567 results about "Neural net architecture" patented technology

The invention relates to the field of road traffic, in particular to a multi-feature fusion fatigue driving detection method and a fatigue driving detection system. The method comprises the following steps: extracting facial features from a face image of a driver; extracting vehicle features from the vehicle driving parameters of the vehicle driven by the driver; and fusing the facial features and the vehicle features to judge whether the driver is in fatigue driving. Extracting facial features by designing a CNN model; extracting basic convolution features; extracting local convolution features; extracting global convolution features; performing pooling operation; aggregating global features; and carrying out dimensionality reduction mapping. A self-encoder is designed to extract vehicle characteristics; a symmetric deep neural network structure is adopted, and high-dimensional time sequence data is compressed to a low-dimensional potential space through nonlinear mapping; through combination and matching of the CNN model and the auto-encoder, the technical defects of feature redundancy, noise interference, information loss and suboptimal decision existing in an existing multi-feature fusion fatigue driving detection system are thoroughly solved.

The invention relates to the technical field of exception prediction, in particular to a storage cabinet exception trend prediction system based on time series data analysis, which comprises a state monitoring module, an interval sensing module, a path reconstruction module, a symptom activation module and an evolution prediction module. According to the method, the state vectors including the temperature, the voltage, the current and the door lock state are constructed and combined with the timestamp information to form the time sequence data sequence, and the dynamic expression mode of state change is established; a jump characteristic is analyzed by using a ratio of a time interval to a state change amplitude, a short-time disturbance path and a trend evolution path are distinguished by combining a jump rate statistical index, and an evolution activation signal is identified based on trend maintenance and non-fallback characteristics. On the basis, a neural network structure with long-time dependent learning ability is introduced to capture an aperiodic thermal anomaly trend in a state sequence, and the accuracy and timeliness of anomaly recognition are improved through multi-dimensional parameter cooperative processing and path construction logic.

The invention discloses a power plant equipment fault prediction method based on a time sequence large model, and the method comprises the following steps: S1, collecting and preprocessing the time sequence data of a multi-source sensor of a power plant, and generating a standardized time sequence data set; s2, constructing a time sequence large model, inputting standardized data, and outputting a future operation state predicted value; s3, comparing the running state prediction value with an actual measurement value to generate a prediction residual sequence; s4, constructing a Bayesian neural network model, inputting a prediction residual sequence, and outputting error probability distribution; s5, optimizing a Bayesian neural network structure and hyper-parameters by adopting an ant colony optimization algorithm; s6, confidence interval estimation is executed, and whether the state is a high-risk state or not is judged; and S7, outputting a running state label, and dynamically acquiring a data closed-loop updating model. According to the invention, high-precision prediction and uncertainty evaluation of the operation state of the power plant equipment are realized, so that the accuracy and response time efficiency of fault early warning are improved.

The invention discloses a reinforcement learning method and system for source network load storage collaborative multi-scene optimization, and the method comprises the steps: constructing a power distribution network optimization model with the minimum cost, and converting a mixed integer nonlinear problem into a solvable mixed integer second-order cone optimization problem; converting a mixed integer second-order cone optimization problem into a reinforcement learning decision model, and performing multi-round assignment on boundary condition parameters by using different operation scene data to form a multi-scene training task set; designing a reinforcement learning decision model multi-scene training loss function, combining the reinforcement learning decision model to construct a strategy neural network, an evaluation network neural network and a scene representation embedded neural network, and completing reinforcement learning decision adaptive to multi-scene optimization; and training a power distribution network multi-scene optimization decision model based on the multi-scene training task set, the multi-scene training reinforcement learning loss function and the neural network structure, and deploying the power distribution network multi-scene optimization decision model to an actual system to complete reinforcement learning decision model application oriented to source network load storage collaborative multi-scene optimization.

The invention discloses a tight reservoir three-dimensional crustal stress field modeling method based on an improved neural network. The tight reservoir three-dimensional crustal stress field modeling method comprises the steps that S1, a unified-format multi-source geological physical data tensor set is constructed; s2, constructing a frequency domain hierarchical enhancement-SIREN implicit neural network structure based on the unified format multi-source geological physical data tensor set; s3, inputting the candidate hyper-parameter configuration into the frequency domain hierarchical enhancement-SIREN implicit neural network to complete one-time model training; s4, aiming at each candidate hyper-parameter configuration, initializing an inner-layer population of the black widow optimization algorithm, completing second model training, and obtaining an optimal model parameter of the frequency domain hierarchical enhancement-SIREN implicit neural network; s5, tight reservoir fracturing parameter optimization and real-time safety window adjustment are achieved. According to the method, the continuous stress field can be quickly generated at the resolution of 1 m, real-time well section updating and fracturing scheme optimization are supported, and the fracturing transformation effect, the fracturing safety margin and the reliability of economic productivity prediction are remarkably improved in practical application.

The invention discloses a lithium battery residual life prediction method and system and terminal equipment, and relates to the technical field of lithium battery health management. The method comprises the following steps: receiving a battery capacity attenuation sequence as an original input sequence, detecting and filtering abnormal data by adopting a 3 sigma criterion, and carrying out noise suppression processing on the battery capacity attenuation sequence through a Dropout mask; and carrying out normalization processing on the preprocessed battery capacity attenuation sequence, dividing the battery capacity attenuation sequence into a training set, a verification set and a test set through a sliding window algorithm, and generating a time sequence characteristic matrix and a corresponding residual service life label. According to the method, a neural network structure fusing trend prior perception and dynamic attention regulation is constructed, a multi-scale capacity modeling strategy is introduced to separate a degradation trend, fluctuation disturbance and high-frequency noise, and compared with a traditional time sequence neural network or a single attention model, pseudo fluctuation characteristics caused by capacity regeneration can be more effectively recognized, and the method is more efficient and more reliable. And the judgment accuracy of the model in a complex degradation scene is improved.

The invention discloses a millimeter wave SAR icing thickness inversion algorithm based on deep learning, and aims to solve the problems of insufficient icing monitoring precision and poor real-time performance under complex meteorological conditions. According to the algorithm, high-quality input is provided for deep learning modeling through data preprocessing including slope distance correction, polarization denoising, back scattering characteristic extraction and data enhancement. A multi-branch deep neural network structure is adopted for feature extraction, and in combination with a Swin Transformer V2 trunk, a Gated MLP module, cross-modal attention fusion and physical consistency constraint, the adaptability of the model to millimeter wave SAR data and multi-modal input is improved. Meteorological parameters and line operation data are integrated through multi-modal data fusion, and model robustness is enhanced. Finally, an icing thickness prediction value is output through real-time reasoning, intelligent early warning is triggered in combination with confidence analysis, end-to-end real-time response is achieved, the intelligent level of safe operation of a power grid is remarkably improved, and the method has wide application value.

The invention relates to the technical field of artificial intelligence and information propagation analysis, and discloses a deep learning-based network information trend prediction method and system.The deep learning-based network information trend prediction method comprises the steps of performing neural architecture search through combination of an evolutionary algorithm and reinforcement learning; an optimal neural network structure suitable for different types of network information is automatically found; dynamic reconstruction of a model structure is realized through an environment perception and event triggering mechanism; according to a deployment environment resource constraint, adopting an importance-perceived neuron self-adaptive pruning technology; turning point features in network information propagation are specially extracted and enhanced; knowledge migration from a large-scale high-precision model to a lightweight model is realized; an online learning and continuous optimization mechanism is adopted to prevent disastrous forgetting; according to the method, the key turning point of network information propagation can be accurately predicted, the prediction accuracy is improved, the early warning time is shortened, and the computing resource consumption is reduced.

The invention relates to the technical field of industrial robot control, and discloses an industrial robot trajectory optimization control method based on an intelligent algorithm. The method comprises the steps that joint position information, tool center point coordinates and a motion time sequence when a robot executes multiple tasks are collected and stored in a track database; after cleaning and screening data, extracting a feature set containing a path point sequence, speed distribution and an acceleration contour; constructing an intelligent optimization algorithm model of a neural network structure, and training by using the feature set to learn a trajectory optimization strategy; analyzing a target position coordinate and a motion constraint condition of the current task to obtain an initial track parameter; and inputting the initial parameters into the trained model, outputting an optimized track sequence containing a path point list and a speed curve, and generating a control instruction to drive the robot to move. The method can adapt to different tasks, improves track rationality and motion stability, and fits industrial production practice.

The invention relates to the technical field of data analysis and processing, in particular to an identity data analysis method based on a spatial frequency sensing fusion network, and the method specifically comprises the following steps: collecting identity data to be detected, and marking real identity information; preprocessing the collected identity data, and dividing the collected identity data into a training set and a test set in proportion; constructing a spatial frequency sensing multi-scale network for real-time deep identity analysis, and training the network; inputting the identity data in the test set into the trained network for forgery detection and prediction, and outputting a probability score that each identity is true or false so as to obtain an identity analysis result. By constructing a lightweight deep neural network structure which combines space and frequency feature perception and has a multi-scale fusion capability, the method can be used for efficiently detecting identity information in actual scenes such as video conferences and social media.

The invention discloses a microseismic source positioning method, device and system, and a storage medium. The method comprises the following steps: dividing a microseismic data sample data set into a training set and a test set; according to the training set, constructing a microseism inversion subnet containing a Swin Transform encoder, and according to the training set, constructing a microseism inversion subnet containing a Swin Transform encoder; the micro-seismic forward modeling subnet realizes seismic wave field continuation by inputting a micro-seismic source position and a speed model and utilizing a recurrent neural network structure and a convolution operator, and establishes a forward modeling subnet based on a wave equation; constructing an inversion-forward modeling closed-loop neural network according to the micro-seismic inversion subnet and the forward modeling subnet based on the wave equation; and inputting the test set into the inversion-forward closed-loop neural network to carry out micro-seismic source positioning. By adopting the technical scheme of the invention, the limitations on physical constraint, feature modeling and anti-noise capability in the prior art are overcome.

A medical code allocation method based on hierarchical association and diversity enhancement comprises the following steps: S1, data preprocessing and graph construction: based on a tree hierarchical classification architecture of ICD codes, adopting a hierarchical graph model to perform visual representation on the tree hierarchical classification architecture, and constructing a co-occurrence matrix of the clinical record text and the ICD codes for the clinical record text; s2, applying a Graph-BERT graph neural network structure to perform graph structure conversion on a tree structure of an ICD code system, and learning based on an encoder structure of a graph transformer to obtain embedded representation of ICD codes; s3, performing word segmentation processing on a clinical text, inputting the processed clinical text into a bioBERT model to obtain a feature vector of the clinical text, and unifying semantic features of clinical records, hierarchical features of ICD codes and related association information between the semantic features and the hierarchical features through a multi-modal fusion mode; s4, after the final feature vector of the clinical record text is obtained, correlation enhancement prediction is performed by using a CorNet network, so that a related probability matrix is generated, a binary cross entropy loss function is expanded, the binary cross entropy loss function, hierarchical diversity loss and semantic diversity loss jointly form a diversified loss function, and the diversified loss function and a classic model are subjected to comparative analysis, so that the final feature vector of the clinical record text is obtained. Therefore, the accuracy and effectiveness of the method are verified.

The invention relates to a basin flood simulation optimization method based on a physical information neural network, and aims to construct a basin flood acceleration solution mode based on PINNs, and ensure the continuity and the physical characteristics of mass conservation during model operation, and the method comprises the following steps: simulating basin flood by using an urban flood space-time simulation model; obtaining a data set required for constructing the physical neural network; physical rules are integrated and simplified into limiting conditions, so that integration of a neural network structure is facilitated; establishing a partial differential equation (PDE) used for describing river flood routing, and taking the partial differential equation as a loss function of the physical information neural network model; constructing and training a PINNs drainage basin flood model constructed on the basis of a CNN (Convolutional Neural Network); and establishing a model evaluation index and performing evaluation. The method has the beneficial effect that the speed and precision of basin flood simulation are improved.

The invention relates to an underwater target detection method based on a pulse neural network, which can realize underwater target detection with high precision and low power consumption. By fusing the cross-stage partial network and the YOLO architecture, the problem of pulse degradation is effectively solved, and the feature extraction capability of the model is enhanced; in order to solve the problem of underwater noise interference, the pulse-based underwater image denoising method is designed, only integer addition is used in the method, a pulse neural network structure can be seamlessly embedded, and the quality of a feature map is enhanced; in order to solve the problem that a traditional normalization method is low in precision in the spiking neural network, separate batch normalization is provided, by independently normalizing a feature map in multiple time steps and optimizing a residual structure, the time dynamic state of the SNN can be effectively captured, and the detection precision is improved. The network shows excellent performance in underwater target detection, and compared with an artificial neural network of the same scale, the network has higher performance and lower energy consumption.

The invention discloses a low-sample neural network structure reliability evaluation system and evaluation method, and relates to the technical field of engineering structure safety monitoring, and the evaluation system comprises a cloud server which is used for constructing a recurrent neural network model containing a static variable embedding mechanism, completing model training and converting a model format; the edge calculation terminal is used for receiving and preprocessing real-time data of the sensor, executing multi-step prediction to output a future time period response sequence, and calculating a future failure probability through virtual Monte Carlo simulation; the sensor assembly is used for collecting structure state time sequence data; and the communication module is used for realizing data interaction and alarm signal transmission operation. According to the method, collaborative modeling of time-varying and static uncertainty is realized by adopting a static variable embedded recurrent neural network model, failure probability distribution is generated at an edge computing terminal in combination with a virtual Monte Carlo technology, failure risk prediction in a future time period is supported, and real-time and accurate reliability early warning can be realized in a resource limited scene.

The invention provides a method for predicting tunnel excavation water inflow through horizontal drilling, and relates to the technical field of tunnel construction water inflow prediction.The method comprises the steps that water inflow time sequence data and geological parameter data are obtained; calculating the permeability coefficient of the position where each monitoring point is located according to the water inflow time sequence data by using a Goodman formula based on rock integrity correction; decomposing the water inflow time sequence data to obtain a plurality of intrinsic mode functions and a residual term; forming a training data set; constructing a prediction model comprising a GRU neural network structure and a physical constraint mechanism, training the prediction model by using the training data set until the loss function converges, and obtaining a trained prediction model; and obtaining current monitoring data, inputting the current monitoring data into the trained prediction model, and predicting the water inflow and the water inflow risk level of the to-be-excavated section. According to the method, refined and high-precision prediction of the water inflow in the tunnel excavation process can be achieved, and potential safety hazards caused by coarse-grained prediction of a traditional method are avoided.

The invention discloses a livestock and poultry health state intelligent evaluation method based on multi-sensor fusion and an AI prediction model, and belongs to the technical field of livestock breeding, and the method comprises the following steps: setting multiple types of sensors in a livestock and poultry breeding area, collecting multi-source data such as temperature, humidity, carbon dioxide concentration, ammonia gas concentration, illumination intensity, weight and current, and carrying out the analysis of the data; wherein the weight data is obtained through the weighing platform, and the current data is used for sensing the operation state of drinking, ingestion, ventilation or lighting equipment; various types of sensors such as temperature and humidity sensors, carbon dioxide sensors, ammonia gas sensors, illumination sensors, weight sensors and current sensors are deployed in livestock and poultry breeding areas, all-directional data acquisition of environments and behavior states is realized, and dynamic modeling and feature extraction of key behavior nodes are realized by introducing a deep neural network structure of a gating circulation unit and an attention mechanism.

The application discloses a memory reinforcement deep unfolding multi-modal image fusion method with downstream task enhancement, which comprises the following steps: collecting infrared images and visible light images, and dividing training set and test set; establishing an optimization target and solving, to obtain an iterative formula; using a neural network to replace a proximal operator in the iterative formula, to obtain a neural network structure; sending the training set into the neural network to obtain a fusion picture, and calculating a total loss according to the fusion picture, the infrared image and the visible light image; updating parameters of the neural network according to the total loss to obtain an updated neural network; inputting the infrared image and the visible light image of the test set into the updated neural network to obtain a fusion image. The application makes the fused image have characteristics easy to be distinguished by a downstream task network, and can realize the best performance on a data set, while performance and interpretability are taken into account, and the application has rationality and applicability.

The invention discloses a tea drying equipment dynamic control method based on deep reinforcement learning, comprising the following steps: S1, collecting parameters of each control unit of tea drying equipment, and generating a state observation data set; s2, constructing an equipment graph structure, and outputting equipment graph structure information; s3, constructing an action value calculation network containing a dried graph neural network structure based on an improved MADDPG algorithm, and generating a global joint feature embedding set; s4, constructing a strategy network set, and generating a control action set; s5, inputting the global joint feature embedded set and the control action set into an action value prediction module, and outputting an action value prediction result; s6, the improved MADDPG algorithm is used for executing action value loss function calculation and parameter reverse updating; and S7, deploying the trained strategy network set to a tea drying equipment control system. According to the method, the MADDPG and drying graph neural network is improved, and dynamic closed-loop regulation and control of the tea leaf drying equipment are realized.

The invention discloses a power network anomaly detection method and system, and relates to the technical field of intelligent power grid monitoring, and the method comprises the steps: constructing a neuron-like power network graph model, carrying out the node processing of equipment, introducing a multi-factor anomaly score function, and dynamically activating an abnormal node through combining with an improved Winner-Take-All mechanism; the connection weight is dynamically optimized through a topology-aware composite gradient descent method, and a multi-layer neural network structure is constructed to realize cross-layer forward propagation and feedback adjustment; and establishing an abnormal path model in combination with the propagation phase difference and the information entropy, identifying an abnormal path by using a propagation score function and a weighted shortest path algorithm, and carrying out fault tracing. According to the power network anomaly detection method provided by the invention, the neuron-like graph model is constructed, and a multi-factor anomaly activation mechanism is introduced, so that comprehensive perception of a power grid node state under a multi-dimensional time characteristic can be realized, and the anomaly recognition sensitivity and the multi-point simultaneous activation capability are effectively improved.

The invention discloses a topological optimization method based on a physical information neural network (PINN). The topological optimization method comprises the following steps of data preparation, neural network design, loss function definition, dynamic sampling strategy and optimization process implementation. Specifically, the method comprises the following steps: firstly, acquiring the center coordinates of each unit, then inputting the unit coordinates into a sine representation network, outputting a unit density value, calculating displacement through finite element analysis, then calculating a flexibility value according to node displacement so as to obtain loss, and finally, carrying out back propagation on the loss and updating the network weight. The method has the advantages that the deep learning technology is combined with the physical information neural network, so that the structural performance is improved, the dependence on experimental data is reduced, and the optimization efficiency and precision are improved. Particularly, the structure of the neural network is different from that of a conventional neural network, and deep embedding of physical information is considered in the design of the neural network, so that the prediction accuracy and the generalization ability of the network are improved.

The invention discloses a low-voltage power distribution network line loss prediction method, a model training method and a related device, and the model training method comprises the steps: (1) employing a neural network structure combining a time sequence feature extraction module and a multi-scale feature extraction module, and capturing the long-term trend of photovoltaic output and load fluctuation, line loss change modes under different time scales are effectively modeled; (2) a deep feature extraction module introduces a deep feature extraction mechanism with mutually different architectures, and the expression ability and generalization performance of the model to a complex nonlinear relationship are improved; (3) a crown porcupine optimization algorithm is adopted to perform joint optimization on model parameters and structures, efficient search can be realized in a high-dimensional non-convex space, local optimum is avoided, and prediction precision and stability are improved; based on the line loss prediction model, the line loss prediction method provided by the invention can be directly used for power distribution network operation situation evaluation and energy efficiency analysis, and has good engineering application value under the background of wide distributed photovoltaic access.

The invention relates to the field of numerical reservoir simulation, and provides a proxy modeling method for embedding a coarse grid simulator, so as to solve the problems of poor space-time extrapolation performance and unstable long-term prediction. According to the method, a coarse mesh numerical simulator is used as a physical prior module to be embedded into a multi-resolution fusion neural network structure (MNN), and a complex nonlinear relation in the physical prior module is captured by means of a Fourier neural operator (FNO), so that efficient, long-time-sequence and stable prediction of multiple physical fields such as pressure, oil saturation and gas saturation is realized. According to the method, a coarse mesh numerical solution is used as an intermediate drive, multi-scale physical information is fused through a modular MNN network, and the physical consistency and the space-time generalization ability of the proxy model are remarkably improved. In the model, the FNO is utilized to extract global frequency domain characteristics of a coarse grid simulation result field, and through verification of multiple cases, the model shows good space-time extrapolation performance under various coarsening scales and well control conditions. The method has the robustness of numerical simulation and the high efficiency of deep learning.

The invention discloses an adaptive task unloading mechanism driven by meta-reinforcement learning in an edge computing environment. According to the mechanism, a system architecture composed of a user equipment layer and an edge server layer is constructed, and task analysis, state perception and strategy optimization processes are combined to realize unloading scheduling optimization of a multi-task dependent structure. Task unloading is modeled as a Markov decision process, a double-layer training mechanism is adopted, local strategy training is realized by utilizing a near-end strategy optimization algorithm, and the generalization ability of the system is improved in combination with cross-task meta-strategy learning. According to the method, the sequence is fused into the sequence neural network structure and the multi-head attention mechanism, and the accuracy and efficiency of unloading strategy generation are improved while the task dependency relationship is modeled. The mechanism has good task adaptability and delay optimization performance in a dynamic edge computing environment, and is suitable for various mobile computing scenes.

The invention discloses a power system short-term load prediction method and device based on deep learning, a computer and a storage medium, belongs to the technical field of power load prediction, and solves the problem that in the prior art, a power system short-term load prediction model cannot be subjected to collaborative optimization through feature selection and a deep neural network structure. The technical problem that the prediction precision and accuracy cannot be improved due to feature redundancy, error accumulation and insufficient periodic feature learning is difficult to solve. The method comprises the following steps: acquiring power load data of a power system; constructing a comprehensive feature selection method; a double-load prediction model is constructed, the first load prediction model is a three-channel LSTM-CNN-based short-term power load prediction model, and the second prediction model is a double-channel LCLA-based multi-step short-term power load prediction model; and performing short-term load prediction by using the prediction model to obtain a final load prediction value. The method is suitable for short-term load prediction of the power system.

The invention belongs to the technical field of artificial intelligence, particularly relates to the technical field of reinforcement learning, and particularly relates to an unmanned system autonomy evaluation method based on reinforcement learning, which can be applied to intelligent unmanned aerial vehicle manufacturing, and comprises the following steps: constructing an unmanned aerial vehicle autonomy evaluation index system; constructing an unmanned system autonomy scoring model based on reinforcement learning; and training an autonomous scoring model of the unmanned system. According to the method, a three-level index system taking the autonomy of the unmanned aerial vehicle as a root node is established, a double neural network structure based on reinforcement learning is designed, and an autonomy score is calculated. A multi-head attention mechanism is embedded in the value network, and adaptive learning of each leaf index weight is realized. According to the method, the importance weights of different evaluation dimensions are dynamically adjusted, and accurate and adaptive autonomy scores are provided. According to the method, the problems of fixed weight and lack of adaptability in a traditional evaluation method are solved, and the evaluation strategy can be dynamically optimized according to actual task requirements.

The invention provides a microgravity environment flying robot sensing and scene understanding method and system, and belongs to the field of robot intelligent sensing and scene understanding. The problems that in an indoor working scene under the microgravity environment, facility layout is crowded, space illumination is complex, and floating objects are blocked, so that image feature extraction is difficult, the sight line of a sensor is blocked, and the three-dimensional environment modeling precision and the target recognition accuracy are affected are solved. A lightweight convolutional neural network structure is adopted, so that the calculation burden is reduced; rGB images and data of a laser ranging sensor are combined in the aspect of multi-modal data fusion, a multi-modal information fusion module is added in target detection and pose estimation, and the method is suitable for recognition and positioning of a target object needed by high-precision operation; in the aspect of task adaptive semantic segmentation, aiming at a specific task scene in a microgravity environment, performing network training based on a specific data set; and through a transfer learning mode, the model has higher adaptability.

The invention provides a personalized federal learning method and system for multi-device heterogeneous data, and the method comprises the steps: transmitting a neural network structure to all clients, so as to enable all clients to carry out local model training; receiving the trained model parameters of each client, and dividing the trained model parameters of each client into non-BN layer parameters and BN layer parameters; all the non-BN layer parameters are aggregated; calculating distribution similarity among the clients according to all the BN layer parameters to obtain a similarity matrix; clustering the clients according to the similarity matrix by adopting an affinity propagation algorithm to obtain a client group with similar feature distribution; performing intra-group aggregation on the BN layer parameters corresponding to each group of clients; sending the aggregated non-BN layer parameters to all the clients, and sending the aggregated BN layer parameters in the groups to the clients in the corresponding groups; therefore, the training stability and the prediction precision in a multi-device heterogeneous environment are improved.

The invention relates to the technical field of electrolytic aluminum monitoring and control, and provides an automatic control method and system for an aluminum electrolysis cell in the electrolytic aluminum process, and the method comprises the steps: A, collecting the operation parameters of the electrolysis cell in real time through a multi-source sensing array; b, constructing a dynamic material balance-heat balance coupling model, and inputting the operating parameters of the electrolytic cell into the dynamic material balance-heat balance coupling model for outputting alumina blanking rate data; c, according to the aluminum oxide blanking rate data change rate, temperature-polar distance cooperative control of the electrolytic cell is synchronously executed, and the optimal polar distance compensation amount is calculated according to the operation parameters; and D, constructing an electrolytic cell resistance change rate prediction model based on the LSTM neural network, and inputting the aluminum oxide blanking rate data change rate into the LSTM neural network structure model for outputting an aluminum oxide concentration compensation strategy. The change condition of the aluminum electrolysis cell in the aluminum electrolysis process can be rapidly and accurately monitored, and real-time control is provided.

The invention relates to the technical field of audio signal processing and biomedical signal analysis, and still has a further optimized space for the recognition of anti-noise requirements, signal individual differences and complex pathological modes in a noise environment. The invention provides a double-path CNN heart sound classification method based on time-frequency and double-spectrum fusion features, and the method comprises the steps: carrying out the preprocessing of an original heart sound signal of a data set which is classified into a normal heart sound and an abnormal heart sound, and obtaining a to-be-recognized heart sound signal; based on dynamic continuous wavelet transform, adaptively selecting parameters to extract time-frequency characteristics, introducing bispectrum analysis, capturing nonlinear characteristics, generating a dual-channel characteristic pattern, and efficiently storing the dual-channel characteristic pattern in an HDF5 format; and based on a designed double-path convolutional neural network structure, respectively processing the extracted time-frequency and double-spectrum features, performing classification after fusion, and training a model in combination with category weighted loss and an optimization strategy to obtain a heart sound classification result. The heart sound recognition accuracy can be improved.