143results about How to "Improve dynamic adaptability" patented technology

The invention discloses an accurate dosing control system for water treatment, relates to the technical field of dosing control, and is used for solving the problem of poor dosing control accuracy in the water treatment process. Comprising a data acquisition module, a condition library construction module, a multi-target processing module and an optimization decision module. According to the method, a sensor is arranged in a water body to collect pollutant parameters, data are preprocessed, the types and the concentrations of pollutants are recognized through a support vector machine model, after recognition, an optimal processing parameter combination is called from a removal condition library, the condition library is dynamically updated by using an Apriori algorithm, and when multiple pollutants coexist, the optimal processing parameter combination is called from the removal condition library. Removal conditions are automatically adjusted according to interaction rules among pollutants, collaborative removal is realized, an optimal decision is solved by constructing a multi-objective optimization model with removal efficiency, medicament cost, processing time and environmental adaptability as targets and combining a genetic algorithm and a particle swarm optimization algorithm, intelligent dosing control is realized, and the system is suitable for large-scale popularization and application. And the accuracy and efficiency of water treatment are improved.

The invention provides a multi-modal intelligent medical scene dynamic reasoning method and device based on medical knowledge, and the method comprises the steps: obtaining multi-modal data of a user, and extracting the multi-modal data as multi-modal features; inputting the text modal features into a graph neural network preset with a medical knowledge graph for feature enhancement to obtain medical knowledge enhanced text features, and inputting the text modal features into a memory network stored in a user personalized health file for feature enhancement to obtain file enhanced text features, splicing the medical knowledge enhanced text features and the file enhanced text features to obtain context enhanced text features; inputting the context enhanced text features and the multi-modal features into a pre-trained multi-modal fusion model to output fusion features; and inputting the fusion features into a dynamic reasoning tree constructed by a plurality of reasoning nodes in a graph structure to determine an optimal reasoning node, and reasoning the fusion features by using the optimal reasoning node to realize dynamic reasoning adaptive to the medical scene.

The invention discloses a multi-mode adaptive fusion navigation positioning method and system in a satellite navigation denial environment, the system adopts GNSS / IMU / visual sensors, and the method comprises the following steps: receiving data of the three sensors and completing modeling; initializing the three sensors; judging the strength of the GNSS signal by utilizing chi-square test; optimization variables are set, a residual matrix is constructed, and nonlinear optimization is carried out; setting a GNSS trust factor when the GNSS signal is weak or denial, and calculating a weight; and weighting and fusing sensor data, outputting optimal state estimation of the system, and realizing continuous navigation and positioning. According to the method, a self-adaptive weight adjustment mechanism is established, GNSS trust factors are introduced, and the weights of multi-sensor data are adjusted in a self-adaptive manner, so that the positioning difficulty under the conditions that the GNSS is subjected to deception interference, suppression interference, building blocking and the like can be overcome, and continuous navigation positioning of an unmanned system (such as a robot, an unmanned vehicle, an unmanned aerial vehicle, an unmanned ship, an unmanned underwater vehicle and the like) is realized.

The invention discloses a function calling method, a related method, a related device, equipment and a storage medium, and the function calling method comprises the following steps: obtaining a historical calling record, and obtaining a preset function matched with a current demand description as a current function; wherein the current demand description is a user demand description currently processed by the agent, and the historical call record comprises a historical function finally called by the agent processing historical demand description; obtaining function similarity between the current function and each historical function; and obtaining a finally called target function for processing the current demand description based on a historical function selected by the function similarity. According to the scheme, the function call for the historical demand can be referred to when the new demand is processed, and the learning ability and the execution efficiency of the intelligent agent can be improved.

The invention relates to a distributed light storage energy control method, and the method comprises the steps: firstly collecting effective illumination data of all photovoltaic modules, constructing an illumination data set through combining historical data, collecting environment data, and associating the environment data with the illumination data, thereby forming an environment influence data set; through a machine self-learning algorithm, an environment influence data set is analyzed, an environment influence factor is generated, the conversion efficiency of a photovoltaic module is predicted, energy conversion data of an inverter and historical input data of an energy storage device are collected, a light energy conversion data set is constructed, a conversion efficiency factor is generated, and the photovoltaic module is analyzed by utilizing the predicted environment data and the environment influence factor. According to the method, effective illumination data is predicted, expected energy storage data is calculated, whether the expected energy storage data meets the power demand or not is judged according to the power demand data, and a power supply distribution strategy is generated, so that the distributed light storage energy system is regulated and controlled, and the energy effective utilization rate of the distributed light storage energy is improved.

The invention discloses a real-name system information management method supporting multi-level authentication, and the method comprises the following steps: collecting basic identity information of a user, and storing the basic identity information through an asymmetric encryption algorithm; according to the use demand and the access authority requirement of the user, generating a multi-level authentication hierarchy; dividing the real-name system information into a basic information area, an authority information area and a sensitive information area according to the sensitivity of the real-name system information; carrying out distributed storage on real-name system information, and carrying out hierarchical encryption on data of different information areas by using a hierarchical encryption strategy; based on a multi-level authentication result and in combination with an access control strategy, authority distribution is carried out on the user, and controlled access of real-name system information is carried out under multi-dimensional verification; user information operation behaviors are monitored in real time, user behavior characteristics are analyzed based on an anomaly detection model, and when risk behaviors are recognized, an additional verification process is triggered or user operation is directly limited. According to the invention, safer, more flexible and more efficient real-name system information management is realized.

The invention relates to the technical field of land resource management, in particular to a land resource management method and system based on big data analysis. The method comprises the following steps: acquiring multi-source land resource acquisition data; carrying out asynchronous parallel data acquisition on the multi-source land resource acquisition data so as to generate multi-source land resource real-time acquisition data; performing request-response optimization on the multi-source land resource real-time acquisition data to generate a self-optimization transmission parameter set; performing precision error correction on the standard multi-source land resource real-time data by using the self-optimization transmission parameter set to generate land resource correction data; performing nonlinear feature enhancement processing on the land resource correction data to generate a nonlinear enhanced feature matrix; and performing dimension compression processing on the nonlinear enhanced feature matrix to generate a multi-dimensional spatial-temporal feature matrix. According to the invention, the comprehensiveness and accuracy of land resource management are improved by improving data acquisition, processing, prediction and planning methods.

The present invention relates to the field of real-time data processing technology, specifically a distributed real-time data stream processing system and method, the system includes a state mapping modeling module, a load dynamic calculation module, a node task adjustment module, and a resource scheduling execution module. In the present invention, by analyzing the data flow change sequence, identifying the frequency and trend of state switching, realizing dynamic identification and classification of flow states, improving the monitoring and understanding capabilities of real-time data, accurately quantifying node resource pressure in resource occupancy and load range analysis, enhancing the predictability of load balancing, providing a scientific basis for resource scheduling, combining task priority sorting and dynamic task adjustment, optimizing node computing resource utilization and task processing response speed, realizing flexible configuration of memory and computing resources, alleviating peak load pressure, improving data processing efficiency and system stability, and improving the system's adaptability and reliability in processing large-scale data flows.

The invention relates to a TPA-LSTM-based shunting area wet and slippery road surface vehicle type-based dynamic early warning and speed guiding method. The method comprises the following steps: S1, collecting traffic flow operation time sequence data in a highway shunting area wet and slippery road surface scene; s2, predicting speed limit values of different vehicle types for safe passing on the wet and slippery road surface in the shunting area; s3, formulating early warning information release content; s4, determining an early warning information issuing mode; s5, determining an information issuing position of the highway shunting area, wherein the information issuing position is comprehensively determined according to an existing shunting area ETC portal layout principle in combination with a warning information issuing standard; and S6, early warning information is pushed to the vehicle end, and the ETC door frame receives the optimal passing speed information of the wet and slippery road surface pushed by the traffic management background and issues the information to vehicle-mounted OBU equipment with a voice broadcast function of the vehicle. According to the method, the real-time performance, the dynamic adaptability, the personalized early warning and the driver response effect are remarkably improved, so that the occurrence rate of accidents in the diversion area in a wet and slippery road scene in rainy days is greatly reduced.

The invention discloses a directional antenna prediction method and system based on a base station position and an air route, and relates to the technical field of wireless communication network optimization, and the method comprises the steps: collecting base station information, building and updating a base station database, obtaining the position of an unmanned plane in real time, and matching the position of the unmanned plane with the position of the base station. The position of a real base station is approached through iterative calculation by using MDT data or network measurement data; and predicting a communication demand and RSRP, and obtaining an optimal antenna direction according to a directional antenna prediction result. According to the directional antenna prediction method based on the base station position and the route, the dynamic adaptive capacity of a system to the flight path of the unmanned aerial vehicle is improved by obtaining the position information and the route track of the unmanned aerial vehicle in real time, the quality and stability of data transmission are improved through a base station approximation algorithm, and the directional antenna prediction accuracy is improved by adjusting the pointing direction of the holder type antenna in real time. The reliability and efficiency of task execution of the unmanned aerial vehicle are improved, and better effects are achieved in the aspects of accuracy, stability and reliability.

The invention discloses a collaborative monitoring system and method for a digital transformation process, and relates to the field of digital transformation monitoring, and the method comprises the steps: obtaining all data needing digital transformation; performing digital transformation according to an actual business process, analyzing a data relationship according to a preset process mining algorithm, constructing a business process model, and performing digital verification to generate a digital verification score; when the digital verification score is greater than a preset threshold value, verification is passed, and the collaboration degree between the activities is calculated; and performing anomaly analysis on the real-time data stream, when an anomaly is identified, marking the data of the activity of which the collaboration degree between the abnormal data and the activity is greater than a threshold value, and generating an early warning signal. According to the method, integration and multi-level association analysis of multi-source heterogeneous data are realized by constructing a direct association relationship matrix and a causal relationship set, and the accuracy and the flexibility of collaborative degree calculation among business activities are comprehensively improved by combining digital verification scores and a dynamic adaptation model.

The invention discloses a digital twin city energy management system based on the Internet of Things, which relates to the technical field of intelligent management, comprises a construction module, an analysis module and a management module, constructs a heat supply topological structure, constructs a node flow balance equation and a pressure difference balance equation, obtains a heat supply flow equation set and a theoretical historical load value of a corresponding node, and obtains the theoretical historical load value of the corresponding node; and constructing an urban heat supply digital twinborn model, comparing the theoretical current load value with the corresponding theoretical historical load value, executing a first operation, and sending a solution strategy. Through the digital twin technology, the efficiency and accuracy of energy management are improved, virtual testing and verification are carried out through a digital twin model, operation decisions and control strategies are optimized, through prediction and simulation, energy waste is reduced, the operation cost is reduced, the dynamic adaptive capacity of the system is improved, and through optimization of energy use, the energy management efficiency is improved. And sustainable development and environmental protection targets of cities are supported.

The invention provides a chronic disease management-oriented dynamic causal model construction and application method, which comprises the following steps of: continuously acquiring multi-modal dynamic health data of a patient, taking the multi-modal dynamic health data as a node set, taking a causal relationship among the multi-modal dynamic health data as an edge set, and establishing a dynamic causal model; the causal intensity between the multi-modal dynamic health data is used as a weight matrix, and a dynamic causal model is constructed based on the node set, the edge set and the weight matrix; for the new multi-modal dynamic health data, calculating marginal causal effect values of the current new multi-modal dynamic health data and all node sets in the dynamic causal model, and updating the dynamic causal model based on the marginal causal effect values; for each time step, the comprehensive causal weight of each side of each time step is calculated according to the long-term component features and the short-term component features of the multi-modal dynamic health data corresponding to the same side, the dynamic causal model is updated based on the comprehensive causal weight, and an application method for the dynamic causal model is provided.

The invention relates to the technical field of sports science and technology, in particular to an AI-based football testing method and system, which comprises the following steps of: training a data set by constructing a plurality of decision trees by adopting a random forest algorithm on the basis of historical match data, randomly selecting a feature subset, and determining an optimal tactical scheme through a voting mechanism, and analyzing the relationship between the tactical effect and the winning rate of the team, and generating a tactical effect and winning rate correlation analysis result. According to the method, historical competition data are deeply analyzed through the random forest algorithm, the complex relation between the strategy effect and the winning rate is revealed, data support of strategy decision is enhanced, the Bayesian network optimizes role allocation according to part of technical characteristics and historical performance, team performance is improved, and a nonlinear dynamic model theory analyzes option and strategy interaction. The enhanced deep learning method adapts to match changes through a deep neural network, and the strategy and configuration flexibility and the ability of the whole queue to cope with complex match conditions are improved.

The invention relates to the technical field of networking methods, in particular to an ad hoc network dynamic networking method and system based on an MAC layer, and the method comprises the steps: obtaining the topological relation information of each node in a network; obtaining distance information between the nodes; generating a topology sensing matrix based on the topological relation information; calculating a priority matrix according to the topology sensing matrix; evaluating the service capability of each node based on the priority matrix and the distance information; calculating the reliability of each node according to the service capability; based on the reliability, time slots are distributed for all the nodes; generating a routing optimization matrix according to the priority matrix, the reliability and the distance information; the priority, the service capability, the reliability, the time slot allocation result and the routing optimization information of each node are output, and by introducing an innovative priority matrix calculation method, the direct connection relationship of the nodes is considered, and the positions and the influence of the nodes in the whole network are captured, so that the network resource allocation is more reasonable, and the routing optimization efficiency is improved. And the overall throughput of the network is obviously improved.

The invention relates to the technical field of semiconductor test equipment, in particular to a full-automatic high-precision probe station, which utilizes multi-physics field coupling sensing and compensation to remarkably improve the positioning precision: by comprehensively collecting multi-physics field data such as a temperature field and a stress field and combining a coupling algorithm to optimize a compensation strategy, multi-factor interference is effectively counteracted, and the accuracy of the probe station is improved. The positioning repetition precision can reach + / -0.05 [mu] m, and the test requirement of the advanced semiconductor technology is met; a dynamic three-dimensional modeling technology and dynamic adaptability enhancement are achieved, a dynamic three-dimensional modeling system with the size affected by multi-physical field data can update the form of a test target in real time, test deviation caused by target deviation or surface fluctuation is avoided, and the adaptability of equipment to a complex test scene is improved. The automation degree is high, the testing efficiency is improved, the whole process of data acquisition, modeling, compensation and testing can be completed without manual intervention, the calibration process is simplified, the testing period is shortened, and the method is suitable for batch detection scenes.

The invention relates to the technical field of smart home / smart home, and discloses a computing device scheduling method, a computer readable storage medium and an electronic device. The computing device scheduling method comprises the steps of obtaining actual device parameters of computing devices in a task execution system; the actual equipment parameters represent the parameters reflecting the actual operation state of the computing equipment in the actual operation environment of the computing equipment; decomposing a to-be-executed task to obtain a plurality of sub-tasks and a task logic relationship among the sub-tasks; respectively issuing the sub-tasks to target computing equipment according to the actual equipment parameters; and controlling the target computing equipment to execute the corresponding sub-task according to the task logic relationship between the sub-tasks. The resource utilization rate is increased, overload or resource waste is avoided, and meanwhile the task execution speed is increased.

The invention discloses a multi-delivery robot path collaborative optimization system and a scheduling method, and aims to solve the problems of non-uniform scheduling load, unsmooth path, weak conflict avoidance capability and poor dynamic adaptability of existing multi-robot scheduling. The system comprises a central scheduling module, a distribution robot terminal, an environment sensing module and a map management module, and realizes task analysis, state monitoring, path planning and conflict coordination. According to the method, region division and task balanced distribution are completed by improving a DARP algorithm, path coverage efficiency and smoothness are optimized by fusing a BA * algorithm, a cattle farming method and a TEB algorithm, multi-type path conflicts are solved by adopting a time window reservation and distributed negotiation strategy, and dynamic adjustment under new task access, robot faults and environment sudden change is supported. According to the method, the load balance, the path optimization effect and the conflict avoidance capability of multi-robot collaborative distribution can be remarkably improved, the efficient and stable execution of the distribution task is ensured, and the method is suitable for terminal distribution scenes such as communities and office buildings.

The invention discloses a resource dynamic allocation method and system for uplink communication of a multi-beam low-orbit satellite, and the method comprises the steps: a user management unit senses that a user accesses a satellite communication network, and transmits a resource request to an uplink resource allocation unit; after receiving the resource request, the uplink resource allocation unit allocates uplink wave beams, time slots and channel resources for the user and respectively sends a resource allocation result to the user management unit and the on-satellite transmission unit, and the user management unit sends the received resource allocation result to the user; and the on-satellite transmission unit schedules a specified beam to receive uplink data of the user in a specified channel and a specified time slot according to a resource allocation result. The method has a high resource utilization rate, and supports multi-user on-demand dynamic scheduling of satellite uplink beam resources.

The invention discloses a big data storage and backup management system based on cloud service, relates to the field of big data backup management, and is used for solving the problems that parameters collected in a parameter library cannot be dynamically adjusted to balance the stability of the system, and the cloud storage environment adaptability of long-term operation of the system is reduced. A weighted analysis model is established by collecting a deviation value between an estimated backup time interval and a reference value, a deviation value between an estimated backup data volume and a reference data volume and an overall round duration, the total number of times of backup interval adjustment is obtained, parameters influencing the backup interval adjustment range are collected and integrated into a parameter library, and the backup interval adjustment range is adjusted according to the parameter library. Setting a data cooling and heating mechanism according to the network delay effect corresponding to all the parameters in the parameter library and the data change frequency, classifying the parameters in the parameter library, formulating a group of fuzzy rules for fuzzy reasoning according to the frequency characteristics of the cold data parameters in the backup interval, determining the calling result of the cold data parameters, and optimizing the utilization rate of storage resources.

The present invention relates to the field of navigation technology, and in particular to a method and system for warning a ship overtaking and collision in a dangerous area. The method comprises: collecting real-time status data of a ship, generating a dynamic risk field by multi-scale analysis, predicting trajectories by using a graph neural network in a non-Euclidean space, optimizing path data in combination with a multi-level Bayesian method, and using reinforcement learning for self-optimization of risk response. The present invention can significantly improve the collision avoidance capability and navigation safety of a ship in a complex marine environment through dynamic risk assessment and adaptive path optimization.

Embodiments of the present application provide a kind of intelligent interview method, device, equipment, storage medium and program product, it is related to artificial intelligence field, specifically related to the application of big model in financial technology field.The method comprises: the interview purpose of target project and the item to be confirmed of interviewee are input into pre-trained outline generation model, and interview outline is generated;According to the sequence of interview question, ask the target interviewee, after asking current interview question, obtain reply information;If the first correlation of reply information and current interview question is greater than or equal to the first threshold value, then the second correlation of reply information and preset answer is calculated, and the user emotion in reply information is extracted;If there is target preset answer with second correlation less than second threshold value or user emotion is preset emotion, then ask question is sent to target interviewee.The present application realizes a kind of interview method based on big model and can dynamically adjust automatic execution, make interview more intelligent, flexible and effective.

The invention discloses an agent use method and device, equipment and a storage medium. The method comprises the steps of obtaining multiple pieces of historical operation data corresponding to a preset game environment and corresponding historical gold coin information; obtaining historical operations corresponding to the game data, wherein the historical operations comprise any one of an equipment purchasing operation, an equipment selling operation, a skill upgrading operation and other operations; inputting the plurality of pieces of historical operation data and the historical gold coin information corresponding to each piece of historical operation data into a to-be-trained operation prediction model, and completing the training of the operation prediction model according to the output prediction operation and the corresponding historical operation; obtaining current equipment information, current skill information and current gold coin information corresponding to the intelligent agent to be used; inputting the current equipment information, the current skill information and the current gold coin information into the prediction operation model, and outputting; the usage information includes one or more of equipment purchase operation, equipment sale operation, skill upgrade operation and other operation.

The invention discloses a blue army task planning method based on knowledge and data driving, and relates to the technical field of military simulation modeling and computer simulation, and the blue army task planning method comprises a sensing layer, a fusion layer, a reasoning layer, a planning layer, an optimization layer and an output layer. Bidirectional interaction between the layers is achieved through data interfaces, and deep fusion of knowledge and data in the whole process is ensured; by constructing a dynamically updated knowledge graph and a knowledge-data mapping mechanism, domain knowledge can guide data modeling in the whole process, meanwhile, the data can optimize a knowledge system in real time, the limitation of a single driving mode is solved, and through testing, in a complex military drill scene, the knowledge and data can be fully fused. The matching degree of the fusion-driven planning scheme and the actual situation reaches 92%, is improved by 25% compared with a knowledge-driven scheme, and is improved by 30% compared with a data-driven scheme.

The application discloses an adaptive tuning device and method of an anti-interference satellite communication antenna, and relates to the technical field of satellite communication, wherein the device comprises a multi-mode reconfigurable surface antenna module, an environment sensing module, a prediction module, a cooperative tuning module and a closed-loop feedback correction module; the application constructs a multi-mode radiation system through the multi-mode reconfigurable surface antenna module and the environment sensing module, obtains an electromagnetic environment and a link state, the prediction module predicts interference and signal deviation based on multi-source data, the cooperative tuning module synchronously executes radiation mode switching and array amplitude and phase pre-tuning according to a prediction result, preset nulls are formed before interference arrives, and the closed-loop feedback correction module collects signal quality in real time and corrects a prediction model and a tuning strategy, so that a prediction type, closed-loop self-optimizing active anti-interference tuning mechanism is formed, interference is not arrived, and tuning is performed in advance, and the anti-interference capability and communication stability under a dynamic complex electromagnetic environment are significantly improved.

This utility model relates to the field of high-voltage electrical system technology, and particularly to a DC copper busbar with single-phase conduction function, comprising: an input copper busbar, an output copper busbar, a power diode, and an insulating material; the input and output copper busbars are arranged in parallel, and the two copper busbars have an overlapping area in a direction perpendicular to the plane of the copper busbars, with a predetermined gap between the input and output copper busbars in the overlapping area; the power diode is disposed within the gap in the overlapping area of ​​the input and output copper busbars, and its two ends are electrically connected to the input and output copper busbars respectively; the insulating material fills the remaining gaps outside the power diode in the overlapping area of ​​the input and output copper busbars. This utility model solves the problems of high cost, reliance on manual operation, and complex structure of existing anti-reverse protection measures.

The invention relates to the field of intelligent data analysis, and provides an insurance abnormal behavior identification method and system based on deep learning, and the method employs a data processing technology based on deep learning to carry out the embedded coding of a data set of personal communication information, and then carries out the full-time-domain cascading. And performing significant aggregation between embedded coding and feature effects on the data set of the doctor-seeing position information, so as to guide significant interaction representation according to conditions between communication embedded cascade features and doctor-seeing position significant aggregation representation to intelligently judge whether the doctor-seeing behavior of the target user is abnormal or not. In this way, complex interaction among different features can be captured, and therefore abnormal behaviors can be recognized more accurately. And meanwhile, the system can be continuously optimized along with the evolution of fraud means, so that better dynamic adaptability is achieved to cope with continuously changing fraud behaviors, and the accuracy and the intelligent degree of insurance abnormal behavior auditing are improved.

The invention belongs to the technical field of file retrieval. According to the sensitive file retrieval method and system based on the white list database, before file retrieval and detection are started, the white list database is queried, and whether a file path is recorded in a record which passes detection of the white list database or not is judged; if the file path exists in the detected record of the white list database, directly skipping a file corresponding to the file path, and entering a detection process of a next file; otherwise, executing strategy detection; after strategy detection is passed, file paths and corresponding strategy identifiers are recorded in a white list database, and files in the white list database are directly skipped during retrieval; according to the method, after the file passes through the detection strategy, the path of the file is recorded in the white list in the database, and when the same detection strategy is executed subsequently, the system can directly skip the recorded file, so that the retrieval efficiency is remarkably improved.

The invention discloses a multi-person picture synthesis method and device, equipment and a medium. The method comprises the following steps: splicing original images at multiple angles to obtain a panoramic image of a current target area; real-time tracking is carried out according to all face position information in the current frame, sound source information is collected synchronously, and a current spokesman in the target area is determined; and generating a multi-person synthetic picture according to the real-time motion picture of the current spokesman and the real-time motion pictures of other people. Multi-angle images are collected for feature point matching and image splicing, a high-quality panoramic image is generated, wide-area visual angle coverage is achieved, a more detailed image is obtained, all face positions are locked, corresponding sound source information is collected, a current spokesman is recognized, a multi-person synthetic picture which highlights the current spokesman and gives consideration to other people is generated, and the multi-person synthetic picture is obtained. The integrity of the panoramic information is guaranteed, meanwhile, the key content is clearly presented, the problems that the visual angle is limited and information is overloaded are solved, and the communication efficiency and watching experience of a teleconference are improved.

The present application relates to the technical field of intelligent control, and discloses a power plant low bypass steam direct heating network circulating water dynamic shunting system, comprising: a data acquisition module, which is used for collecting power plant operation parameters and heating network demand parameters in real time; a model solving module, which is used for inputting the collected parameters into a preset multi-objective optimization model, and obtaining an optimal steam shunting ratio through online calculation of the model; an instruction generation module, which is used for converting the optimal steam shunting ratio into real-time control instructions for a shunting regulating valve group; an execution adjusting module, which is used for adjusting the shunting regulating valve group according to the control instructions, dynamically changing the flow of low bypass steam injected into the heating network circulating water, and discharging the remaining steam according to the original bypass system; and a safety monitoring module, which is used for independently monitoring key safety parameters, and when a safety threshold is exceeded, interrupting the main control process and executing a preset highest priority safety interlocking action. Under the premise of guaranteeing the heating quality of the heating network, the economic benefits of low-grade waste heat recovery are maximized.