796 results about "Cluster based" patented technology

A method, system, and device for wind speed prediction and layout optimization in wind power generation are provided. The method includes: obtaining a basic wind resource dataset of a target region; constructing a physics-informed neural network model based on the basic wind resource dataset; obtaining wind speeds data at a specific location in a velocity field based on the physics-informed neural networks and constructing a training dataset; training the physics-informed neural network model based on the training dataset; reconstructing a wind speed distribution within the velocity field and predicting wind speeds for a next time period with a wind farm using the trained physics-informed neural network model; and optimizing a layout of a wind turbine cluster based on a reconstructed wind speed distribution within the velocity field. The present application reconstructs a two-dimensional velocity field of the wind farm by training the PINN and enables accurate ultra-short-term wind speed prediction.

The invention provides an intelligent decision-making method based on multi-dimensional risk coupling modeling of new energy ships and related equipment. The method comprises the following steps: fusing multi-modal data of each new energy ship to obtain a target multi-modal feature vector; mining risk implicit association strength by using a large language model, and constructing a risk knowledge graph; performing risk time sequence evolution prediction by adopting a dynamic Bayesian network based on the atlas to obtain a single-ship risk prediction result; constructing a graph structure according to the single ship risk and the operation parameters, and carrying out space coupling modeling by adopting a graph convolutional network to obtain a cluster risk prediction result; and generating an optimal operation and maintenance strategy by adopting reinforcement learning based on a cluster risk prediction result and a reward function by taking a high-fidelity digital twinborn body as a virtual environment. Therefore, according to the method, the problem of closed-loop adaptive control from risk deduction to decision response in a complex navigation scene is effectively solved by constructing a unified modeling mechanism of multi-dimensional risk coupling and linking real-time control strategy generation.

The invention belongs to the technical field of computer software, and particularly relates to a heterogeneous database unified deployment management method based on a cluster management platform, which comprises the following steps: predefining metadata of various database services; the method comprises the following steps: pre-constructing a multi-version program package warehouse, and establishing a structured warehouse directory system; obtaining a chip architecture and an operating system type of the target node, and generating environment fingerprint information; dynamically selecting a database installation package matched with the current environment and a dependency item set of the database installation package by mapping a warehouse directory system; through a unified database adaptation layer, a standardized operation interface is called, and installation, configuration and start-stop operation of the database are executed; a visual deployment and arrangement function based on component roles is provided by expanding a Web service interface of the management platform; and according to a deployment arrangement result and the environment fingerprint information, calling a deployment script to complete automatic deployment and configuration of the database cluster. According to the invention, unified adaptation, deployment and full-life-cycle management of various domestic databases are realized.

The invention discloses an Internet of Things cluster data analysis method based on big data and artificial intelligence, and relates to the technical field of data analysis, and the method comprises the steps: carrying out the initialization and parameter synchronization processing of an edge side feature coding sub-model and a cloud space-time depth model through a space-time diagram sequence, obtaining a collaborative modeling parameter set, and obtaining a collaborative modeling parameter set; carrying out self-supervised reconstruction and prediction task training on the cloud space-time depth model based on the collaborative modeling parameter set, obtaining a depth representation model, learning abnormal mode parameters by using the depth representation model, constructing an abnormal scoring function, obtaining an abnormal detection model, and after receiving real-time multi-source heterogeneous data at an edge node, carrying out real-time reconstruction and prediction task training on the cloud space-time depth model; performing state prediction and anomaly score calculation by using the anomaly detection model to obtain a real-time anomaly score result; through the intelligent data analysis method, the accuracy, the real-time performance and the interpretability of Internet of Things cluster data analysis are improved.

The invention relates to the technical field of cloud services, in particular to a purchase quotation method and system based on clustering and association rules, and the method comprises the following steps: collecting historical purchase order data of internal products of an enterprise and external related market data, the historical purchase order data comprising product description; carrying out cleaning and standardized preprocessing on the collected data; the method has the advantages that through clustering based on product description, products such as servers and switches with different descriptions but similar essence are sorted and classified, data chaos caused by description differences is eliminated, and association rules mined on the basis are more targeted. Compared with a procurement quotation mode purely depending on experience, the procurement quotation method has the advantages that the procurement quotation accuracy is greatly improved, an enterprise can purchase required products at a more reasonable price, and cost waste caused by price judgment errors is avoided.

The invention discloses a manned aircraft / unmanned aerial vehicle pursuit deep learning control method simulating a pigeon positive and negative feedback mechanism. The method comprises the following steps: step 1, giving mathematical models of a manned aircraft and an unmanned aerial vehicle and performing environment initialization; step 2, building a quaternion PID controller of the unmanned aerial vehicle; 3, a manned aerial vehicle / unmanned aerial vehicle hierarchical decision-making architecture is established; step 4, mathematical modeling of a pigeon positive and negative feedback mechanism; 5, designing a manned aerial vehicle / unmanned aerial vehicle cluster reward function simulating positive and negative feedback of the pigeon flock; 6, planning and designing an escape unmanned aerial vehicle path; step 7, predicting an escape unmanned aerial vehicle path; and 8, generating a manned aerial vehicle / unmanned aerial vehicle cluster pursuit learning strategy based on the actor-commentator network. According to the invention, an efficient, flexible and extensible bionic intelligent pursuit control framework is constructed, and the pursuit task execution efficiency and cooperative combat capability of the manned aerial vehicle / unmanned aerial vehicle heterogeneous cluster in a complex dynamic environment are improved.

The invention provides a public opinion video tag aggregation method and system based on artificial intelligence, and relates to the technical field of artificial intelligence. Comprising the following steps: acquiring pictures and text information in a short video, and performing semantic alignment; different large language models are adopted to generate preliminary labels for the pictures and the text information after semantic alignment; clustering the pictures and the texts after semantic alignment to obtain clusters; calculating the labeling probability of each primary label type in the current cluster by each large language model, and selecting the primary label with the highest probability sum as a clustering label of the current cluster; calculating the reliability weight of each large language model in the current cluster based on the clustering label of the current cluster; and based on the reliability weight, calculating the weighted support degree of all the large language models to different preliminary label types of each piece of data in the current cluster, calculating the weighted label of the current data, and further determining a final label. According to the method, the condition of few labels or no labels can be effectively processed, and the manual workload is greatly reduced.

The invention belongs to the technical field of computers, and particularly relates to a resource allocation method and device, equipment and a storage medium. The method comprises the following steps: S1, acquiring a target resource occupation task and real-time state data of a plurality of clusters; s2, calculating a cluster resource utilization rate based on the total resource configuration amount and the resource usage amount; calculating a residual resource proportion of the cluster based on the resource residual amount and the resource demand parameter; calculating a load sensitivity coefficient of the target resource occupation task based on the task priority, the resource demand parameter, an execution duration threshold and a task dependency identifier; and S3, screening the clusters according to preset screening conditions corresponding to different task guarantee levels, determining the clusters meeting the preset screening conditions as candidate clusters, monitoring in real time, triggering resource migration based on quantitative migration conditions, and forming a closed-loop process of data acquisition-index calculation-screening distribution-preemption coordination-monitoring migration.

Computer-implemented methods for recommendation prioritization for a container orchestration system. Aspects include receiving a set of recommendations for a cluster of a container orchestration system. Aspects also include selecting an optimal recommendation from the set of recommendations using a scored knowledge transform graph. Aspects further include generating a confidence score for the cluster based on the optimal recommendation. Aspects also include determining a category of a readiness assessment model for the cluster using the confidence score. Aspects further include modifying a computer resource of the cluster based on the category of the readiness assessment model.

The invention discloses an active fluctuation collaborative stabilizing method and system for a high-proportion distributed new energy power grid, and relates to the technical field of power grid dispatching. According to the method, a physically consistent weather prediction model is established through multi-source meteorological data fusion, and a regional fluctuation propagation rule is accurately captured; identifying a high-risk fluctuation cluster based on dynamic time warping and spectral clustering, simulating a fluctuation propagation path by using a digital twin platform, and quantifying resource requirements; energy storage resource configuration is optimized by adopting mixed integer programming and a column generation algorithm, and multi-dimensional stability verification is carried out through a digital twin environment; a self-adaptive optimization mechanism based on reinforcement learning is established, continuous evolution of the system is realized, the technical bottlenecks of a traditional method in the aspects of fluctuation perception, resource allocation, system self-adaption and the like are solved, a collaborative stabilization mechanism with accurate prediction, intelligent recognition and decision optimization is formed, and a complete solution is provided for safe and stable operation of a high-proportion new energy power grid.

In various examples, systems and methods are disclosed that relate to the generation of images of cluster-based structures. For example, a system can obtain scene data associated with a scene of a three-dimensional environment, the scene comprising a plurality of objects; determine a first set of surfaces and a second set of surfaces, each surface of the first set of surfaces and the second set of surfaces corresponding to at least one object of the plurality of objects; and update the surfaces of the first set of surfaces based at least on a classification associated with the first set of surfaces. In examples, the system can generate an image based at least on updating the first set of surfaces. Updating the first set of surfaces can include tessellating the primitives of each surface of the first set of surfaces in accordance with a tessellation factor.

The invention discloses a kubernetes cluster resource adjustment method and device, equipment and a medium, and relates to the technical field of computers, and the method comprises the steps: collecting a target index, and analyzing the target index to construct a resource demand portrait of an application program; the target indexes comprise performance indexes, resource indexes and state indexes of all nodes and application programs of the kubernetes cluster; the resource demand of each application is predicted, when the node resource pressure index exceeds a target threshold value, a scheduling strategy is determined based on the resource demand and the resource demand portrait, and the to-be-migrated Pod is scheduled to a target node based on the scheduling strategy to complete resource adjustment; if all the nodes in the cluster do not meet the resource demand, calling a pre-registered node pre-registered to the cluster based on a preset node resource pool, adjusting the schedulable state of the pre-registered node, expanding the capacity of the cluster according to the adjusted node, and adjusting the resource based on the expanded cluster. And dynamic adjustment of resource allocation is realized.

The invention relates to the technical field of industrial internet, in particular to a control system and method for stirrer production. Comprising an edge data preprocessing unit, a flexible production instruction distribution unit and a working condition diagnosis and protection control unit. By constructing an integrated control parameter mapping model, an industrial operation signal is converted into a collaborative sensing digital primitive; the production task is compiled into an intelligent production execution process cluster based on a resource scheduling algorithm, and equipment is driven to execute differentiated production; and performing situation assessment on the digital primitives by using the anomaly detection model, and generating a regulation and control instruction to execute process logic recombination and safety protection. According to the invention, the problems of low production information integration level and tight data in each link of a production control system are solved, and flexible scheduling and active security defense in the production process are realized.

The invention discloses a production and manufacturing enterprise cluster chain type interaction carbon emission reduction method based on dynamic electrical carbon factors, and relates to the technical field of industrial carbon emission reduction and energy management systems. According to the method, a four-layer decision-making framework of a global coordination layer, a cluster coordination layer, an enterprise execution layer and an equipment control layer is constructed, and dynamically updated electric carbon factors are introduced, so that accurate identification, carbon emission accounting and collaborative optimization of enterprise clusters on a product chain are realized. The core of the method is that carbon quota allocation, production scheduling and resource allocation are carried out among hierarchies by utilizing a multi-objective and double-layer optimization model, so that the total carbon emission of a cluster is minimized while the economic benefit is ensured. A dynamic monitoring and feedback mechanism is also established, the power carbon factor weight can be adjusted and a resource allocation plan can be triggered according to real-time data such as the proportion of clean energy of the power grid, closed-loop regulation from a macroscopic strategy to microscopic operation is formed, the overall carbon footprint of an industrial chain is effectively reduced finally, and the participation barrier of small and medium-sized enterprises is broken.

In order to solve the problems of reliability reduction and task interruption caused by individual faults when an unmanned aerial vehicle cluster executes a task, the invention provides an unmanned aerial vehicle cluster task recovery method based on time sequence prediction, and post remedy is converted into pre-prevention. The system adopts a distributed architecture, a single machine continuously collects neighbor multi-dimensional states to form a sequential sequence, and a future window failure probability is predicted on an airborne LSTM; when the predicted value exceeds a dynamic threshold value, recovery is actively triggered, and an optimal position covering scheme and an executor are determined by adopting distributed negotiation with the optimal gain-energy consumption ratio as a criterion; and the executor adjusts the track in advance to realize seamless replacement. According to the method, smooth handover can be completed before failure, the task interruption time is remarkably shortened, and the resource utilization rate and the cluster task success rate in a complex environment are improved.

The invention provides an automatic fault tolerance method for distributed training of a model in a large-scale cloud native scene, and relates to the technical field of data management, and the method comprises the steps: collecting and based on hardware monitoring data of each node in a cluster, scheduling a distributed training task to each node, and starting check point storage; monitoring the running state of each training task, collecting a training log and a chip acceleration platform log, and collecting CPU, memory and acceleration card resource index data of each node and each training task; building a model training fault detection classification model by combining a supervision and machine learning method on the basis of the collected logs and hardware index data; and based on the model training fault detection classification model, judging whether each running training task has a fault and the fault type, and if the node equipment has a fault, rescheduling and loading the latest check point data to finish training task breakpoint continuous training and automatic fault tolerance. According to the invention, model breakpoint continuous training is realized, and the stability and efficiency of model training are improved.

The invention relates to the technical field of equipment fault prediction and adaptive scheduling. The industrial IoT equipment fault prediction and adaptive scheduling method based on the edge AI comprises the following steps: dividing edge nodes into equipment communication clusters based on equipment type isomorphism, and carrying out lightweight privacy aggregation processing on a directional pruning gradient at a preset cluster head node to generate an encrypted cluster-level gradient packet; performing double-stage global model updating operation on the encrypted cluster-level gradient packet at a cloud coordination layer to generate a new generation of fault prediction model; deploying the new-generation fault prediction model to an edge node, and triggering a preventive scheduling instruction set based on a judgment result that a fault risk entropy value output by the new-generation fault prediction model exceeds a dynamic threshold value, the technical effects of improving the privacy security of model training, reducing the communication overhead of edge heterogeneous equipment and enhancing the collaboration of a prediction result and scheduling control are achieved.

The invention discloses an MoE expert deployment system and method based on a wafer-level chip, and the system comprises a statistical module which is used for carrying out the statistics of expert co-activation probability distribution reflecting the cross-layer cooperative activation relation between experts in MoE and the communication demands of a calculation core on the wafer-level chip; and the clustering module is used for performing clustering based on the expert co-activation probability distribution and communication requirements by taking the minimization of communication traffic across physical regions and the minimization of intra-group communication load difference of expert groups borne by the physical regions as targets to obtain an expert grouping scheme of MoE and a physical region layout mapping scheme corresponding to wafer-level chips. According to the method, clustering can be carried out according to the cross-layer co-occurrence rule of the experts, and the experts subjected to high-frequency cooperative activation are constrained in the same physical region, so that a large amount of global communication is converted into local exchange, and the overhead of long-distance and high-delay cross-region communication is remarkably reduced. Meanwhile, load distribution is optimized in combination with communication requirements, and local hotspots are effectively avoided.

The invention discloses an unplanned operation early warning method and system fused with multi-source data mining. According to the method, vehicle tracks, operation plans and electronic fence data are collected, cleaning and map matching preprocessing are carried out on the tracks, and then track sequences are mapped into low-dimensional embedded vectors through a track representation learning model; recognizing an abnormal behavior mode through clustering based on the vector, and constructing a dynamic risk index and predicting the future state of the vehicle in combination with real-time data; inputting the abnormal mode, the risk index and the prediction result into a fuzzy logic decision maker for multi-source fusion, outputting a comprehensive risk level and triggering graded early warning; a reinforcement learning mechanism is adopted to dynamically optimize decision maker parameters according to the early warning effect, and finally early warning information is pushed to a management platform and feedback is received to form closed-loop management. According to the invention, accurate and adaptive early warning of unplanned operation behaviors is realized, and the intelligent level of operation safety supervision is significantly improved.

In various examples, systems and methods are disclosed that relate to the generation of images of cluster-based structures. For example, a system can obtain a depth buffer for a scene based at least on the performance of one or more ray tracing operations. The system can then determine an update to a position of a camera involved in performing the ray tracing operations and reproject the points represented by the depth buffer to generate an updated depth buffer. In examples, the system can then update at least one object of the plurality of objects based at least on a hierarchical depth buffer associated with the updated depth buffer and one or more tessellation rates.

The embodiment of the invention discloses a cluster tenant resource scheduling method and device, electronic equipment and a storage medium, relates to a cloud computing technology and can be used in the field of financial science and technology, and the method comprises the following steps: predicting a future resource demand of a target tenant namespace based on historical resource use data of the target tenant namespace in a cluster, the target tenant namespace comprises a tenant namespace bearing a key service; determining a real-time resource quota of the target tenant namespace based on the available resources of the shared pool and the future resource demand predicted value, the service level and the comprehensive priority of the target tenant namespace; and performing real-time resource scheduling on the target tenant namespace based on the real-time resource quota. According to the embodiment of the invention, the service quality of key services in a multi-tenant mixed deployment scene can be improved, and the resource utilization rate can be effectively improved.

The invention relates to the technical field of data processing and analysis, in particular to a knowledge graph entity accurate identification relation system for a service system, which comprises the following steps of: analyzing database table data, extracting an attribute change track and generating an attribute evolution sequence to provide a time sequence data basis for subsequent analysis and ensure the traceability of attribute evolution; meanwhile, the adaptability of the system to heterogeneous data is enhanced; clustering and dividing attribute clusters based on an attribute change mode, establishing a cross-cluster association index, revealing a potential association rule between attributes, and supporting rapid classification and dynamic adjustment of attribute relationships in a service scene; dynamic analysis of intra-cluster and cross-cluster attribute influence is realized by using a bidirectional attribute propagation model, and coupling strength between attributes is accurately quantified through iterative calculation and feedback calibration; by constructing a dynamic attribute topological graph and monitoring attribute value changes in real time, association reconstruction is triggered to generate an attribute association group set, dynamic changes of a service system are adapted, and association accuracy is improved.

The invention discloses a wind power plant power prediction method and device based on space-time collaboration and a storage medium, and belongs to the technical field of deep learning, and the method comprises the steps: obtaining historical wind field data and geographic space features of a wind power plant, the wind field data comprising fan operation data and meteorological feature data in the same time period; combining the historical wind field data, the geographic features and the corresponding total power labels of the wind power plant into a sample set, and training a wind power plant power prediction model by using the sample set; wherein the wind power plant power prediction model extracts local and overall double-layer similarity based on historical wind power plant data, obtains a time period cluster with high similarity with meteorological feature data from historical fan operation data according to the double-layer similarity, and performs spatial clustering on the time period cluster based on historical geographic features; and inputting the to-be-measured wind field data and the to-be-measured geographic features into the trained wind power plant power prediction model to obtain the total power prediction value of the wind power plant, thereby improving the wind power prediction precision and efficiency of the complex terrain.

In general, techniques for dataset clustering and artificial intelligence (AI)-assisted theme extraction are described. In an example, a method comprises computing, by a data management platform, chunk embeddings for respective chunks obtained from a dataset; generating, by the data management platform, based on the chunk embeddings, a cluster hierarchy having a plurality of clusters, each cluster of the plurality of clusters including one or more of the chunk embeddings; generating, by the data management platform, using a machine learning model, a theme for a cluster of the plurality of clusters, the theme generated by the machine learning model based on respective chunks of the one or more of the chunk embeddings included in the cluster; and outputting, by the data management platform, an indication of the theme for the cluster.

The invention discloses a serialization-based point cloud over-segmentation method, device and equipment and a medium, and relates to the technical field of point cloud over-segmentation. The point cloud over-segmentation method comprises the following steps: obtaining point cloud data set preprocessing; serializing the preprocessed point cloud data into a Hilbert curve; dividing the serialized point cloud data into a plurality of initial segments based on spatial continuity, and obtaining a multi-scale segment set; the feature similarity between adjacent segments is calculated, and a similarity matrix is obtained and established to provide a basis for subsequent super-point clustering; carrying out super-point clustering on the initial segments based on a self-adaptive updating algorithm; updating super-point features through a cross attention mechanism, establishing dynamic association between points and super-points, and obtaining a new super-point structure; and according to the updated super-point features, constructing a super-point graph, performing enhancement processing through a graph convolutional network, fusing features from the backbone point cloud network and corresponding multi-level super-point features, then transmitting the fused features to a segmentation head for segmentation, and obtaining semantic segmentation output.

The invention discloses a patent clustering method and system based on a pre-trained heterogeneous graph neural network, and relates to the technical field of data analysis. In the pre-training stage, a patent heterogeneous graph network is constructed based on target domain patent data; performing message transmission through the meta-path based on the current heterogeneous graph network to obtain patent meta-path representation; calculating a Transform attention score based on the representation; and pre-training the patent network based on the attention aggregation meta-path representation. In the downstream fine tuning stage, on the basis of patents in the other field, patent meta-path representations are obtained in accordance with the pre-training stage, downstream meta-path representations are aggregated on the basis of attention scores of pre-training, and the sensitivity of downstream clustering tasks is enhanced by using prompt vectors; the patents are clustered based on the aggregated characterization; and sending the patent clustering data to the target terminal equipment according to the sequence. According to the method, pre-training and prompt learning are combined to reinforce low-resource adaptability, patents can be accurately clustered, and the user field technology retrieval and analysis efficiency is improved.

The invention relates to the field of data analysis, in particular to a straw replacement nutrient release prediction method under a water-fertilizer coupling condition, and the method comprises the steps: carrying out the collection and segmentation processing of soil environment time sequence data, and obtaining an environment segment feature data set; performing quantitative evaluation on the temperature and humidity collaboration of the environment segment to obtain a collaborative window proportion correction factor; analyzing the environment fragment main limiting factor structure to obtain a main limiting factor structure correction factor; obtaining a decomposition mode correction distance by performing cooperative and restrictive joint correction on the standard Euclidean distance; a nutrient release prediction result is obtained by carrying out decomposition mode clustering and calibration analysis on the environment fragment, so that the problem of mode confusion caused by the fact that a cooperative difference and a main limiting factor difference cannot be distinguished by an existing clustering method based on statistical characteristics is solved.

The invention relates to the technical field of database query optimization and distributed system data processing, and discloses an intelligent routing optimization method and system for multi-table association query, and the method comprises the steps: a query topology analysis unit analyzes an object relation mapping configuration file to construct a query topological graph; the environment parameter detection unit collects real-time state data of the server cluster based on the query topological graph, wherein the real-time state data comprises a pre-estimated cache hit rate, associated sparseness and a target table data line number estimation value; the routing strategy decision-making unit compares the real-time state data with a preset threshold value, and selects a cache-driven step-by-step query strategy, a column-type compression transmission strategy or an asynchronous intermediate table unloading strategy; and the dynamic execution control unit executes data acquisition and assembly operation according to a selected strategy. According to the method, the problems of low database computing resource utilization rate and application server memory overflow caused by the adoption of a static strategy in a traditional framework are solved, and the system response speed and the service availability of multi-table association query are effectively improved.

The invention belongs to the technical field of power distribution network planning, and particularly relates to an active power distribution network multi-level dynamic partitioning method and device based on cluster division, and the method comprises the steps: firstly carrying out the low-voltage class cluster division, respectively constructing an electrical modularity fusion degree index, a power balance degree index and an intra-cluster and inter-cluster coupling degree index, and carrying out the weighting to form a cluster division comprehensive index; dynamically adjusting and updating the corresponding weight; performing cluster division based on a community discovery algorithm; and then performing high-voltage class cluster division, defining a comprehensive objective function of a high-voltage layer, performing high-voltage layer initialization, performing initial partitioning and feasibility check, defining a small cluster number constraint, iteratively optimizing cluster division by adopting a gradient descent method until convergence, and outputting a final cluster division scheme. According to the method, through low-voltage layer fine division, super-node modeling, high-voltage layer optimization and cross-level cooperative adjustment, and in combination with improved modularity, power balance degree and intra-group and inter-group coupling degree, the cluster structure characteristics of the power distribution network are remarkably optimized.

The invention relates to the technical field of power grid monitoring, in particular to an operation monitoring method and system for a distributed power distribution network, and the method comprises the steps: generating a plurality of multi-dimensional monitoring surfaces based on a topological structure, real-time electrical parameters and a load side dynamic response simulation result of the distributed power distribution network; determining a plurality of groups of electrical parameter time sequence chains and generating a parameter chain cluster based on the interaction data of the multi-dimensional monitoring surface and the distributed power distribution network and the load side dynamic response data; and the multi-dimensional monitoring surface is a monitoring carrier which is appointed by a user and covers a distribution network topology area and a load response area. According to the method, the load absorption capacity of the distribution network and the power characteristics of the operation state unit can be accurately evaluated by analyzing time series data such as load fluctuation, power change and voltage change, the dynamic response characteristics of the distribution network can be deeply analyzed, and possible load risks can be recognized in advance.