55 results about "Initial load" patented technology

The invention discloses a power efficiency optimization method based on intelligent load management. The method comprises the following steps: S1, generating an initial load characteristic matrix; s2, decomposing the initial load characteristic matrix into a short-term behavior characteristic matrix and a long-term trend characteristic matrix through pseudo-symmetry decoupling; s3, multi-level load demand prediction is carried out through a fusion structure of the multi-scale convolution and the recurrent neural network; s4, coding a cognitive gene sequence, triggering gene mutation and recombination operation when significant change occurs, and generating a load distribution strategy; s5, a load distribution strategy is transmitted into a causal reasoning and self-supervision feedback control module, and if the deviation between feedback data and a prediction result exceeds a threshold value, a self-learning mechanism is triggered; s6, dynamically distributing the load priority, and generating a decision path of load management; and S7, performing adaptive optimization through a multi-layer feature feedback closed-loop optimization system. The intelligent load management method is adopted, and the power efficiency is optimized through multi-layer feature decoupling and closed loop.

The invention discloses an electric power marketing management information platform daily electric power fitting method and related equipment, and relates to the technical field of electric power data management, and the method comprises the steps: obtaining historical load data, external environment data and equipment operation state data of a target region; constructing an initial load fitting model based on the historical load data; extracting a date characteristic factor according to a preset time classification rule, and dynamically correcting the initial load fitting model based on the date characteristic factor to generate a corrected load model; generating a multi-source fusion feature based on the external environment data and the equipment operation state data; and inputting the multi-source fusion features into the corrected load model, and outputting a target load prediction result.

The invention relates to a method and a device for reducing response time delay of a deep learning model and a storage medium, which are applied to the technical field of artificial intelligence, and comprise the following steps: acquiring a historical load through sliding window sampling, acquiring initial load prediction data by using a dynamic joint prediction mechanism based on the historical load, obtaining an error compensation value through the corresponding actual load data sequence, and performing error compensation on the initial load prediction data to obtain final load prediction data; determining the total number of instances required for a period of time in the future based on the final load prediction data sequence; the number of instances which are currently operated is adjusted to match the total number of instances required in a period of time in the future, so that the response time delay of large-scale deep learning model reasoning work is reduced; through accurate workload prediction and resource scheduling, the response time delay of model reasoning can be effectively reduced, the achievement rate of a service level target is improved, and the service cost is reduced.

The invention relates to the technical field of dynamic load balancing processing and communication, and discloses a dynamic load balancing transmission method based on synchronous adjustment, which comprises the following steps: acquiring a synchronous parameter set of each communication node in real time; calculating an initial load weight of each communication node through a dynamic feedback algorithm; a priority coefficient is introduced, and the initial load weight is dynamically corrected according to the real-time level of the transmission request; a final load weight is generated through normalization processing, the sum of the load weights of all the nodes is made to be 1, and a dynamic load distribution strategy is generated according to the final load weight; and dynamically distributing the transmission request of the terminal equipment to each communication node according to the load weight proportion of each node by adopting a weighted polling algorithm according to a dynamic load distribution strategy, thereby completing dynamic load balancing of the transmission request. According to the method, the synchronous parameter set of each communication node is acquired in real time, and the load weight of the node is accurately calculated and dynamically corrected, so that an efficient and accurate dynamic load balanced distribution strategy is realized.

The invention provides a bill component design period and operation period micro-front-end extension method and device. The method comprises the following steps: developing a micro-front-end extension module according to a preset extension development specification; the micro-front-end extension modules are separated and packaged into different resource files according to different use scenes; registering the micro-front-end extension module as an operation period component or a design period component, and recording an access path and metadata of a corresponding resource file; when a receipt application is started, analyzing the configuration information of the receipt to obtain a core function and an extension point definition, and querying and identifying an extension list needing to be loaded by using registered extension description information and an access path of a resource file; and according to the extension list and the corresponding resource file path, dynamically loading the registered resource file of the running period component or the design period component as required. Through on-demand loading and dynamic module loading, a user only needs to load necessary extensions when opening a receipt, and the initial loading time is remarkably shortened.

The invention relates to a self-adaptive wind load passive slowing-down design method based on threshold triggering. The method comprises the steps that aerodynamic parameters during non-load-shedding are calculated according to a nominal program angle; calculating an initial load shedding program angle ideal value; when the flight Mach number is within the allowable load shedding range and the qalpha value exceeds the current amplitude limiting value, load shedding is carried out; only the part exceeding the qalpha amplitude limiting value is subjected to load shedding; and carrying out amplitude limiting processing on the angle correction of the load shedding program in combination with the dynamic characteristics and maneuverability of the rocket body. According to the method, the load shedding opportunity is triggered through the qalpha amplitude limiting value, amplitude limiting processing is carried out on the load shedding program angle and the angular velocity correction amount, the structural strength of a rocket body can be fully utilized, the dynamic lag characteristic and the actual control capacity of the rocket are considered, the rocket body can track the corrected program angle with high precision, and the method has high application value in engineering.

The invention relates to the technical field of data processing, and discloses an assembly line design method based on a small enterprise credit factory mode, which comprises the following steps of: initializing task and resource parameters, collecting a credit approval task set and an approval role set, defining processing time, priority weight and latest completion time for each task, and completing the task according to the processing time, the priority weight and the latest completion time. Defining a resource capability and an initial load for each approval role; and constructing a multi-objective optimization model, comprehensively considering task completion time, a resource utilization rate and task priority completion quality, and converting an optimization objective combination weight into a comprehensive objective function. According to the invention, a technical scheme based on customer admission management and detailed approval paths is adopted, and precision of risk assessment and hierarchical management of loan customers is realized. The approval nodes can be dynamically matched according to the risk levels of the customers, and the problem of resource waste caused by homogenization processing of high-risk customers and low-risk customers in an existing process is solved.

The invention discloses a bus load prediction method and device, a storage medium and electronic equipment. Relates to the technical field of data processing, the technical field of electronics and other related technical fields, and comprises the following steps: aggregating a plurality of buses in a target power grid to obtain a virtual bus corresponding to the target power grid; obtaining a plurality of first historical bus loads of the virtual bus in a historical time period; predicting the load of the virtual bus according to the plurality of first historical bus loads to obtain an initial load prediction result; and determining a target load prediction result corresponding to each bus in the target power grid according to the target matching factor and the initial load prediction result. Through the bus load prediction method and device, the problem that the prediction efficiency of the bus load is relatively low due to the fact that the load of each bus in a power grid is predicted manually in the prior art is solved.

The invention relates to a power equipment load scheduling optimization method and device. The method comprises the following steps: carrying out load operation data acquisition through monitoring units deployed in power equipment to obtain a distributed load operation data set, and obtaining strategy game relationship network data between the equipment based on the distributed load operation data set; obtaining initial load adjustment intention intensity data of each power device according to the policy game relationship network data between the devices; sending the initial load adjustment intention intensity data and the global collaborative scheduling optimization target to a distributed negotiation chain to obtain final load adjustment strategy data after negotiation consensus; and obtaining an optimized distributed load scheduling model according to the final load adjustment strategy data after negotiation of the consensus, distributing the optimized distributed load scheduling model to each power device, and executing decentralized power device load scheduling optimization. By adopting the method, the load scheduling of the power equipment can be optimized.

The invention discloses a power distribution load prediction method and system, and relates to the technical field of graph calculation, and the method comprises the following steps: obtaining load data and auxiliary features of a user, and constructing a load portrait; according to the load portrait, selecting a model subset from a pre-training model library, migrating the selected model, and outputting an initial load prediction result; constructing a load association graph according to the spatial adjacency and behavior similarity between the users, and spreading load characteristics in the load association graph; regulating and controlling the model of the target user according to the propagated load characteristics in combination with an initial load prediction result; obtaining a user load data distribution change after regulation and control, and identifying a behavior mutation; through multi-model integration, individual fine tuning, load association graph construction, incremental learning and drift detection, the problems of low precision, poor dynamic adaptability and slow response to sudden change in traditional power distribution load prediction are effectively solved.

The application discloses an AI acceleration card reasoning computing power benchmark test method and device, obtains reasoning load data and hardware configuration parameters, generates a load anchor point sequence through business scene feature joint labeling, constructs an initial load test set accordingly, quantifies concurrent throughput backoff features to generate a computing power benchmark atlas, directionally decomposes cumulative accuracy deviation in the test set, identifies two components of computing power sensitivity and bandwidth sensitivity, respectively obtains computing power correction and bandwidth bottleneck anchor points through utilization rate false reporting traceability and memory conflict frequency domain association, generates accuracy correction instructions and implements accuracy re-correction on the initial test set, performs cold start path loss detection and accuracy gradient monotonicity failure identification based on the corrected test set, locates false reporting consistency blind area anchor points, executes global optimization on the computing power benchmark atlas, and finally outputs benchmark test results, thereby improving the accuracy and business adaptability of AI acceleration card reasoning computing power benchmark test results.

The invention relates to a multi-center load balancing method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring performance indexes of a plurality of data centers; setting a fitness function based on the performance indexes, and randomly generating a plurality of initial load schemes; based on the fitness function, optimizing the plurality of initial load schemes in an iteration mode to obtain an intermediate load scheme; the intermediate load scheme is converted into an initial pheromone, ant colonies are placed in a plurality of data centers in an iteration mode, action paths of the ant colonies are obtained, the initial pheromone is iteratively optimized according to the action paths, optimized target pheromone is obtained, and the optimized target pheromone is converted into a target load scheme; and allocating tasks to the plurality of data centers based on the target load scheme. By adopting the method, multiple data centers can efficiently process large-scale requests.

The invention discloses a load prediction method and system based on user order demands, and relates to the technical field of power load prediction. Calculating a standard production load of a single product based on the historical production log and the historical power load data, and determining a characteristic operation power value corresponding to each product type according to the standard production load; generating a baseline load prediction curve based on planned starting and ending time and a characteristic operation power value in the production plan data; combining the baseline load prediction curve and the historical power load data to obtain an input feature, and inputting the input feature into the time sequence prediction model to obtain an initial load prediction curve; determining an efficiency correction factor and an environment correction factor based on the historical production log and the historical power load data; and correcting the initial load prediction curve according to the efficiency correction factor and the environment correction factor to obtain a final load prediction curve. By implementing the technical scheme provided by the invention, the accuracy of power load prediction of the industrial enterprise is improved.

The present invention relates to the technical field of dynamic load balancing processing and communication, and discloses a dynamic load balancing transmission method based on synchronous adjustment, comprising: obtaining a synchronization parameter set of each communication node in real time; calculating the initial load weight of each communication node through a dynamic feedback algorithm; introducing a priority coefficient, and dynamically correcting the initial load weight according to the real-time level of the transmission request; generating a final load weight through normalization processing so that the sum of the load weights of all nodes is 1, and generating a dynamic load distribution strategy based on the final load weight; according to the dynamic load distribution strategy, using a weighted polling algorithm, dynamically distributes the transmission request of a terminal device to each communication node according to the load weight ratio of each node, thereby completing dynamic load balancing of the transmission request. The present invention obtains the synchronization parameter set of each communication node in real time, accurately calculates and dynamically corrects the node load weight, thereby realizing an efficient and accurate dynamic load balancing distribution strategy.

The invention discloses a flexible load regulation and control method and system based on global optimization and edge collaboration, and belongs to the field of power systems, and the method comprises the steps: predicting a power grid flexible load prediction value corresponding to a target regulation and control day according to the flexible load data of a current power grid; iteratively optimizing the initial load regulation and control solution according to the power grid flexible load predicted value and a preset random swarm intelligent optimization algorithm to obtain an optimal load regulation and control solution; according to the adjustable power range corresponding to the flexible load of each edge device, decomposing the optimal load regulation and control solution to obtain a first load regulation and control solution corresponding to each edge device; according to the difference value between the actual value and the predicted value of the flexible load of each edge device in the target regulation and control day and a preset distribution rule, adjusting each first load regulation and control solution to obtain the compensation load capacity of each edge device; and according to each compensation load capacity, each edge device is controlled to execute flexible load regulation and control. According to the invention, the problem of poor flexible load regulation and control effect in the prior art can be solved.

The invention relates to an operation control method and device of a fused salt heat storage system, computer equipment and a readable storage medium. The method comprises the following steps: acquiring historical operation basic data of the fused salt heat storage system; iteratively training an initial load prediction model of the fused salt heat storage system according to the historical operation basic training data to obtain a target load prediction model; after the target load prediction model is tested to reach the standard based on the historical operation basic test data, inputting the real-time operation basic data into the target load prediction model, and performing prediction through the target load prediction model to obtain an operation load prediction value of each discrete prediction time interval within the load prediction time domain length; generating a corresponding operation control rule for each discrete prediction time interval according to all the operation load prediction values; and performing operation control on the fused salt heat storage system within the load prediction time domain length according to all the operation control rules. By means of the method, the control effect of operation control over the fused salt heat storage system is improved.

The invention discloses a big data network communication coordination method, and relates to the technical field of network communication, and the method comprises the specific steps of model construction, request receiving and load monitoring, load shunting and complexity evaluation, priority calculation and dynamic sorting, and request processing and threshold control. According to the method, the hardware performance index values of the nodes are collected regularly, the comprehensive load value of the current node is calculated and compared with the initial load threshold value for judgment, when the comprehensive load value exceeds the initial load threshold value, the load state of the adjacent nodes is inquired, and candidate nodes lower than the initial load threshold value are screened out; reasonable request distribution is achieved by calculating the distribution proportion through the optimal distribution coefficient formula, the design can dynamically adjust request distribution according to the real-time load condition, the problems of node overload and resource idleness are effectively avoided, the processing efficiency and stability are remarkably improved, and under the high-concurrency scene, the service life of the system is prolonged. And a request distribution strategy can be quickly responded and adjusted.

The invention discloses a dynamic optimization method and device for space networking, equipment, a medium and a program product, and relates to the technical field of space networking, and the method comprises the steps: generating a first service range of each node according to an initial feature of each node, the initial feature comprising an initial load and an initial communication demand; according to the real-time characteristics of each node, the first service range of each node is dynamically adjusted to obtain a second service range of each node after dynamic adjustment, and the real-time characteristics comprise a real-time load and a real-time communication demand; and combining the second service ranges of the nodes to obtain third service ranges of the nodes, and providing network services for network devices in the third service ranges of the nodes through the nodes. According to the invention, the adaptability of a traditional space networking scheme is improved.

The present invention relates to the field of wind power generation technology, and provides a blade calibration method and device, wherein the method comprises: obtaining an initial load curve; performing positive and negative segmentation processing on the load in the initial load curve, dividing the first load curve into multiple second load curves with the target extreme point in the first load curve obtained by the segmentation processing as the segmentation point; adjusting each second load curve, and integrating the adjusted multiple second load curves into a third load curve; and performing strength calibration on the blade in the target direction based on the third load curve. To solve the problem of relatively low efficiency in strength calibration of blades, each second load curve is adjusted to achieve local optimization, and fast and accurate fitting of each second load curve is achieved. The adjusted multiple second load curves are integrated into the third load curve, which can achieve global optimization, improve the fitting accuracy and fitting speed as a whole, and improve calibration efficiency.

The present invention discloses a packet error rate optimization method, system, device, and medium based on 5GNR packet coding and load balancing. The method includes: dividing a source data packet into code blocks to obtain a plurality of subcode blocks, adding an outer code to each subcode block, adding error correction coding to obtain a plurality of error correction coding blocks, packet encoding the corresponding bits of all error correction coding blocks to obtain a check packet; calculating an initial load balancing value; using a load balancing distribution technique to average the number of bits to be removed to each error correction coding block and the check packet, predicting the rise and fall of the load balancing trend between the error correction coding blocks based on the initial load balancing value, so that when the number of bits removed from the error correction coding block exceeds the initial load balancing value, the bits are forwarded to other error correction coding blocks, and the bits to be removed are preferentially distributed to the error correction coding blocks with bit-associated loads below a set value, and the BCJR algorithm is used to parallelly process the error correction coding blocks after the bits are removed for decoding. The present invention ensures consistent performance among all error correction coding blocks and reduces the packet error rate of the data packet.

The invention discloses a load collaborative management method, system and device based on multi-dimensional perception, and the method comprises the steps: obtaining multi-dimensional data which can affect a load through a deployed edge node; extracting features of the multi-dimensional data in different dimensions to form an initial load feature vector; inputting the initial load feature vector into a pre-trained model to obtain a load feature vector output by the model; clustering the load nodes based on the load feature vectors, and correspondingly forming group-level units by clustering groups obtained by clustering; and selecting a representative node of each group-level unit, receiving a scheduling instruction through the representative nodes, and coordinating the allocation of resources in the group-level units. Due to the fact that the multi-dimensional data including the environment, the user behavior, the equipment state and the like influencing the load are collected, the high-quality data are extracted from the multi-dimensional data, the load feature vector is formed and used for determining the load management decision, and the accuracy of the load management decision is improved.

The invention relates to the technical field of distributed control, in particular to a multi-target balanced load coordination control method for a distributed virtual power plant, and the method comprises the steps: analyzing the similarity between historical daily load data, introducing an LOF algorithm to evaluate the abnormal degree of the similarity of historical curves, and determining the abnormal degree of the similarity of the historical curves; determining the reference of a predicted value obtained through historical data at the same time in previous days; determining an initial load prediction value at the next moment according to the data change trend before each moment, predicting a first prediction value at each moment by adopting a prediction algorithm according to the load data at the same moment in the previous day, and evaluating the predictability of the data at each moment by analyzing the regularity of the data at each time point in the previous day; through multi-source information complementation, a final load demand is determined, and then load value regulation and control of each power supply load of the virtual power plant are performed, so that the accuracy of virtual power plant load prediction is significantly improved, the precision and robustness of load prediction are improved, and subsequent precise regulation and control are supported.

The present invention discloses a method for loading game resource files based on access order sorting, which relates to the technical field of game engine resource scheduling and is used to solve the problems of unreasonable resource loading order and increased loading delay. The method constructs a dependency graph by parsing resource metadata, uses nodes to represent resource files and directed edges to represent dependency relationships, performs loop detection and topological sorting to obtain structural parameters, combines topological structure with access frequency, uses Pareto sorting to generate an initial loading sequence, monitors the decompression ratio and call frequency of loaded resources in real time, calculates activity, collects the number of requests for unloaded resources and the depth of the dependency chain, and estimates demand intensity. An adjustment factor is constructed based on the Pareto level, activity, and demand intensity, and the loading sequence is reordered and tasks are reallocated. Dependency consistency is verified, key resources are loaded first, and dynamic scheduling optimization is achieved.

The invention discloses a multi-user heterogeneous load identification method and system. The method comprises the following steps: acquiring first data of target electric equipment, and establishing an initial load sample library according to the first data; pre-training a plurality of first models and a plurality of second models, and performing first cooperative training on all the first models and the second models to obtain a third model; and performing real-time load identification according to the third model in combination with the initial load sample library. The cooperative training method does not need to limit the completely consistent structure of each user model, and can fully cooperate with the multi-user heterogeneous model to improve the trained effect; the effect of knowledge distillation for cooperative training of the multi-user load identification model is effectively improved; the designed improved convolution module can realize high-accuracy load identification under the condition of extremely few parameters and calculation amount, the lightweight degree of a compressed model in knowledge distillation is effectively improved, and the practicability of cooperative training is improved.

The present application relates to the technical field of load balancing, and discloses a multi-model-based adaptive load balancing method and system. The method includes: obtaining a multi-dimensional resource information data set of multiple server nodes and performing implicit fuzzy K-means clustering to obtain a fuzzy partitioning result; generating a target similarity matrix based on the fuzzy partitioning result and constructing an effective load topology map; performing historical load records and network topology analysis to generate dynamic load prediction results; modeling multiple server nodes in the effective load topology map as a multi-agent cooperative game, obtaining the cooperation strategy of each server node and establishing a multi-model load balancing mechanism; performing load distribution calculation on the dynamic load prediction result based on the multi-model load balancing mechanism to obtain an initial load distribution strategy; performing convex optimization processing on the initial load distribution strategy to obtain an optimal load distribution plan. The present application improves the accuracy of adaptive load balancing.

This invention provides a method and optimization method for acquiring bolt preload fitting information, a terminal, and a computer storage medium for acquiring bolt preload fitting information. The method for acquiring preload fitting information includes: based on the bolt's stress loading strategy, performing a preload acquisition process at each loading stage to obtain a preload value corresponding to each target stress value; and acquiring bolt preload fitting information based on each target stress value and its corresponding preload value. The method for optimizing preload fitting information includes: based on the bolt's stress load and the initial number of loading stages, repeatedly executing the preload fitting information acquisition and optimization process; during repeated execution, updating the number of loading stages based on the execution results to update the stress loading strategy, thereby acquiring optimized preload fitting information. This allows the preload fitting information to better match the actual information, improving the accuracy of the preload fitting information.

The present invention discloses a method for load-aware scheduling of data flow roles in a heterogeneous environment. First, a data flow graph model representation is input, then the load factor is initialized, a perception queue is established, and the LAQPSA algorithm is used for scheduling and enqueueing. Finally, the scheduling result is output. The role division mechanism proposed by the method of the present invention further describes the differences of heterogeneous resources, defines the communication overhead of roles through the method of enqueueing predecessor roles, better defines the differences of computing platforms using the extreme value description method, makes it easier to mine the parallelism in the data flow graph by establishing a perception queue, and the initial load factor and fluctuation value maintain the scheduling balance of the computing platform on the basis of exerting the processing capabilities of heterogeneous platforms, further reducing the overall scheduling cross value and increasing the scheduling speedup ratio, meeting the scheduling requirements in a heterogeneous environment, and at the same time effectively improving the scheduling performance of data flow roles and mining the concurrent characteristics of the data flow graph.

The invention discloses a game resource file loading method based on access sequence sorting, relates to the technical field of game engine resource scheduling, is used for solving the problems of unreasonable resource loading sequence and increased loading delay, and aims to construct a dependency graph by analyzing resource metadata, represent resource files by nodes and represent dependencies by directed edges. Executing loop detection and topological sorting to obtain structure parameters, generating an initial loading sequence by adopting Pareto sorting in combination with a topological structure and an access frequency, monitoring a decompression ratio and a calling frequency of loaded resources in real time, and calculating the activeness; and collecting the number of request times of unloaded resources and the dependency chain depth, and estimating the demand intensity. And based on the Pareto level, the activeness and the demand intensity, constructing an adjustment factor, reordering a loading sequence, reallocating tasks, checking dependency consistency, preferentially loading key resources, and realizing dynamic scheduling optimization.

The application discloses a power distribution load prediction method and system, and relates to the technical field of graph calculation, and comprises the following steps: obtaining load data and auxiliary features of a user, and constructing a load portrait; according to the load portrait, selecting a model subset from a pre-trained model library, and migrating the selected model to output an initial load prediction result; constructing a load correlation graph according to spatial adjacency and behavior similarity between users, and propagating load features in the load correlation graph; according to the propagated load features and the initial load prediction result, regulating and controlling the model of a target user; obtaining the distribution change of the regulated and controlled user load data, and identifying behavior mutation; through multi-model integration, individual fine-tuning, load correlation graph construction, incremental learning and drift detection, the application effectively solves the problems of low precision, poor dynamic adaptability and slow reaction to sudden changes in traditional power distribution load prediction.

The invention discloses a tide financial service scheduling method and system, computer equipment and a medium. The method comprises the steps of collecting original real-time load data of a plurality of heterogeneous target financial businesses in past fixed duration, and performing cleaning processing to obtain real-time load data; and projecting the real-time load data to a corresponding time slot of the sliding time window according to the timestamp. And combining the load data volume of the same service type in each time slot and carrying out logarithmic processing. And correcting the load count in each time slot by using the time penalty curve of the target financial service. And correcting the initial load score according to the service importance to obtain a load score. And normalizing the load scores, and determining the real-time load ratio of each service. And allocating computing resources to each service according to the proportions. According to the method, resources are dynamically evaluated and allocated through a mechanism of a time window and a time slot in combination with a time attenuation model and a service weight model, so that the flexibility and fairness of resource allocation are improved.