. This utility model has the advantages of simple structure and quick switching.

The utility model provides a gear shaft with lubrication belongs to mechanical technical field. It has solved the problem of complex structure of existing gear shaft. The gear shaft with lubrication of this utility model includes the hollow axle body, the outer gear is formed on the axle body upper end, the axle body lower end is vertically equipped with a ring of keyway, the lower end inner wall of axle body has a section of taper guide surface, the guide surface and axle body coaxial arrangement, the keyway upper end extends to the guide surface, the inside wall of axle body is horizontally formed with the support block, the support block upper side is equipped with the recessed cavity, the recessed cavity bottom wall is vertically penetrated with a ring of along the axle body circumferential distribution oil hole, every oil hole is opposite the guide surface setting, the side wall of axle body is equipped with the oil injection hole of making the recessed cavity communicate with the outside of axle body. The gear shaft with lubrication of this utility model simple structure.

The application discloses a track tensor completion and abnormality repairing method based on singular value weighting and truncation, and relates to the technical field of flight track data processing.The application fully utilizes easily-obtained ADS-B track data, and mines the internal law through intelligent learning capability; meanwhile, aiming at the complex problem that missing and abnormality are coupled with each other in the track data, a robust tensor model is innovatively constructed, which fuses an adaptive weight mechanism, singular value truncation and sparse abnormality constraint, so that the collaborative and accurate processing of missing completion and abnormality repairing is realized; further, an optimization solving strategy based on an alternating direction multiplier method is proposed, and through the augmented Lagrange method, low-rank track tensors, sparse abnormality tensors and auxiliary variables are efficiently block-iteratively optimized, so that the complete track can be accurately recovered in a complex data environment.

The application discloses a complex problem knowledge externalization modeling system and method based on human-computer conversation, and belongs to the technical field of computers. Through compliance preliminary screening and structured coding of multi-source cross-modal knowledge, a dynamic conflict graph with double time markers is constructed, and the conflict detection sensitivity is dynamically adjusted based on the user cognitive state, solving the problems of traditional human-computer conversation system knowledge fusion black box and cross-modal conflict detection deficiency. The decision logic is encapsulated as a visual decision path graph through a dynamic arbitration engine, and the presentation complexity is adjusted according to the user cognitive state, realizing the transparency and interpretability of the arbitration process. At the same time, the user intervention result is obtained in real time and strategy recalculation is triggered, combining with the ethical pre-check and example library mechanism, a closed-loop system of deep coupling of conflict detection, dynamic arbitration, user intervention and ethical compliance is constructed, promoting the co-evolution of each module and improving the credibility and intervenability of the system.

The application relates to the field of battery packs, in particular to a design method of a power battery pack sealing structure, which comprises the following steps: step 1, determining a risk point; step 2, analyzing a sealing failure reason of the risk point and determining the failure reason; and step 3, quantitatively designing according to the failure reason; the application adopts a complete problem analysis framework: firstly, the risk point (an insert and a shell) is identified, then the failure reason (insufficient compression amount, torque attenuation, resonance and the like) is analyzed in depth, and then the complex problem is summarized into three core design elements (a sealing gasket, a fastening torque and a shell mode). The systematic method can comprehensively check potential failure modes in the design stage, and avoids the sealing failure risk from the source.

This application discloses a multi-decision collaborative control method, apparatus, electronic device, and computer program product for streaming video understanding. The method includes performing visual token compression on neighboring interval data and historical interval data respectively to obtain the current memory state; calculating a readiness score based on the current query to be answered and the current memory state; in response to the readiness score not being lower than a score threshold, inputting the current query to be answered and the current memory state into a fast visual language model to obtain a first answer result; in response to the first answer result being marked as an upgraded answer, inputting the current query to be answered and the current memory state into a slow inference model to output a second answer result. In this application, controlled upgrades between the fast path and the slow inference path can maintain real-time constraints while also considering the inference accuracy of complex problems, dynamically improving the performance and efficiency balance of streaming video understanding, and better meeting the usage requirements of streaming video scenarios.

Embodiments of the present disclosure disclose a model training method, a dialogue reply method, devices, equipment and media. A specific embodiment of the method comprises: using a first training data set to perform model training on an initial full-duplex dialogue model to obtain a trained first full-duplex dialogue model; using a second training data set to perform model training on the first full-duplex dialogue model to obtain a trained second full-duplex dialogue model; and using a third training data set to perform model tuning on the second full-duplex dialogue model to obtain a full-duplex dialogue model, wherein the third training data is preference data for inference token granularity. This embodiment is related to artificial intelligence, and using the trained full-duplex dialogue model can achieve dynamic matching of inference granularity and problem complexity, reduce inference token redundancy for simple problems, guarantee inference depth for complex problems, reduce invalid overhead in the dialogue reply process, and improve reply efficiency.

This invention relates to a flexible flow shop scheduling method, system, and computer-readable storage medium based on an environment model, belonging to the field of flexible flow shop scheduling and control technology. This invention decouples the construction of the environment model and the training of the agent. First, the environment model is trained independently using historical scheduling data, enabling it to accurately predict the shop state transitions under given states and actions. Then, the parameters in the environment model are fixed, and the environment model is used as a real interactive environment to interact with the agent. This method not only overcomes the dependence of model-free reinforcement learning on a high-fidelity simulation environment but also solves the convergence difficulties and computational complexity problems caused by the coupling training of the model and policy in traditional model-based reinforcement learning. It combines high sample efficiency with policy flexibility, significantly improving scheduling efficiency and system adaptability, and providing a feasible technical solution for intelligent scheduling of flexible flow shops.

The utility model discloses a steering column upper connecting plate punching device which comprises a base, a shell is fixedly connected to the top of the base, an inner cavity is formed in the shell, and a one-way threaded rod is rotationally connected to the interior of the shell. In the actual use process, the one-way threaded rod in the shell is rotated to drive the connecting block to move, when the connecting block moves to the inner end of the sliding block, the inclined surface of the connecting block is used for pushing the sliding block to move towards the two sides, the sliding block drives the extruding and pushing plate to conduct inner diameter clamping and fixing on the connecting plate body located in the inner cavity, and the spring is in the stretching state in the process; the device solves the problems that an existing connecting plate is unstable in fixation and complex in operation, a traditional mode cannot adapt to connecting plates of different sizes, or the fixing process is tedious, the device can quickly and stably fix connecting plate bodies of different inner diameters through cooperation of all parts, machining convenience and accuracy are improved, a reliable foundation is provided for subsequent machining, and the machining efficiency is improved. And the overall working efficiency is improved.

The application relates to the technical field of data analysis, and provides a manufacturing complex problem data analysis method and system based on a knowledge graph, which comprises the following steps: matching parameter change signs with early performance degradation states of equipment components, integrating cross-field knowledge to assist in judging the action relationship between the parameter change signs and the component performance degradation, and confirming component degradation information; constructing an influence path from equipment operation conditions in processed data to component performance degradation and then to product quality influence according to the component degradation information, the processed data and product quality parameters in the processed data; and generating corresponding prompt information according to the influence path and the degradation degree indicated by the component degradation information. The application has the effect of improving the stability of a manufacturing production line and product quality.

This invention discloses an adaptive differential evolution algorithm for optimizing passive radar station deployment, comprising the following steps: establishing a passive time difference station deployment simulation scenario; step 2: population initialization; step 3: fitness value calculation; step 4: genetic operation. On the one hand, this invention addresses the problem of numerous variable parameters and complex parameter settings in particle swarm optimization (PSO) algorithms by incorporating iteration counts, fitness values, and weights into parameter generation, making the algorithm's search and convergence more reasonable. On the other hand, this invention considers the impact of station location error, time difference error, and communication distance between observation stations on the deployment results during experimental simulation, making the modeling process more realistic.

This invention provides a winch, belonging to the field of binding equipment technology. It solves the problem of complex structure in existing winches. This winch includes a support frame, a reel rotatably connected to the support frame, and an operating sleeve with a mounting bracket fixed to its end. The mounting bracket is rotatably sleeved on one end of the reel. Several pawls are hinged to the mounting bracket, and an elastic element acting on the pawls is also provided on the mounting bracket. Several circumferentially distributed locking holes are formed on the outer peripheral wall of the reel. All locking holes are square holes, and all locking holes penetrate the inner peripheral wall of the reel. The free end of the pawl has an L-shaped locking groove. When the operating sleeve rotates clockwise, the locking groove of the free end of the pawl engages with one edge of the locking hole opening. When the operating sleeve rotates counterclockwise, the free end of the pawl disengages from the locking hole under the pressure of the other edge of the locking hole opening. This winch simplifies the structure while ensuring reliable operation.

The application discloses a dynamic retrieval enhancement generation method based on size model cooperation, and belongs to the technical field of artificial intelligence. The method utilizes a first small language model to perform semantic analysis and structured decomposition on a natural language task locally according to a current upstream and downstream environment, and then utilizes a second small language model to generate each preliminary answer associated with a sub-task sequence according to a locally preset knowledge database, and then utilizes a cloud large language model to verify and correct each preliminary answer, so as to guide the second small language model to generate a final answer of the natural language task, thereby solving the problem that a current end-cloud scheduling mechanism is difficult to adapt to complex and changeable actual application scenarios, and the processing effect of an end-cloud cooperative system on complex problems is poor in a retrieval enhancement generation process.

The utility model discloses a multi -functional section bar, including section bar main part, its cross section is arched structure, the inside center of the vault of arched structure is equipped with along the length direction through distribution's screw hole, the both ends of arched structure converge to form two symmetrical flat mouth part to the inside, the middle part of arched structure is equipped with the support cross board that extends along the length direction, the assembly clamping groove is formed between support cross board and two flat mouth part. Through the structure form that arched structure and its both ends symmetrical flat mouth part, the middle support cross board form assembly clamping groove, combine the arc -shaped guide screw hole that vault is through distribution, solved the cost high, the management complex problem that traditional hanging system needs many section bar independent design leads to, thereby obtains multi -functional integration, installs self -centering positioning and stress even distribution's effect, realizes the monomerization load of main rail support, lamp tube installation and clothes hanger suspension.

The application discloses an Excel data knowledge question and answer method based on table structure semantic cooperation, and belongs to the technical field of intelligent question and answer, and comprises the following steps: defining the hierarchical semantic structure, metadata glossary and task category of a cell, and processing an Excel file into a two-dimensional table; generating a cooperative knowledge body for each cell in the two-dimensional table to construct a composite knowledge base; performing multi-constraint analysis on a user question; generating an evidence set based on constraint-driven hierarchical mixed retrieval; guiding a large language model to generate an answer by using a structured prompt; and checking the answer. The application cooperatively models the semantic and structural relationship of table text to improve the expression capability of the real semantics of Excel data, drives hierarchical mixed retrieval through multi-constraint analysis, improves the retrieval accuracy and integrity under complex problems, introduces an explicit executable reasoning mechanism, improves the accuracy and stability of numerical question and answer, constructs an evidence closed-loop checking and self-correction mechanism, and improves the credibility of the answer.

This invention discloses a power device packaging optimization method based on a BPNN-WOA fusion model, relating to the field of power device thermal management. It addresses the problems of traditional optimization methods, such as neglecting nonlinear coupling of parameters, difficulty in global optimization, and handling complex discrete variables. This method relies on a database of heat dissipation influencing factors and a high-fidelity 3D thermal simulation model. Parameters are selected and standardized to generate a sample dataset, which is then used to train a BPNN thermal resistance prediction surrogate model. The WOA algorithm's whale position vector is mapped to a combination of packaging parameters, and the BPNN output is mapped to a fitness value. The optimal parameter combination is obtained through global iterative optimization using WOA, and verified by simulation after inverse normalization. If the deviation exceeds a threshold, additional samples are added to retrain the model. This method can capture the synergistic effect between parameters, achieving global optimization in a continuous-discrete hybrid design space, significantly reducing packaging thermal resistance, and improving optimization efficiency and accuracy.

This utility model discloses a flipping device for parts processing, including a flipping motor, a hydraulic cylinder, and a swing clamping component driven by a hydraulic rod. The swing clamping component is linked to a clamping plate mounting column via two swing rods. The clamping plate mounting column is detachably connected to a corrugated clamping plate or an arc-shaped clamping plate. During operation, the hydraulic rod pushes the swing base, causing the clamping plates on both sides to open and close synchronously to clamp the workpiece. The flipping motor drives the mounting plate and the workpiece to rotate at any angle. This device adopts a replaceable clamping plate design to adapt to parts of different shapes. The screw and locking nut cooperate to achieve flexible adjustment of the clamping plate angle, solving the problems of poor compatibility and complex operation of traditional devices.

The utility model relates to the technical fields of building engineering discloses a building engineering assembly type retaining wall, including wall and constructional column, the left and right direction of wall one side is seted up and is connected groove, the other side of wall is provided with the boss of being connected with the butt joint groove is adapted, the inner chamber of wall is provided with the strengthening mechanism, is provided with the auxiliary mechanism for improving the connecting stability of multiple wall in the up and down direction on the strengthening mechanism, the outside of wall thickness direction is provided with the protective wall, is provided with the connecting mechanism for improving the connecting stability between wall and protective wall between the protective wall and strengthening mechanism. The utility model discloses building engineering assembly type retaining wall is made by prefabricated technology, and through setting up the boss, the butt joint groove, the auxiliary rod and the auxiliary cover, make multiple " occlusion " point position between wall, to improve the connecting stability between wall, solved the retaining wall on -site processing longer, the process complex problem of hour.

The application belongs to the technical field of problem mathematical solution, and provides a mathematical application question solving method and system based on expression constraint and preference optimization, which comprises the following steps: obtaining a problem of a mathematical application question, a solving expression and a corresponding numerical answer; taking the solving expression as a constraint to guide a first large language model to generate a derivative problem with consistent structure; constructing a complex problem according to the derivative problem, guiding the first large language model to generate multiple candidate solving paths for the complex problem to score; integrating the multiple candidate solving paths based on the scoring results to form a training data set; constructing a preference training sample based on the training data set, training a second large language model through a direct preference optimization strategy to obtain a trained second large language model; inputting a mathematical application question to be solved into the trained second large language model to generate multiple candidate reasoning paths and corresponding candidate results; screening the candidate results to output a final problem solving result.