95results about How to "Optimization parameters" patented technology

The invention provides a photovoltaic power station design system and method based on ecological collaborative optimization, and relates to the technical field of data processing. The system comprises an initial configuration module which is used for generating an initial layout structure of a photovoltaic module, performing energy-saving and consumption-reducing calculation and economic analysis according to generating capacity data, and obtaining initial energy-saving and consumption-reducing data and initial economic analysis data; the secondary optimization module is used for optimizing the parameters of the photovoltaic module based on the economical efficiency analysis data and determining energy-saving and consumption-reducing data and economical efficiency analysis data through the generating capacity optimization data; the ecological evaluation module is used for determining ecological benefit data of the photovoltaic power station according to the ecological influence parameters; and the collaborative optimization module is used for setting a power generation income weight and an ecological benefit weight so as to optimize the photovoltaic module layout structure and generate a photovoltaic power station design scheme based on ecological collaborative optimization. The method is used for realizing collaborative optimization of power generation benefits and ecological benefits of the photovoltaic power station, so that the power station gives consideration to ecological protection while improving the power generation efficiency.

The invention relates to the technical field of hydraulic engineering, geotechnical mechanics and geological disaster prevention and control engineering, and discloses a landslide surge physical and numerical hybrid model construction method and system and application. Analyzing the landslide instability evolution motion process, the landslide warehousing first wave height and the surge propagation process of the landslide surge; performing a numerical simulation test according to a second result of the landslide surge numerical model test; adjusting the particle size and the kernel function in the landslide surge numerical simulation test, so that the first wave height, the opposite-bank arrival height and the surge pressure distribution in the landslide surge numerical simulation test and the landslide surge physical model test are consistent; and taking the adjusted landslide surge numerical model as the obtained landslide surge hybrid model. The system comprises a first result acquisition module, a second result acquisition module, a test parameter adjustment module and a hybrid model acquisition module. According to the method, different calculation working conditions can be conveniently analyzed.

The invention discloses a smart city street lamp intelligent control system and method. The method comprises the following steps: performing multi-aspect inspection on each hardware device through a comprehensive intelligent control module and a communication module; the environment monitoring module is used for collecting multi-source data in real time and integrating and analyzing the multi-source data through the data fusion module; generating a dynamic dimming strategy of the common road section and the high-priority road section according to the collected data and regional requirements; brightness adjustment and on-off control are executed one by one through the single lamp control module, and equipment operation parameters are collected through the real-time feedback and monitoring module. The remote management platform is used for monitoring the running state of the street lamp in real time and dynamically adjusting strategy optimization analysis; different operation scenes are simulated through a system simulation test, actual deployment is carried out in a pilot area, and parameters and operation logic are optimized. Through modular design, intelligent decision making and data driving modes, the management efficiency and the energy-saving level of urban street lamps are remarkably improved, and reliable technical support is provided for smart city construction.

The invention belongs to the technical field of equipment fault diagnosis, and discloses a variable working condition bearing fault diagnosis method based on a domain adversarial hybrid graph convolutional network. According to the specific technical scheme, the method comprises the specific steps that firstly, a convolutional neural network is used for conducting feature extraction on input data, obtained features serve as nodes, secondly, an adjacent matrix is obtained through a graph generation layer, an association graph is constructed according to the input matrix, and thirdly, the association graph is obtained. Inputting the constructed graph into a multi-receptive-field mixed graph convolutional network to extract structural information, capturing local structural information and global structural information of the graph from the perspective of the graph convolutional network and graph sampling and aggregation by using the multi-receptive-field mixed graph convolutional network, and finally, extracting the structural information of the graph from the perspective of the multi-receptive-field mixed graph convolutional network. And fault classification is carried out on the obtained features through a transfer learning method, so that the fault diagnosis capability of the model on a target domain is improved, and the expression capability, the generalization performance and the robustness of the model are remarkably improved.

The invention relates to the technical field of intelligent high-speed vehicle passing control, and provides a method for predicting a vehicle driving track under highway passing control based on digital twin tracking simulation, and the method comprises the steps: collecting highway digital twin system, target vehicle and environment data in real time, and carrying out the integration to form an input data set; a future driving track of a vehicle is predicted by using a pre-trained LSTM deep neural network model, and an actual track is collected through devices such as a GPS and the like for verification. And when the prediction deviation exceeds the preset range, the model is dynamically adjusted to improve the accuracy. Prediction results and deviation analysis are transmitted to a background control terminal and a driver mobile terminal in real time, and traffic control, accident early warning and driving aid decision making such as front road condition prediction and suggested driving speed providing are supported. The highway traffic efficiency and safety are effectively improved, the driver experience is enhanced, the development of an intelligent traffic system is promoted, the operation cost is reduced, and remarkable social and economic benefits are achieved.

The invention provides an intelligent laboratory man-machine interaction control method and system, and relates to the technical field of control, and the method comprises the steps: collecting experimenter and equipment images through a visual collection device, recognizing equipment operation state parameters based on a deep learning model through a processing server, including a switch state, an indicating lamp state, a display screen state and a fault indication state, and transmitting the state parameters to a server; and comparing the identification result with a preset parameter to generate an evaluation result. And the processing server receives the user voice instruction, extracts the content of the control instruction, and determines the target equipment and the control action. And verifying the legality of the instruction and judging whether an execution condition is met by combining an equipment state evaluation result. And a cooperative control strategy is generated according to the control instruction and the evaluation result, the cooperative control strategy comprises a control time sequence, parameters and a data interaction mode, and the control instruction is sent through the field bus network to realize multi-device cooperative control. According to the invention, intelligent management and control of laboratory equipment are realized, and the experiment efficiency and safety are improved.

The invention relates to the technical field of electronic packaging, and discloses an intelligent dispensing control method and system based on machine vision and a storage medium. The method comprises the following steps: acquiring a dispensing part image, and performing feature extraction on the dispensing part image to obtain feature information; according to the feature information, carrying out dispensing position area calculation to obtain the area of a dispensing area; calculating the surface roughness of the dispensing part according to the feature information of the image to obtain roughness information; according to the feature information, the dispensing position and the roughness information, the dispensing amount is calculated, and dispensing parameters are obtained; according to the dispensing parameters and the roughness information, the dispensing amount is adjusted, and the effective dispensing amount is obtained; according to the dispensing parameters, the effective dispensing amount and the dispensing area, dispensing operation is carried out, and a dispensing result is obtained; and according to the dispensing result, the dispensing quality is monitored and adjusted based on visual feedback control. According to the method, the dispensing precision of the dispensing machine can be improved, parameters are optimized, and efficient dispensing control is achieved.

The invention provides an auxiliary pathological analysis method for a microscope system, and belongs to the technical field of pathological analysis, and the method comprises the steps: firstly carrying out the gray normalization, denoising and other preprocessing of an obtained pathological section image, and carrying out the illumination equalization processing through a region adaptive histogram equalization algorithm; then, color features of the image are extracted in an LAB color space, a color representation center is calculated, a feature mapping matrix is established, and meanwhile texture features are extracted and fused with the color features; and carrying out dimension reduction on the fused feature vector by using a principal component analysis algorithm, and reserving important feature information. And then, establishing a convolutional neural network model, inputting the feature vectors after dimension reduction, performing training and cross validation by using a labeling data set, and improving the model performance through iterative optimization. According to the invention, the technical problem that the diagnosis and judgment of a doctor are seriously influenced due to inconsistent color expression caused by the difference of shooting conditions of pathological section images observed by a microscope is solved.

The invention discloses a transformer fault diagnosis research method and system, and relates to the technical field of transformer fault diagnosis, and the method comprises the steps: introducing an SHAP value to carry out the feature optimization of transformer fault data based on a decision tree and an integration method thereof; dividing the optimized data into a training set and a test set, and performing data cleaning and oversampling processing on the training set; and optimizing the deep mixed kernel extreme learning machine based on the improved dung beetle algorithm, constructing a transformer fault model, and performing fault diagnosis by using the transformer fault model. According to the feature system provided by the invention, the input features of the transformer fault diagnosis model can be effectively optimized, transformer fault data can be effectively utilized, and the influence of redundant features on the diagnosis model is reduced. Compared with other heuristic artificial intelligence algorithms, the improved dung beetle algorithm provided by the invention has more excellent optimization ability, and can find more excellent parameters of the deep mixed kernel extreme learning machine in the same number of iterations and range.

The invention relates to the technical field of spring steel treatment, and particularly discloses an ultrahigh-strength spring steel quenching and tempering treatment process, which is characterized in that a temperature monitoring system is arranged for monitoring a temperature rise stage in spring steel quenching and tempering treatment; the method specifically comprises the following steps that firstly, real-time temperature data of the temperature rising stage in spring steel quenching and tempering treatment are obtained, and analysis and judgment are conducted; according to the method, the temperature data in the temperature rising stage are obtained in real time, the deviation between the temperature data and the standard temperature is calculated, namely the temperature deviation value, the temperature deviation value is compared with the first temperature deviation threshold value and the second temperature deviation threshold value, and a temperature normal signal, a to-be-analyzed signal or a temperature abnormal signal is recognized and generated; therefore, abnormal conditions in the heating process can be found in time and processed, and meanwhile, the to-be-analyzed signals can be further processed and analyzed, so that parameters in the heating process are optimized, and the processing effect is improved.

The present invention discloses a friction compensation control method for an electro-hydraulic servo system. The method first collects signals such as system flow, oil supply pressure, acceleration, etc. of the electro-hydraulic servo system, models the system friction using an improved LuGre friction model, identifies its parameters offline using a genetic algorithm, and uses the estimated value generated by the identification model for feedforward compensation; then, a hyperbolic sine function with better smoothness is introduced to design the nonlinear state error feedback and extended state observer part of the variable coefficient self-disturbance rejection controller; and on this basis, a fuzzy adaptive control method is combined to adjust the nonlinear state feedback coefficient through error and error differential; the "contradiction between rapidity and overshoot" is solved, and the stability and finite time convergence of the system are guaranteed. This method effectively improves the system's compensation capability for friction nonlinearity, improves the system's motion performance, and improves the system's tracking accuracy and anti-disturbance capability.

The invention discloses a rusted high-strength bolt connection pretightening force intelligent self-optimization prediction method and system, and belongs to the technical field of bolt pretightening force detection.The method specifically comprises the steps that a finite element database of bolt connection pretightening force loss is established, and a machine learning prediction model based on a neural network is trained; when the recognition accuracy of the machine learning prediction model reaches a preset value, pre-tightening force loss of the target bolt is monitored in real time, and real-time geometric feature information of connection of the target bolt is obtained; when the similarity between the real-time geometric feature information of the target bolt connection and a finite element model in the finite element database is greater than or equal to a preset value, calling the pre-tightening force loss of the existing finite element database; when the similarity is lower than a preset value, the trained machine learning prediction model is utilized to output actual bolt connection pre-tightening force loss in real time; the bolt looseness detection device realizes automatic and marker-free bolt looseness detection, is high in bolt image acquisition angle and distance tolerance, and can meet bolt looseness detection at different positions.

The invention discloses a general text matching form method and system based on a large language model, and relates to the technical field of natural language processing, the method comprises the following steps: obtaining and preprocessing text data of a user side, and defining an input field of a form based on the preprocessed text data; in combination with the preprocessed text data and the defined form input field, cue words are obtained, a pre-constructed large language model is used for analyzing the cue words, analysis results are obtained, and a semantic similarity algorithm is used for matching the analysis results; and outputting a matching result according to a preset format, and verifying the output result to optimize the large language model. According to the invention, through text data acquisition and preprocessing, the text processing efficiency is improved; meanwhile, the form input field is defined based on the preprocessed text data, so that the form can more accurately reflect the structure and content of the text data, the flexibility of form definition is enhanced, and the requirements of different application scenes are met.

The invention discloses an image processing system based on artificial intelligence, and relates to the technical field of image processing, a user inputs an original image into the system by using an image input module, the system receives original image data and displays the image by using a display module, and the image processing module adjusts the original image based on the original image data by using an image adjustment processing module. After brightness and contrast adjustment, noise suppression, edge enhancement and histogram equalization are carried out, the quality of an image displayed on a display module is gradually improved, the defects of a traditional method are overcome, a more intelligent, efficient and high-quality image processing effect is achieved, and particularly, when complex and changeable images are processed, the image quality is improved. The system can automatically analyze the image content and dynamically adjust the processing parameters so as to achieve the optimal processing effect, the image processing system based on artificial intelligence brings a new thought and method for the image processing field, and the system has wide application prospects and important research value.

The invention relates to a gearbox fault diagnosis method and system. The method comprises the following steps: acquiring fault data of a gearbox; performing feature extraction by using the convolutional neural network fault data to obtain fault features; inputting the fault features into an extreme gradient lifting classifier to obtain a fault diagnosis result of the gearbox; wherein the extreme gradient lifting classifier adopts an improved big sugarcane-mouse algorithm to carry out parameter optimization, and the improved big sugarcane-mouse algorithm is combined with a nonlinear weight, an adaptive factor, sine and cosine and a sparrow warning mechanism. According to the invention, the features of the gearbox fault data are mined by using the convolutional neural network, accurate fault classification and identification are carried out on the basis of the features by using the extreme gradient lifting classifier, and the hyper-parameters of the extreme gradient lifting classifier are optimized by introducing the improved big sugarcane-mouse algorithm. And the optimization capacity, the stability and the convergence speed are improved, so that the overall performance of the model is effectively improved.

The invention relates to the technical field of cross-modal retrieval, and discloses a cross-modal retrieval model training method and a cross-modal retrieval method.The training method comprises the steps that a cross-modal retrieval model is constructed; training the cross-modal retrieval model by adopting a low-parameter training method; integrating a Hash layer in the cross-modal retrieval model to learn Hash mapping; constructing a sorting order among training samples of the sorting perception loss optimization input cross-modal retrieval model; the retrieval method comprises the following steps: inputting a query sample into a cross-modal retrieval model; performing feature extraction on a query sample by the cross-modal retrieval model; the cross-modal retrieval model converts the extracted high-dimensional features into low-dimensional and continuous binary-like codes; the binary-like codes are processed and converted into binary codes; comparing the similarity between the binary code of the query sample and the binary code of a storage sample in a database; and outputting a retrieval result semantically related to the query sample. The method is high in retrieval efficiency and good in retrieval performance.

The invention belongs to the technical field of bearing performance testing, and particularly relates to an offset test bench for testing comprehensive performance of a tilting-pad radial sliding bearing under high-temperature and high-specific-pressure working conditions, which is used for testing the comprehensive performance of the tilting-pad radial sliding bearing (test bearing) under the high-temperature and high-specific-pressure working conditions. Comprising a rotor shaft, a driving system, a loading system, a lubricating system and a testing system, the driving system is used for driving the rotor shaft to rotate; the loading system is used for loading a load to the test bearing; the lubricating system is used for supplying temperature-adjustable lubricating oil to the accompanying test bearing, the loading bearing and the test bearing; the test system is used for acquiring test data. Compared with the prior art, the test device solves the problems that a test device in the prior art cannot reach high-temperature and heavy-load test conditions, so that a bearing performance result under a high-temperature and high-specific-pressure extreme working condition (shot by a missile) cannot be obtained; the tilting-pad sliding bearing high-temperature heavy-load test bench constructed by the scheme realizes high-temperature heavy-load working condition assessment of the bearing.

The invention provides a detection method and system for a deeply-forged face image, and the method comprises the steps: constructing a face image data set and a deeply-forged image detection model which comprises a router, an expert model and a Transform module; the face image data set is adopted to carry out joint training on the router, the expert model and the Transform module; and inputting a to-be-detected face image into the trained deep counterfeit image detection model to obtain the probability of whether the to-be-detected face image is a deep counterfeit image or not. According to the method, the robustness and generalization ability of deep counterfeiting detection are enhanced, and the weakness of the existing method in facing diversified counterfeiting technologies and antagonism disturbance is solved, especially for complex propagation scenes and environments with many confrontation factors; by combining the multi-expert model and the Transform module, the accuracy and robustness of detection are improved, and the generalization ability and adaptability of the model to different types of forgery are improved.

The invention relates to the technical field of titanium tetrachloride production, in particular to a method for recovering vanadium from titanium tetrachloride vanadium removal slag and application of the method. The method for recovering the vanadium from the titanium tetrachloride vanadium-removed slag comprises the following steps: (a) carrying out acid liquor leaching on the vanadium-removed slag to obtain vanadium-containing leachate and leaching slag; (b) adding an oxidizing agent into the vanadium-containing leachate for oxidative hydrolysis and pyrolysis vanadium precipitation to obtain a precipitate and a vanadium precipitation mother solution; (c) leaching the precipitate by using alkali liquor, precipitating vanadium by using a precipitant, and roasting to obtain vanadium pentoxide; and (d) the vanadium precipitation mother liquor is subjected to extraction and reverse extraction, and a vanadium-containing solution is obtained. According to the method for recovering the vanadium from the titanium tetrachloride vanadium removal slag, corrosive gases such as HCl and Cl2 are not generated in the preparation process, a large amount of strong alkaline waste residues are not generated, the method has the advantages of being economical and environmentally friendly, the recovered vanadium pentoxide is good in quality and low in impurity content, and the total recovery rate of the vanadium is high.

The utility model provides a lengthened valve rod high platform hard sealing ball valve which comprises a valve body, a heightening support and a valve rod, a packing groove is formed in the side wall of a rod hole, a first containing groove is formed in the top of the valve body, a limiting ring is arranged at the bottom of the heightening support, a first hard sealing ring is arranged in the first containing groove, and a second hard sealing ring is arranged at the bottom of the packing groove. The outer ring of the second hard sealing ring abuts against the side wall of the packing groove, and the inner ring of the second hard sealing ring abuts against the valve rod. Through cooperation of the first hard sealing ring, the first containing groove and the limiting ring, the heightening support is sealed, the first hard sealing ring can provide a better limiting function, the heightening support is limited, the situation that the heightening support is low in price in the installation process is avoided, and the centrality of the heightening support is improved. And on the basis of ensuring the centrality of the heightened bracket, the valve rod is limited through the design of the hard sealing ring II, so that the valve rod has better centrality, and various related parameters of the valve are improved.

The invention discloses an inspection parameter optimization method and device based on cloud side cooperation, and relates to the technical field of inspection parameter optimization, and the method comprises the steps: receiving an inspection instruction sent by a cloud, determining an inspection target according to the inspection instruction, determining target terminal equipment according to the inspection target, obtaining a communication code of the target terminal equipment, and transmitting the communication code to the cloud side cooperation. Establishing an exclusive data transmission channel according to the communication code, receiving state information uploaded by the target terminal device from the exclusive data transmission channel, determining an operation state of the target terminal device according to the state information, and when the operation state is an abnormal state, determining an abnormal amplitude of the operation state, and generating an optimization parameter from the abnormal amplitude through a parameter optimization model, and optimizing the operation parameter of the target terminal based on the optimization parameter. Through the above mode, remote inspection of a plurality of terminal devices at different positions can be realized, and parameters of each terminal device can be optimized.

The application relates to the technical field of sewage treatment, in particular to a pressure dosing method and system for sewage treatment. The application first establishes a stable gas pressure environment in a dosing device, and uses a high-precision pressure sensor to collect pressure data in real time; based on Bernoulli equation, a total pressure-dosing flow relationship model and a dosing pipe static pressure-dosing flow relationship model of the dosing device are constructed, and the applicable flow intervals of the two models are determined; after a target flow value is input, the corresponding model is selected according to the interval to calculate the target total pressure or static pressure value; the system adopts a hierarchical PID control strategy, a primary PID is responsible for total pressure adjustment, and a secondary PID is responsible for static pressure fine adjustment, so that the rapidity and accuracy of pressure control are cooperatively ensured; meanwhile, the actual dosing amount is continuously monitored through a flow measurement unit, and when the flow deviation exceeds the set range, a calibration program is automatically started, and the pressure-flow model parameters are optimized in real time; finally, through closed-loop control of pressure and flow, high-precision adjustment of the dosing flow is realized.

A direct-current energy-consuming apparatus, a method and a device for controlling a direct-current energy-consuming apparatus, and a storage medium are provided. The direct-current energy-consuming apparatus includes an energy-consuming resistor (1), a power module (2), an inductor (3) and a control module (4). The inductor (3) is arranged in a branch formed by the energy-consuming resistor (1) and the power module (2) connected in series. The control module (4) is configured to switch off the power module (2) when it is detected that a voltage between two ends of the direct-current energy-consuming apparatus is greater than a first voltage threshold, and switch on the power module (2), when it is detected that a voltage between two ends of the direct-current energy-consuming apparatus is less than a second voltage threshold, where the two voltage thresholds are determined based on a rated voltage of a corresponding direct-current power transmission system. The inductor (3) is arranged in the branch of the direct-current energy-consuming apparatus, parameters of a main circuit of the direct-current energy-consuming apparatus are optimized, and the power module (2) is controlled to be switched on and off, reducing a current change rate on the energy-consuming resistor (1), reducing the impact on the direct-current bus, and significantly reducing capacitor consumption, thereby achieving the advantages of high performance, high reliability and low cost.

The invention provides a low-delay wave beam direction finding technology realized based on an FPGA. The low-delay wave beam direction finding technology comprises the following steps: establishing a wave beam direction finding simulation mathematical model according to a wave beam direction finding algorithm in combination with specific hardware parameters and indexes and feature information of a to-be-detected flight target; converting the established wave beam direction finding simulation mathematical model into an FPGA program; the method comprises the following steps: establishing a hardware test platform, and programming an FPGA (Field Programmable Gate Array) program into an FPGA baseband board; and performing operation according to the direction testing process to complete direction testing. Compared with the prior art, the delay caused by direction finding is greatly reduced by virtue of the strong parallel processing capability of the FPGA chip and the high-speed clock, and meanwhile, the resolution of direction discrimination is improved. Through application of the FPGA and accurate mathematical modeling, the method has significant advantages in the aspects of direction finding precision, real-time performance, system expansibility, anti-interference capability and the like, and can effectively cope with complex flight target direction finding tasks. Therefore, the technology has wide application prospects in the fields of aerospace, military reconnaissance, unmanned aerial vehicles and the like.

The invention discloses a tool wear monitoring method based on IWOA-GRU, belongs to the field of metal cutting monitoring, and aims to solve the problem of low tool wear state monitoring precision caused by incapability of capturing a long-term dependency relationship in the prior art. The method comprises the following steps of 1, collecting tool machining process data to obtain an original data set; 2, processing the original data set, and dividing the original data set into a training set and a test set; 3, inputting the training set into the GRU model for training; 4, optimizing a weight matrix of the GRU model by adopting an IWOA whale population algorithm; and 5, the trained IWOA-GRU model is used for monitoring tool wear in the tool machining process, and the tool wear state is judged according to the tool wear value Yt output by the model.

The present invention discloses a seedling root-soil separation test bench, comprising a test frame, a seedling clamping and conveying device, a soil shaking device and a soil trough platform, wherein the seedling clamping and conveying device is arranged at the upper part of the test frame, the seedling clamping and conveying device clamps the branches and trunks of the seedlings and conveys them to the rear, the soil shaking device is arranged at the middle part of the test frame, the soil shaking device shakes and pats the root soil on the seedling root system to make it shake off, the soil trough platform is arranged at the lower part of the test frame, and the soil trough platform collects the fallen soil on the seedling root system. The present invention can conduct comprehensive tests on different forms of soil shaking mechanisms, obtain the optimal soil shaking mechanism parameters and reduce the seedling root system damage rate through the test, and provide an experimental basis for the design of the subsequent soil shaking mechanism of the seedling lifting equipment.

The invention relates to a black box classification interaction intervention method, electronic equipment and a storage medium, and the method comprises the steps: firstly carrying out the operation of a black box network classification model through a concept bottleneck model which can be understood by a user, then carrying out the intervention correction of the prediction interpretation of the concept bottleneck model based on the understanding of the user, and then carrying out the intervention correction. After the concept weight matrix of the concept bottleneck model is corrected, the original black box network is subjected to reverse distillation, and error correction knowledge is migrated; according to the method, the black box classification model can be explained, the human-neural network mutual understanding framework is constructed according to the explanation, and based on the framework, interactive error correction of the human and the neural network can be realized.

The present invention discloses a stability analysis method and system for a torque motor drive system, and specifically relates to the technical field of torque motor drive. The method gradually increases and decreases the time delay in a simulation environment to obtain response change data of a controller output and a motor drive system input, evaluates the response sensitivity of the controller under different time delay conditions, and evaluates the response rate stability of the motor drive system in combination with a load mutation scenario. The real-time performance of the controller response to the motor drive system is evaluated through a machine learning model. When the real-time performance is low, an abnormality detection mechanism is used to identify unstable time periods, and controller parameters are dynamically adjusted, thereby effectively reducing oscillation, over-modulation and instability problems caused by time lag effects, improving the stability and reliability of the motor drive system in high-speed or precision control applications, and significantly improving the real-time response capability of the controller in a dynamic environment to avoid system loss of control or damage.

This invention provides a method and system for constructing a multi-agent cooperative interception decision model in the field of agent control. The method includes determining the actions that multiple agents independently execute at the current decision moment; updating the relative motion state based on the actions to obtain the relative motion state at the next decision moment and determining the individual immediate reward; determining the total reward of the agent after executing the action based on the individual immediate reward of the agent and the individual immediate rewards of neighboring agents; generating an experience tuple for the agent at the current decision moment by combining the state, action, total reward, and state at the next moment; calculating a loss function based on the experience tuple and optimizing the parameters of the agent's deep Q-network to obtain an optimized deep Q-network; and determining the multi-agent cooperative interception decision model based on the optimized deep Q-network of each agent. This method improves the efficiency of constructing the cooperative interception decision model.

The embodiment of the invention provides an Internet information pushing method and system based on big data service, and the method comprises the steps: obtaining an Internet interaction data sequence of a basic template and a business derivation template from the big data service, and carrying out the parameter learning and optimization of an Internet interest prediction network through the data. And finally, a target Internet interest prediction network is generated, so that the accuracy of interest point prediction is improved, and accurate grasp of personalized demands of the user is realized. By generating and fusing various interactive knowledge graph features, the network parameters are further optimized, so that the pushed information better meets the actual interests and requirements of the user, and the Internet information pushing efficiency and the user satisfaction are remarkably improved.