54 results about "Echo state network" patented technology

The invention relates to a boiler flue gas temperature dynamic compensation method based on CFD-reinforcement learning, and belongs to the technical field of boiler operation control. The method comprises the steps that boiler operation parameters and flue gas temperature data are collected in real time, a multi-scale coupling three-dimensional CFD model is constructed, a macroscopic gas-solid two-phase flow-combustion reaction-radiation heat transfer coupling algorithm and a microcosmic pulverized coal particle combustion dynamics algorithm are fused, and hearth and flue temperature distribution is simulated and output; building a CFD-reinforcement learning fusion agent based on a CFD model, taking similar working condition agent parameters as initial values, combining historical and virtual data staged training, and reinforcing the weight of a key temperature measurement point through a focusing regulation and control strategy; an echo state network-predictive control collaborative algorithm is adopted, real-time data are input to obtain an initial adjustment amount, and after security constraint optimization, a compensation execution mechanism is controlled to act. The flue gas temperature is accurately and dynamically compensated, the adaptability to complex working conditions is improved, and equipment safety and operation efficiency are both considered.

The invention provides a method and a system for optimizing power closed-loop response time of a wind and light storage system. The method comprises the following steps: collecting multi-source historical operation data of the wind and light storage system within continuous preset duration; carrying out global optimization on key parameters of the echo state network by adopting a sethead whale migration algorithm; based on the collected data, outputting an advanced prediction result of the output of the wind turbine unit, the output of the photovoltaic unit and the change trend of the energy storage SOC in a future time period through the optimized echo state network; taking the advanced prediction result as input, and based on a weighted objective function taking power tracking error minimization and power closed-loop response time minimization as targets, adopting an adaptive snake vulture optimization algorithm to solve the objective function to obtain optimal control parameters of the wind-solar storage system; the optimal control parameters are analyzed into real-time control instructions through the FPGA controller so as to drive the energy storage unit, the wind generating set and the photovoltaic generating set to execute power regulation and control, and collaborative optimization of power closed-loop response time shortening and tracking precision improvement is achieved.

The invention relates to an engine flame tube life prediction method, equipment and a program product. The method comprises the following steps: acquiring a crack propagation parameter of a target flame tube to be subjected to life prediction; inputting the crack propagation parameters into a pre-trained life prediction agent model, and outputting a residual life prediction value of the target flame tube through the life prediction agent model; wherein the life prediction agent model is obtained by training the echo state network by taking a flame tube crack propagation data set obtained by simulation as a training sample, and the flame tube crack propagation data set comprises a plurality of groups of crack propagation parameters and corresponding residual life prediction values. According to the method, the complex nonlinear relationship between the crack form and the life can be efficiently mapped, and the residual life prediction value can be quickly and accurately output; moreover, the method does not need to depend on complex real-time simulation calculation, can be directly connected with actual monitoring data, and effectively reduces the operation and maintenance cost.

The invention relates to an adaptive multi-scale decomposition echo state network chaotic time sequence prediction method suitable for meteorological prediction, and belongs to the field of chaotic time sequence prediction. Comprising the following steps: performing multi-scale decomposition on an input sunspot number sequence by using an HP filter; each sub-sequence obtained through decomposition is distributed to an ESN sub-network; inputting a weight in the ESN sub-network by adopting a Xavier weight initialization method, introducing a composite activation function in reservoir nonlinear updating, and performing state splicing by adopting a state splicing strategy; a sparse random matrix is constructed, a unit matrix scale factor is introduced, and the spectral radius is strictly controlled; each ESN sub-network independently trains an output weight, ridge regression is adopted for solving, integration and reconstruction are carried out after corresponding component prediction is completed, and if a target sequence is a sunspot activity index, the model finally outputs a sunspot number prediction value in a future time step; according to the method, high-precision, stability and robustness prediction of the chaotic time sequence can be realized.

The invention belongs to the technical field of artificial intelligence and network predictive maintenance, relates to a fault prediction method based on a multi-source migration echo state network, and aims to solve the problem of low prediction precision caused by insufficient early samples of a fault, and the method comprises the following steps: firstly, extracting migration knowledge from a plurality of historical fault source domains by using the echo state network; secondly, measuring the similarity between the source domains and the target domain and the difference between the source domains by adopting a dynamic time warping distance, and selecting a plurality of similar source domains with complementary information; then, a plurality of prediction sub-models are established in the target domain in combination with migration knowledge of the selected source domain, and a final prediction model is obtained through integrated learning; and finally, performing future value prediction on the key variables acquired in real time by using the model. According to the method, multi-source historical information can be fully utilized under the small sample condition, the fault prediction precision and generalization ability are remarkably improved, and the method is suitable for fault prediction scenes with multi-source time series data in industrial processes, mechanical equipment, power systems and the like.

The application discloses a power distribution network stability control method based on multiple agents, comprising the following steps: deploying agents, collecting data and constructing graph data; establishing an improved graph echo state network and forming a structured input sequence; inputting the input sequence for time evolution and forming a dynamic echo state vector; exchanging the dynamic echo state vector with neighboring agents to splice and form comprehensive time sequence state information; each agent analyzes the comprehensive time sequence state information to generate a local stability control instruction; when a disturbance occurs, the comprehensive time sequence state information is updated and a disturbance updated local stability control instruction is generated; and the dynamic echo state vector is updated, the comprehensive time sequence state information is constructed and the local stability control instruction is generated periodically and continuously in a normal operation cycle. The application realizes power distribution network stability control by using an improved graph echo network, and has the advantages of strong real-time performance and fast disturbance response.

Procedure for determining the reliability of the modeling of the dynamics of a traffic intersection, comprising the steps: S1 - Recording sensor data regarding the movement of road users at the traffic junction; S2 - Using a trained deep learning network model with the sensor data, wherein the deep learning network model includes at least one echo-state network unit (ESN), and wherein a result of the deep learning network model is a model of the dynamics of the traffic node; S3 - Determining a learned regularization coefficient of the deep learning network model, where the regularization coefficient indicates the reliability of the modeling of the dynamics of the traffic junction.

A transformer health state evaluation method based on a leaky-integrator echo state network includes the following steps: collecting monitoring information in each substation; performing data filtering, data cleaning and data normalization on the collected monitoring information to obtain an input matrix; inputting the input matrix into a leaky-integrator echo state network to generate trainable artificial data, and dividing the artificial data into a training set and a test set in proportion; constructing a deep residual neural network based on a squeeze-and-excitation network, and inputting the training set and the test set for network training; and performing health state evaluation and network weight update based on actual test data. Considering that a deep learning-based neural network needs a large amount of data, the present disclosure uses the leaky-integrator echo state network to generate the artificial training data.

The invention belongs to the technical field of intelligent household appliances, and particularly relates to a water consumption prediction method, device and equipment of a water heater and a computer storage medium. By inputting collected actual water consumption information of a first time period into a pre-trained echo state network, the network can output predicted water consumption information for a plurality of future time periods. According to the method, the problem of inaccurate water supply caused by uncertainty of water use habits of a user in the actual use process of the water heater is effectively solved, so that the water heater can adjust the working state of the water heater more intelligently, and the provided water quantity is ensured to be matched with the actual demand of the user.

The invention provides a photovoltaic power station fault detection method, which comprises the following steps of: acquiring power generation operation, environment and equipment state multi-dimensional data of a photovoltaic power station, and performing standardization, dimension reduction and missing value repair by adopting quantum principal component analysis; constructing a causal relationship network based on a QPCA result, mining a dependency relationship among variables, and generating feature set data; training the feature set by using an echo state network model, and constructing an anomaly recognition model in combination with density peak clustering and an isolated forest algorithm; inputting the preprocessed data to the model in real time, and outputting abnormal data through clustering analysis; analyzing an abnormal influence range and a propagation path by combining a causal network, and quantifying a risk level; triggering an alarm for high-risk abnormity, calling a knowledge base to generate a disposal scheme and notifying operation and maintenance personnel; and processing feedback data is used for updating model parameters to form closed-loop optimization. The data processing efficiency is improved through a quantum algorithm, the detection precision is enhanced by fusing causal reasoning and deep learning, and dynamic risk assessment and model self-optimization are realized.

The application discloses a dynamic quantitative feeding control system and method based on real-time working condition feedback of a reaction kettle, and comprises the following steps: collecting reaction kettle temperature, pressure, flow and other data, performing denoising and standardization, and generating working condition input data set; calculating heat release intensity, energy residual and material residual based on the working condition input data set, and constructing a physical characteristic vector; inputting the physical characteristics into an improved echo state network, and outputting temperature and pressure prediction; constructing temperature, pressure and rate boundary conditions to generate a safety constraint set according to the prediction results and equipment limitations; using self-evolution waveform optimization based on the safety constraint to generate a candidate waveform set and select the optimal feeding curve; inputting the optimal waveform to control feeding, establishing a trust propagation graph, calculating a score gate and triggering a safety interlock. The application realizes accurate prediction and safety control of reaction kettle feeding through the improved echo state network and self-evolution waveform optimization.

The application provides a sequence prediction method and system of a pooling echo state network based on compressed sensing, belongs to the technical field of network information prediction, acquires sequence data required by a prediction task, constructs a pooling echo state network model based on compressed sensing, trains the network model, inputs the acquired sequence data into the trained pooling echo state network model, and obtains a prediction result; a pooling layer and a compressed sensing layer are added to a reservoir pool of the pooling echo state network model, the pooling layer is used for readjusting the weight of the state of the reservoir pool node, and the compressed sensing layer is used for sparse transformation and random subsampling of the node; based on the mechanism of compressed sensing and the pooling algorithm, the application provides a mechanism which can effectively reduce redundant nodes and improve the active performance of the nodes, effectively improves the model performance of the ESN, and makes the reservoir pool active while reducing the calculation amount, and improves the accuracy and operation efficiency of model calculation.

The application discloses a blast furnace gas generation amount prediction method based on an echo state network, and relates to the technical field of steel production. First, the application takes a mutual information entropy correlation coefficient as an evaluation index of influence factors of the blast furnace gas generation amount, sets an hour delay time for the influence factors and the blast furnace gas generation amount, and selects parameters with strong correlation as main influence factors of the blast furnace gas generation amount according to the degree of correlation. Then, historical production data of the previous three hours are selected as training data, and an echo state network model is trained and saved. Finally, data of the latest one hour are selected as model input of the echo state network to predict the blast furnace gas generation amount in the future one hour. The application can realize prediction of the blast furnace gas consumption, and provides strong guidance for operators to judge the change of the future blast furnace gas generation amount.

The invention is entitled "Dynamic Nonlinear Optimization of Battery Energy Storage Systems." The invention provides a system and method for optimizing a battery energy storage system (BESS) that can involve: inputting data to an echo state network, the data relating to operation of a battery energy storage system and including one or more of: load data, renewable energy data, non-renewable energy data, carbon emissions, carbon credits, and weighted energy costs; and optimizing the operation of the battery energy storage system based on the data input to and processed by the echo state network and relating to the operation of the battery energy storage system.

The invention discloses an industrial robot adaptive cooperative control method based on a multi-agent system, and the method comprises the steps: building an agent communication topology, and initializing an agent; collecting sensing and execution data of the intelligent agent, and preprocessing to obtain an ontology and domain operation state set; voxel discretization is carried out to form a space-time voxel occupation view; constructing a structure sharing track body object, and outputting an agent beat time window; constructing an improved echo state network model to obtain a prediction quantity and a compensation quantity; based on a time elastic band algorithm, obtaining an agent space-time trajectory; and executing the time-space trajectory of the intelligent agent to complete distributed cooperative control. By combining online prediction compensation of an improved echo state network model with rolling track updating of a time elastic band algorithm, real-time consistent cooperative control of tasks, tracks and beats of the industrial robot in a shared operation space is realized.

The application discloses a kind of accurate trigger synchronous multichannel vibration data acquisition system, comprising: data acquisition and pre-processing module, for reading data and generating vibration state sample and action state pair;Mutual inhibition reserve pool echo state network module, for executing dynamic feature extraction to vibration state sample;Event sensitive trigger gate module, for generating gated activation vector and timing screening is carried out to reserve pool activation state, output event response feature vector;Reverse memory drive DRQN model module, for receiving event response feature vector and executing forward recursion and reverse recursion, output Q value estimation and policy action;Error training module, for calculating the Q value error of policy action and generating training signal;Experience replay module, for constructing priority experience replay queue based on Q value error and selecting training segment to update DRQN model parameter.The application realizes the high-precision trigger and synchronous acquisition of multichannel vibration signal, improves the stability of feature and policy decision performance.

The invention provides a mechanical arm trajectory tracking control method fusing model prediction and sliding mode control. The mechanical arm trajectory tracking control method comprises the steps that a discrete sliding mode controller is constructed, and a discrete sliding mode surface based on trajectory tracking errors is designed; calculating a sliding mode control law; constructing a dynamics prediction model based on an echo state network, and performing online learning and updating by taking historical state data and a sliding mode control law of the mechanical arm as input so as to predict a state track of the mechanical arm in a future time domain; constructing a model prediction controller embedded with sliding mode control, and introducing a prediction state and a sliding mode control law into an optimization objective function of the model prediction controller; and under the model prediction controller, an optimization problem is converted into a quadratic programming problem to be solved, a smooth optimization control quantity meeting physical constraints is obtained, and the control quantity acts on the mechanical arm system. The high-frequency buffeting problem of sliding mode control is effectively solved, and meanwhile high-precision trajectory tracking of the mechanical arm under strong nonlinear interference is guaranteed.

The application discloses a kind of adaptive control optimization methods of manipulator, it is related to intelligent control technical field, utilize manipulator ontology perception network to obtain the real-time time series signal stream of driving pressure and current, real-time time series signal stream is converted into continuous time sequence characteristic matrix, and equivalent rigid body dynamics model is constructed by physical structure data, and state variable and error variable are defined, continuous time sequence characteristic matrix is input into echo state network, and reserve pool state vector is output by reserve pool structure mapping, and reserve pool state vector is defined as characteristic base function vector, Bellman error is calculated based on error variable, and reinforcement learning evaluation network is constructed using characteristic base function vector, weight estimation value and disturbance estimation parameter are calculated, aggregate uncertainty function is obtained, output control moment is calculated based on aggregate uncertainty function, and actual output moment of actuator is calculated by output control moment, to complete the control of manipulator.

The invention provides a power intelligent training and pushing all-in-one machine platform, and relates to the technical field of power AI training and pushing integration, and the platform comprises a data collection unit which is used for collecting an original operation data set of power equipment through a multi-modal data collection module; the data processing unit is used for processing the original operation data set and outputting a power equipment structured training data set and a real-time reasoning data set; the instruction generation unit is used for training and reasoning an improved deep echo state network training engine based on the data set, and generating a power equipment operation state evaluation result and an operation control instruction; and the equipment regulation and control unit is used for displaying the operation state evaluation result through the visual data interaction module, and transmitting an operation control instruction to the control terminal based on a communication link to complete closed-loop regulation and control, so that training and pushing integrated efficient cooperation can be realized, the development threshold is reduced, the computing power utilization rate and the operation and maintenance efficiency are improved, and the development cost is reduced. Intelligent upgrading of power core services is supported, and application stability and reliability are enhanced.

The present application relates to a kind of network control system time delay compensation method based on neural network sliding mode control, steps are as follows: establishing echo state network;Echo state network training update;Introduce myxomycete algorithm to optimize echo state network model, form the combination prediction model of echo state network based on myxomycete algorithm optimization Prediction network time delay value;According to network time delay value, sliding mode function is designed, the state equation when system is in sliding surface is obtained, and system control quantity is solved.The present application solves the control problem of linear system with input delay, effectively improves the prediction accuracy, uses the predicted result to combine sliding mode control algorithm to output future control quantity, compensates network control system time-varying time delay, improves the tracking ability of network control system signal.

The embodiment of the application provides a kind of substation metering system state prediction method, device and electronic equipment, it is related to sampling algorithm technical field.The method comprises: obtaining the historical data of the substation including environmental state quantity and corresponding metering quantity;The historical data is preprocessed;The preprocessed data is processed by closed loop clustering algorithm to obtain data classification under different scenarios, and the preprocessed data after data classification is divided into training set and validation set;The training set is used to train the combined model, and the validation set is used to verify the trained combined model;The combined model includes random forest algorithm, convolutional neural network algorithm and echo state network algorithm;The combined model after verification is used to output corresponding prediction result based on input environmental state quantity and corresponding metering quantity.The embodiment provided in the application effectively improves the intelligent level of substation operation and maintenance.

The application provides a mobile phone extreme charging wideband variable current adaptive regulation method based on deep learning, and relates to the technical field of mobile phone extreme charging. The method comprises collecting multi-dimensional operation data of the mobile phone extreme charging scene. A local linear embedding algorithm is used for feature extraction. A state of charge evaluation model based on a deep belief network is constructed to evaluate the operation state of the mobile phone extreme charging system in real time. An echo state network is used to construct a dynamic correlation relationship model of multi-dimensional data. A multi-objective optimization model is established, and a multi-objective artificial bee colony algorithm is used to solve the multi-objective optimization model to select the optimal variable current regulation strategy of the current charging condition. The variable current parameters are adjusted. The application is constructed through deep fusion of multi-dimensional data and a deep learning model, and the variable current regulation strategy is optimized by combining an intelligent algorithm, so that the charging efficiency is improved, the battery temperature rise and circuit loss are reduced, the intelligence and adaptability of the variable current regulation are enhanced, and the battery safety is ensured and the system service life is prolonged.

The present disclosure provides a method and system for predicting effluent ammonia nitrogen (NH4—N) and an electronic device. The method includes: obtaining data to be tested; and inputting the data to be tested into a trained deep echo state network, to obtain predicted NH4—N concentration. A method for establishing the deep echo state network includes: establishing an original network, where the original network includes a plurality of input variables and reservoirs, and a principal component analysis (PCA) mapping layer is added between adjacent ones of the reservoirs; initializing the original network to obtain an initialized network; performing parameter optimization on the initialized network by a matrix generation method of singular value decomposition and a competitive swarm optimizer (CSO) algorithm, to obtain an optimized network; and training and testing the optimized network, to obtain the trained deep echo state network.

This invention discloses a method and apparatus for managing cluster computing power devices based on deep learning, specifically including: S1, collecting operational status data of the cluster computing power devices; S2, performing time alignment on the operational status data to form a computing power status input sequence and an energy consumption and heat status input sequence; S3, constructing an echo state network model of two parallel reservoirs; S4, performing recursive mapping in each reservoir to generate a reservoir state matrix; S5, expanding the reservoir state matrix to form a joint state component set; S6, calculating the mutual information values ​​between the state components and the management output variables; S7, filtering the state components according to sparse constraints to form a sparse fused state vector; S8, constructing an output layer linear mapping structure based on the sparse fused state vector; S9, performing recursive least squares online updates on the output layer parameters and generating management decisions. This invention uses two reservoirs and online updates to achieve cluster management.

The application discloses a seasonal frozen region railway roadbed settlement prediction method based on an improved echo state network. First, a multi-sensor monitoring platform is built, railway roadbed settlement, air humidity and multi-depth soil humidity time series data are collected, and correlation analysis and feature screening are performed. Then, the VMD (Variational Mode Decomposition) and sample entropy reconstruction method are used to adaptively denoise and enhance the multi-scale features of the screened input feature sequence. Then, an improved echo state network (IESN) model is constructed. The model introduces a small-world network topology to generate a reserve pool connection weight matrix, and adds a configurable delay mechanism to enhance the ability to capture complex time series dynamic characteristics. Finally, an improved ivy algorithm (IIVYA) is proposed, which is a fusion of multi-elite reverse learning, Cauchy mutation and stable climbing strategy, which is used for global and efficient optimization of the hyperparameters of the IESN model to obtain the final prediction model.

The application relates to an echo state network-based adaptive fault-tolerant control method for a non-strict repetitive system, which comprises the following steps: constructing a nonlinear dynamic system with an actuator fault, and setting a hypothesis condition; introducing a desired error trajectory, constructing a dynamic error equation, and defining a nonlinear function in the equation; approximating the nonlinear function by using an echo state network; proposing an adaptive iterative learning fault-tolerant algorithm, combining a barrier Lyapunov function, deducing a control input, constructing a controller, and filtering redundant batches; and constructing a barrier composite energy function to verify the stability and convergence of the designed adaptive iterative learning fault-tolerant algorithm. The nonlinear dynamic system constructed by the application can still realize effective tracking and control of the system state to the desired trajectory under the conditions that the actuator has additive or multiplicative faults, the system state is limited, and external disturbances exist.

The invention discloses a multi-mobile-robot adaptive iterative learning inclusion control method under false data injection attack, and belongs to the technical field of multi-agent control. The method mainly aims at the inclusion control problem of the multi-mobile-robot system under the false data injection attack. The control scheme comprises the following steps: establishing a multi-mobile-robot system model and a state equation of the multi-mobile-robot system model; establishing a relationship between an original system state and an attacked state; defining an expected error trajectory and coordinate transformation; constructing a Lyapunov function, introducing an echo state network, and designing a virtual controller of a first subsystem; constructing a Lyapunov function, introducing an echo state network, and designing a control input signal; constructing a composite energy function, introducing a projection operator, and designing an adaptive law; according to the adaptive iterative learning inclusion control method disclosed by the invention, all followers can converge to a convex hull formed by a leader along with the increase of the number of iterations under the false data injection attack of the multi-mobile robot system.

This invention discloses a method for evaluating the operational status of a mining drilling rig system based on an echo-state Bayesian neural network. This method addresses the challenges of dynamic feature extraction, model overfitting, and the lack of uncertainty quantification in deterministic assessments of drilling rig status under conditions of strong vibration and high noise in underground environments. It employs an echo-state network with a leakage integral mechanism to suppress noise and extract slowly varying dynamic features from multi-source sensor time-series data, constructing a high-dimensional feature vector through dual-view feature aggregation. A Bayesian neural network with Concrete Dropout is introduced to achieve state probability mapping based on variational inference, and Monte Carlo sampling and prediction entropy are combined to quantify cognitive uncertainty and provide early warning of high-entropy anomalies. This invention effectively filters out high-frequency noise, suppresses overfitting in small samples, and enables proactive early warning of abnormal operating conditions. It achieves high classification accuracy on hard coal seam drilling datasets, and its assessment accuracy and decision reliability are significantly superior to traditional deep learning models.

The invention provides a total phosphorus removal rate and aeration energy consumption prediction method based on an adaptive sparse echo state network, and relates to the technical field of intelligent information. The method comprises the following steps: complementing missing data by using a data reconstruction method based on distance characteristics, establishing a multi-task echo state network to share task information, optimizing network parameters by adopting an adaptive sparse strategy, and realizing collaborative accurate prediction of the total phosphorus removal rate and the aeration energy consumption. The method solves the problems that in the prior art, in the sewage treatment process, nonlinearity is high, mechanism modeling is difficult, and blocking missing exists in operation data, so that a neural network model based on a single-task, complex and redundant structure cannot fully utilize multivariable potential information, the total phosphorus removal rate and aeration energy consumption are difficult to accurately and simultaneously predict, and the operation efficiency is low. The calculation burden is large; and the operation efficiency is low.

The application provides a sunken skin stretch forming path planning method, and belongs to the technical field of stretch forming path planning, and the method comprises the following steps: S1, acquiring the design parameters and forming constraint conditions of the sunken skin; S2, establishing a finite element model of the sunken skin; S3, initial stretch path global planning based on an echo state network-cell mapping hybrid algorithm; S4, finite element simulation verification of the initial stretch path; and S5, stretch path iterative optimization based on a fractional order Kalman filter-dove search hybrid algorithm. The application solves the problems of low stretch forming path planning precision, many forming defects, poor process stability, weak technical reproducibility and the like of the existing sunken skin stretch forming path planning.