5478results about How to "Reduce occupancy" patented technology

The invention discloses a cloud edge cooperative control method, which comprises the following steps: establishing a cloud edge cooperative architecture, collecting data by edge nodes, and selecting a corresponding data compression strategy to compress the data; generating an edge-cloud collaborative uploading data stream; the cooperative control layer executes adaptive load balancing based on a cloud-side node state, task characteristics and a network condition which are monitored in real time so as to realize task allocation and scheduling between the cloud and the side; the cloud receives the data stream transmitted by the edge node, decompresses and processes the compressed data according to the metadata information, completes the task and feeds back an execution result; a node performance prediction model is constructed by using system state monitoring information, a task allocation strategy and a resource allocation and data compression strategy are adjusted in real time according to a prediction result, and cloud-side collaborative optimization is realized. Efficient data transmission, flexible task scheduling and self-adaptive resource management are realized, and the performance of the system in a complex and changeable environment is comprehensively improved.

The invention relates to the technical field of big data storage, in particular to a storage control method and device based on big data. The method comprises the following steps: acquiring an original data stream and application demand information; performing data dynamic correlation analysis on the original data flow according to the application demand information to obtain a dynamic update data correlation graph; performing feature vector generation on the dynamic update data association graph to obtain a data feature vector; performing intelligent quality evaluation on the data feature vector to obtain a data quality label; performing timeliness evaluation according to the data feature vector, and performing access popularity evaluation to obtain a timeliness score and an access popularity score; and acquiring system resource load condition information. The storage resource utilization rate, the data access performance and the overall efficiency of the system are improved by solving the limitation of a traditional method in the aspects of static data management and resource allocation.

The invention relates to the technical field of raw material management systems, in particular to a raw material management system for building engineering construction. The management system comprises a raw material information management module, a purchase management module, an inventory management module, a use management module, an optimization allocation module, a raw material quality evaluation module, an environmental impact evaluation and material selection module, a supply chain transparency module, a life cycle management module and a raw material demand prediction and inventory optimization module. An automatic purchase and supplier collaborative optimization module; the system uniformly manages data information of links of raw material purchasing, inventory, use and the like through a centralized data platform, forms a complete material management chain, avoids information dispersion and lag, realizes real-time sharing of data of each link, improves transparency of management and accuracy of decision making, and meanwhile, greatly improves management efficiency. The complexity and error rate of manual operation can be remarkably reduced through automatic purchase order generation, intelligent inventory management and supplier collaboration.

The invention provides a construction site collision risk assessment method based on a digital twinborn scene, and relates to the technical field of building simulation and collision detection. According to the method, after collision risk evaluation factors are screened out by constructing a construction site digital twinborn scene, correlation analysis is carried out by using a weighted Heraonian average operator, and a determined weight is obtained; constructing a membership function of each collision risk evaluation factor, performing fuzzy reasoning, establishing a construction site collision risk evaluation mechanism, and outputting an evaluation result; dynamically generating and updating the working space of the digital twinborn scene according to the evaluation result, reconstructing the digital twinborn scene, and executing voxelization collision detection to obtain an accurate collision detection result; and according to an accurate collision detection result, in combination with the generated dynamic working space, calculating a collision time-space amount to obtain a final risk value. The method is high in collision detection speed and accurate, is applied to a dynamic complex environment of a construction site, and can achieve objective and accurate evaluation of space-time collision risks.

The invention discloses an intelligent security and protection monitoring method and system based on image recognition. The method comprises the following steps: S1, acquiring multi-modal data; s2, performing preprocessing to generate complete target information; s3, performing real-time detection on the generated complete target information, and performing primary processing on a conventional scene; s4, based on an improved visual SLAM technology, obtaining three-dimensional information of multiple cameras; s5, when the edge device detects a potential abnormal condition, performing abnormal behavior analysis on the data by using a Swin Transform model, and optimizing a detection result in combination with the three-dimensional information of the multiple cameras; s6, performing adaptive learning on historical data by using a depth deterministic strategy gradient algorithm, and dynamically adjusting a threshold value of abnormal behavior detection; and S7, after an abnormal condition is detected, processing data through differential privacy protection and an end-to-end encryption technology, automatically controlling security and protection equipment, and giving a real-time alarm. According to the invention, image recognition and deep learning technologies are fully utilized, and accurate real-time monitoring in a complex scene is realized.

The invention discloses a brain tumor MRI image segmentation model and method based on an improved Swin UNETR network, and belongs to the technical field of image segmentation. The brain tumor MRI image segmentation model is obtained by improving a Swin UNETR as a baseline model and comprises an encoder, a jump connection and a decoder, the decoder comprises a plurality of convolution fusion modules and an attention mechanism module, feature information of different levels is fused in the decoder by using the plurality of convolution fusion modules, and the attention mechanism module is used for identifying the feature information of the different levels of the feature information of the different levels of the feature information of the different levels of the feature information of the different levels of the feature information of the different levels of the convolution fusion modules. And the attention mechanism module is used to focus the feature information obtained by the plurality of convolutions. By adopting the scheme of the invention, the problem that the existing medical image segmentation technology based on deep learning is insufficient in brain tumor MRI image segmentation precision and is difficult to realize accurate segmentation especially in the face of different types of tumors and variability can be effectively solved, the calculation amount is small, and the image segmentation efficiency can be effectively improved.

The invention discloses a communication transmission control cabinet based on big data, and relates to the technical field of communication transmission, and the control cabinet comprises the steps: introducing a real-time network state perception and prediction module, collecting the data of a link through an embedded sensor, carrying out the prediction through combining with an LSTM time sequence model, carrying out the estimation of a future link state, introducing a multipath transmission mechanism, and carrying out the real-time network state perception and prediction. When tasks are distributed, bandwidth, delay and load conditions of paths are synthesized, batch uploading is performed through caching and compression strategies of edge computing nodes, occupation of links is effectively reduced, when link performance is reduced, a path switching mechanism can select alternative paths with optimal performance in real time, influences of link interruption on communication tasks are avoided, and meanwhile, the communication efficiency is improved. According to the high-real-time task priority strategy, an optimal path is adopted, delay of key tasks is reduced, non-real-time tasks are compressed and processed in batches at edge nodes, resources are not competed with real-time tasks, interference is avoided, and the requirements for ship navigation and real-time monitoring tasks are met.

The invention discloses a VR-based unmanned aerial vehicle inspection method and system, a storage medium and a program product, and the method comprises the steps: calculating a low-polygon terrain model, and carrying out the subdivision smoothing processing, and obtaining a three-dimensional terrain image; matching and registering the panoramic image sequence and the three-dimensional topographic image to obtain a point cloud map; generating a virtual mining area environment map; calculating image quality parameters; when the image quality parameter is lower than a preset threshold value, generating a semi-real and semi-virtual fusion display picture; transmitting the semi-real and semi-virtual fusion display picture to a VR helmet for display, and synchronizing the attitude information of the unmanned aerial vehicle to the VR helmet; when it is determined that a fatigue feature appears according to the brain wave data, the unmanned aerial vehicle is controlled to be switched to a preset autonomous flight mode; constructing a virtual anti-collision boundary; and obtaining an optimized flight path according to the virtual anti-collision boundary, and controlling the unmanned aerial vehicle to inspect according to the optimized flight path. The continuity and reliability of the inspection task are improved, and the safety risk of the inspection task is reduced.

The invention provides a decoding and coding method, device and equipment, and the decoding method comprises the steps: decoding a first code stream of a current image block, and obtaining a coefficient hyper-parameter feature of the current image block; for each stage sub-block of the current image block, determining a probability distribution parameter of the stage sub-block based on the coefficient hyper-parameter feature, and decoding a second code stream of the current image block based on the probability distribution parameter to obtain a residual feature of the stage sub-block; based on a mean value prediction network, determining mean value features of the sub-blocks in the stage; determining a reconstruction feature of the stage sub-block based on the mean feature of the stage sub-block and the residual feature of the stage sub-block; performing feature aggregation on the reconstruction features of the sub-blocks in each stage to obtain the reconstruction features of the current image block; a reconstructed image block of the current image block is determined based on the reconstruction feature of the current image block. Through the scheme of the invention, the coding performance and the decoding performance are improved.

The invention provides an image rendering method and device, electronic equipment and a storage medium, and the method comprises the steps: calculating a distance between a virtual image obtaining device and a to-be-rendered object through determining the to-be-rendered object in a to-be-rendered scene, and determining the distance as a first loading factor; determining a rendering picture pixel proportion of each to-be-rendered object based on preset view cone size information, determining the rendering picture pixel proportion as a second loading factor, generating a rendering loading priority sequence according to the first loading factor, and determining a model level resource file and a texture level resource file according to the second loading factor, loading the resources to a video memory for rendering according to the sequence; the priority is determined by calculating the distance between the camera and the to-be-rendered object, it is ensured that a high-quality model and texture of a close-range object are loaded preferentially, the instantaneity of the rendering loading process is improved, texture and model grade division is conducted through the pixel proportion of the to-be-rendered object in the picture, and the rendering efficiency is improved. The occupation of a video memory is reduced, and the overall rendering performance of the system is improved.

The invention provides a meteorological monitoring method based on a GRU neural network and differential privacy federated learning. The meteorological monitoring method comprises the following steps: collecting local meteorological data to generate a meteorological time sequence data sequence; selecting a proper model structure according to the heterogeneity of a client, carrying out local meteorological monitoring model training based on a GRU neural network, and carrying out parameter updating on a local model through an LPDP-FedProx federal learning algorithm; the central server performs model aggregation and updating according to a global model weighted average aggregation algorithm after acquiring model parameters of the client; and the client receives the updated global model parameters and carries out the next round of local training until an optimal monitoring model is obtained. According to the method, the personalized differential privacy algorithm and the FedProx federated learning algorithm are combined, the LPDP-FedProx federated learning algorithm is provided, a federated learning framework capable of effectively processing heterogeneous data is constructed, the privacy of local data is effectively protected, and the security of federated learning is improved.

The invention provides a WEB browsing real-time rendering method and system based on distributed rendering, and relates to the technical field of distributed rendering, and the method comprises the following steps: receiving a rendering request from a user terminal browser, obtaining page data of a target webpage according to the rendering request, preprocessing the page data, analyzing the preprocessed page data, and displaying the analyzed page data. Obtaining a DOM tree structure and a CSSOM tree structure of the page, and constructing a rendering tree adaptive to the user terminal according to the DOM tree structure, the CSSOM tree structure and the equipment parameters; dividing the rendering tree into a plurality of sub-rendering trees, transmitting sub-rendering results generated by the plurality of distributed rendering nodes to a synthesis service node, performing image compression, transmitting the compressed rendering result data to a user terminal browser, receiving and decompressing the rendering result data in the user terminal browser to obtain a rendering result bitmap, and displaying the rendering result bitmap in the synthesis service node. And displaying the information to the user for real-time interactive browsing of the user.

The invention relates to the technical field of data storage management, in particular to a data storage method based on a large model. The method comprises the following steps: performing input class cluster fragmentation optimization on a target pre-training large model to obtain storage modulus fragmentation data; constructing an intelligent fragmentation storage strategy by using the storage modulus fragmentation data; obtaining target model training data; performing intelligent fragmentation storage on the target model training data by using an intelligent fragmentation storage strategy, and performing redundancy compression optimization to obtain dynamic redundancy storage index data; and training data storage space optimization is carried out according to the dynamic redundancy storage index data, and an optimized data storage strategy is obtained. According to the method, intelligent fragmentation storage and refined data redundancy compression are performed on the data, so that efficient and stable model training data access is realized, the problem of low data storage and access efficiency in a large model training process is solved, and the model training efficiency and reasoning real-time performance are remarkably improved.

The invention discloses a data storage system of a digital country based on cloud computing, which comprises a client, a data transmission module and a cloud computing center, relates to the technical field of data storage, stores hot data on a high-performance hot data storage layer, ensures that frequently accessed data can be quickly responded, improves the system performance, and at the same time, stores the hot data on a high-performance hot data storage layer. The cold data is stored in a low-performance cold data storage layer, the storage cost is reduced, effective balance of performance and cost is achieved, the data is automatically migrated among different storage layers along with change of data access frequency, so that the performance and the cost are balanced, the frequently accessed data is migrated into a cache, the access speed of the data is increased, and the data access efficiency is improved. And establishing a relationship between a time interval and a data type of historical data, predicting data which are possibly accessed in a given time interval, adding the data into the cache, dynamically adjusting the cache, and ensuring the validity of the cache.

The invention relates to the technical field of education informatization, and discloses a digital teaching resource library intelligent management and application method and system. The method comprises the steps of constructing a digital resource library, constructing a resource application big data analysis model to calculate a resource quality index, a popularity index and an update index, performing intelligent resource sorting and elimination based on the indexes, analyzing the correlation between a user retrieval request and resources, and performing intelligent recommendation. According to the method, efficient management and accurate application of teaching resources are realized, the resource utilization rate and the teaching effect are improved, the storage space and the management cost are saved, and data support is provided for education decision making.

The invention provides a vehicle mobile charging control method, electronic equipment and a storage medium, and belongs to the technical field of new energy vehicle charging. The method comprises the following steps: receiving a mobile charging request, and determining a target parking space coordinate associated with the mobile charging request; determining at least one callable robot based on the first state information of each robot in the mobile robot set; determining at least one callable charging pile based on the second state information of each charging pile in the mobile charging pile set; determining a target robot and a target charging pile according to the first position information of each callable robot, the second position information of each callable charging pile and the target parking space coordinate; and the target robot is called to carry the target charging pile to the target parking space coordinate, so that a vehicle on the target parking space is charged through the target charging pile. The problems of low charging pile utilization rate and long user waiting time in the prior art can be solved.

The invention discloses a water environment management system based on cloud computing, and relates to the technical field of environment monitoring. In order to solve the problems that in the prior art, water environment emergencies cannot be responded in real time, the requirement for rapid early warning cannot be met, regular manual data monitoring is limited by manpower, material resources and environmental conditions, high-frequency and large-range coverage cannot be achieved, and therefore the continuity and integrity of monitoring data are affected. Monitoring data are collected and transmitted to the cloud computing platform in real time, continuous and high-frequency monitoring of the water environment is achieved, the water quality change can be rapidly recognized through the instant analysis capacity of the cloud computing platform, abnormal key monitoring points are accurately recognized, the data collection and transmission time interval is dynamically adjusted according to the abnormity severity degree and the influence range, and the monitoring accuracy is improved. Therefore, key information can be acquired in time, unnecessary data transmission is avoided, resources are saved, abnormal conditions are tracked and managed, and problems can be effectively solved.

The invention discloses a DBA management method and device in an FTTR scene and an optical line terminal. The method is applied to a PON MAC chip, wherein the PON MAC chip comprises a CPU unit and a logic unit. The method comprises the following steps that: a CPU (Central Processing Unit) determines the uplink service bandwidth of each AllocID in each sub-period in the current DBA period; the CPU unit takes at least part of the sub-periods as sub-distribution periods, and determines BWMap data of the first sub-distribution period and the second sub-distribution period; the CPU unit sends the length of the silent window, the BWMap data of the first sub-allocation period, the BWMap data of the second sub-allocation period and the total length of the current DBA period to the logic unit; the logic unit determines the number of the remaining sub-allocation periods in the current DBA, and takes the BWMap data of the second sub-allocation period as the BWMap data of each remaining sub-allocation period to obtain complete BWMap data in the current DBA period; and the logic unit sends the BWMap data of the current DBA period, so that the problem that the DBA cannot adapt to the use requirement of the FTTR scene is solved, and the low-delay requirement of the DBA is met.

The invention belongs to the technical field of distributed data access, discloses an RDMA (Remote Direct Memory Access) distributed data transmission method supporting high-concurrency message transmission, and aims to improve system performance in a high-concurrency message transmission environment. The method comprises the following core steps: 1) peer-to-peer server design: each device is an independent server, and connection is established by a connection channel monitored at any time; 2) establishing a connection resource pool in advance; 3) creating and registering an RPC message; 4) transmitting the RDMA to a memory mailbox pool of a target server; 5) the client and the server perform high-concurrency processing; 6) dynamically polling the server; (7) zero-copy transmission of data is achieved through the RDMA technology. According to the invention, the dynamic expansion capability of the architecture is improved by adopting a connection resource decoupling mode, QP jitter can be reduced and the stability of the system can be enhanced by registering a memory of a continuous large page at one time, and the message transmission efficiency under the distributed network architecture is effectively improved by adopting a combined sending strategy.

The invention discloses a training method and device of a distributed computing system, a program product and a medium, which are used for solving the problems of performance difference and low communication efficiency of heterogeneous computing nodes. According to the scheme, the calculation capability of each calculation node is fully considered, so that dynamic allocation and iterative optimization of tasks are realized; specifically, the problems of idle resources and low efficiency are avoided by acquiring the computing power of each computing node and allocating the training task matched with the computing power to the computing node; the time length of single iteration is determined, and the number of iterations in the communication interval is adjusted according to the time length, so that each computing node can perform data exchange at a proper time, network bandwidth occupation and delay caused by frequent communication are reduced, the overall efficiency of the training process is optimized, mutual waiting between the computing nodes is avoided, and the network performance is improved. When the computing node runs the training process, the power consumption and heat generation are reduced, and the communication overhead is also reduced, so that the long-term stability and reliability of the computing node are improved.

The invention provides a dense warehouse adopting embedded elevators and a storage solution of the dense warehouse, the elevators are embedded between storage bins of the dense warehouse, and guide rails with single paths are arranged on the elevators. According to the elevator, the guide rails can directly communicate with the rails of the storage units of all the layers in the dense warehouse, so that cargoes at the inlet position are directly guided into the elevator through the rails during warehousing, and the cargoes are transferred to the corresponding storage layers to be stored; and during warehouse-out, the goods in the storage layer are directly guided into the elevator through the track and are transferred to the corresponding exit position. On the basis of the embedded elevator, the occupied storage space can be effectively reduced, and the storage utilization rate is increased; and front and rear buffer storage positions are formed by front and rear tracks of the lifting platform, so that on one hand, the planning design of a warehouse-in and warehouse-out line can be simplified, on the other hand, the warehouse-in and warehouse-out efficiency of automatic warehouse space can be improved through linkage scheduling of different walking devices, and the construction cost and the construction period are reduced.

The invention discloses a mine digital asset storage management method, which comprises the following steps of: extracting characteristics of mine data, dividing the mine data into hot data, warm data and cold data according to the characteristics, selecting different storage devices for the data, and dynamically adjusting migration of the data among the different storage devices based on access frequency and business requirements; future data storage requirements are predicted through a machine learning algorithm, and storage resources are scheduled; performing full life cycle management on the data, performing business value evaluation on the data, and adjusting storage priority and resource allocation based on the data value; carrying out full-life-cycle encryption on the data, and recording data operation through a block chain technology; energy consumption of the storage equipment is monitored in real time, and storage energy efficiency and carbon footprint management are adjusted. The problems of low storage efficiency, high storage cost and insufficient energy consumption optimization caused by data type diversity, access frequency difference and insufficient storage resource utilization in mine data management can be solved.

The invention discloses a 3D point cloud target identification method based on a bidirectional long-short-term memory network, which comprises the following steps: firstly, preprocessing original point cloud data, carrying out feature embedding on the preprocessed point cloud data, then carrying out data feature extraction based on the bidirectional long-short-term memory network, carrying out local feature aggregation and global feature aggregation on the data, and finally carrying out target identification on the 3D point cloud target. Feature extraction is enhanced, and the data expression ability is improved; and constructing a point cloud data identification model based on a full connection layer, training and evaluating the model, and performing point cloud data identification based on the model. According to the scheme, under the condition that the calculation precision is not reduced, the long-term dependency relationship between the 3D point cloud sequences is captured with few calculation resources and memory resources, aggregation is carried out based on the extracted features, and accurate recognition is achieved.

The invention discloses a protocol calculation method and device for aggregate communication, a calculation card and a storage medium. In the first data transmission period, the communication unit receives first protocol sub-data sent by the first node; the protocol information processing unit determines a protocol type, a data address and first data sent by the first node according to the first protocol sub-data; the protocol calculation unit reads local data corresponding to the data address from the storage unit according to the data address; performing protocol calculation on the first data and the local data according to the protocol type to obtain second data; in a second data transmission period, the protocol information processing unit reads second data in the storage unit and generates second protocol sub-data according to the second data, and the communication unit sends the second protocol sub-data to a second node; and after a plurality of data transmission cycles, the data corresponding to the data address completes protocol calculation of all nodes. And the balance of protocol calculation and chip self-calculation tasks in the aspects of calculation power and storage is realized, and the overall performance is improved.

The invention relates to a gob-side entry retaining grouting reinforcement technology in the field of mining engineering, provides a method for grouting reinforcement of a reuse roof of a gob-side entry retaining of a coal mine and accurate control of the grouting amount, and aims to realize real-time monitoring and control of the grouting amount in the grouting process. According to the method, pressure monitoring, flow velocity monitoring, low-frequency electromagnetic wave detection, ultrasonic detection and on-site drilling slurry discharge monitoring are combined, the grouting process is detected in real time through multiple monitoring means, and it is ensured that cement slurry fills cracks and does not flow into a goaf. And meanwhile, cement paste with different particle sizes is adopted for grouting step by step according to the size of the crack, and paste leakage is avoided while the filling effect is guaranteed. The method is suitable for grouting operation under complex geological conditions, has the characteristics of high efficiency and accuracy, and remarkably improves the stability and safety in the use process of a roadway.

The invention discloses a diagnostic data remote transmission method based on a regional remote diagnosis and treatment platform. The method comprises the following steps: S1, collecting diagnostic data of a patient; s2, preprocessing the acquired diagnosis data; s3, carrying out data transmission by utilizing a short-distance wireless communication technology and a long-distance mobile communication technology; s4, performing data encryption and identity verification by adopting AES and RSA encryption algorithms; s5, after the regional remote diagnosis and treatment platform receives the diagnosis data, decryption and identity verification are carried out on the data; s6, performing image data analysis by using a convolutional neural network, processing character data by using a natural language processing technology, predicting and classifying physiological parameters by using regression analysis and a classification algorithm, and performing fusion analysis on multi-modal data in combination with a deep learning model; and S7, transmitting the analyzed and processed diagnosis result back to a local data terminal through a long-distance mobile communication technology. The system has the advantages of high transmission stability, high safety, high transmission efficiency and high diagnosis accuracy.

The invention provides an integrated containment pressure suppression, filtration and discharge system. The integrated containment pressure suppression, filtration and discharge system comprises a containment, a discharge valve, a discharge pipeline, a spray pipe, a first-stage pressure suppression water tank, a second-stage pressure suppression water tank, a baffle plate steam-water separator, an expansion tank, a rupture disk structure, a combined metal fiber filter layer, a radioactivity monitor, a first stop valve and a second stop valve. The two-stage pressure suppression water tank is arranged, the high removal efficiency of airborne radioactive substances is achieved through two-stage bubbling water washing, then the combined metal fiber filter is arranged by means of an existing cabin, and the requirements for rapid pressure reduction and radioactive substance removal of the containment under the LOCA accident working condition can be met; meanwhile, the problem that the containment filtering and discharging system of the existing nuclear power station occupies too many resources is solved, and the economical efficiency is high.

The utility model provides a hip structure and a robot, and relates to the technical field of robots, in particular to the technical field of hip structures of robots. A hip structure includes: a first hip actuator and a second hip actuator for performing hip movement of a robot; wherein the first hip joint actuator and the second hip joint actuator are obliquely arranged. The size of the hip structure is reduced, the space compactness of the hip of the robot is improved, the flexibility of the hip is improved, and the movement range of the hip is expanded.

The invention provides an automatic sprinkling irrigation system which comprises a water storage device, the water storage device comprises a water storage piece and a rainwater collection piece, the rainwater collection piece is arranged at the upper end of the water storage piece, the rainwater collection piece is provided with a collection part, and the collection part is communicated with the water storage piece through a flow guide channel; the power supply device is arranged on the water storage device and comprises a photovoltaic panel, the sprinkling irrigation device comprises a water pipe and a nozzle, one end of the water pipe communicates with the water storage device, and the other end of the water pipe communicates with the nozzle; the control device comprises a sensor and a control unit. According to the technical scheme, the underground water storage part and the ground rainwater collection part are combined, rainwater is efficiently collected and stored, and evaporation loss and ground space occupation are reduced. The sprinkling irrigation device is connected with the water storage piece through a water pipe, sprayers can be flexibly arranged according to requirements, and uniform irrigation is achieved. The control device integrates a sensor and a control unit, monitors soil humidity, illumination and other environmental parameters in real time, automatically adjusts an irrigation strategy, ensures timely and appropriate irrigation, and solves the problem of water resource waste.

The invention discloses a water quality prediction method based on multi-source data, and constructs a water quality prediction model based on a long short-term memory network (LSTM) and a graph neural network (GNN) of multi-source data (surface water monitoring data, meteorological data, hydrological data and pollution source on-line monitoring data), so that short-term real-time on-line prediction of surface water quality factors can be realized. According to the method, on the basis of multi-source data such as section water quality monitoring data, historical meteorological data, forecast meteorological data, hydrological monitoring data, pollution source online monitoring data and upstream and downstream relations, multiple main water quality influence factors such as water quality, meteorology, hydrology, pollution sources and upstream and downstream section water quality are comprehensively considered; the LSTM or the graph neural network is adopted to establish a water quality prediction model, and the comprehensive influence of various influence factors and the time-space relationship on the water quality is simulated, so that the simulation result is more accurate and reliable.