38 results about "Process output" patented technology

The invention discloses the technical field of constructional engineering and intelligent control, and relates to a concrete hydration heat self-adaptive control method and system based on digital twinning. The method comprises the following steps: S1, constructing a concrete finite element model, simulating a hydration heat process, and outputting simulation data of temperature and temperature stress; s2, based on the analog data, optimizing the arrangement scheme of the cooling water pipe and the sensor through a composite convolutional neural network; s3, arranging and collecting real-time temperature and temperature stress data according to the scheme; s4, in combination with simulation and real-time data, predicting temperature change by adopting a BP neural network and a long-short term memory network, and constructing a dynamic prediction model; s5, based on the prediction result and the real-time data, constructing a WebGL digital twinborn platform, visualizing the data and correcting a prediction model; and S6, dynamically adjusting parameters of the cooling water pipe according to a correction result to realize hydration heat self-adaptive control. The invention further provides a system for realizing the method, the concrete temperature and the temperature stress can be accurately controlled, and the engineering structure safety and the quality control efficiency are improved.

The invention provides an industrial sensor intermittent fault detection method and device and a medium. The method comprises the following steps: acquiring a process input vector of a target industrial process at a current moment and a process output measurement value measured by a sensor to be measured; determining a process output predicted value according to the process input vector and the virtual sensor model; calculating a current residual between the process output measured value and the process output predicted value; if the current residual error exceeds a target early warning threshold value, marking an early warning signal, and generating a first fault mark under the condition that the continuous triggering times of the early warning signal on a time sequence reach a first preset times; based on the residual sequence from the historical starting moment to the current moment, the posterior probability that the sensor to be detected is in a fault state at the current moment is calculated through a hidden Markov model, and a second fault mark is generated under the condition that the posterior probability is larger than or equal to a preset probability threshold value; and if at least one of the first fault mark and the second fault mark is established, determining that the to-be-detected sensor has an intermittent fault.

The invention provides a task flow dynamic optimization method and system based on data feedback, and the method comprises the steps: obtaining process feedback data, determining the process execution time consumption and a processing result through a production monitoring system log, recognizing the current processing capability of a process, and obtaining the to-be-processed workpiece amount of a downstream assembly node and the frequency fluctuation amplitude of an upstream process output workpiece; acquiring the queuing workpiece quantity and the real-time load quantity of each process corresponding to the rearranged process processing execution sequence, analyzing to obtain a process processing capacity matching degree index, and determining a global workpiece circulation stability evaluation result; and obtaining total number change data of workpieces completed in each process from the dynamic optimization implementation scheme, judging the system processing throughput improvement degree, and determining the process load balancing state of workpiece circulation.

The invention provides an industrial large and small model collaboration framework, a reasoning method, a training method, a device and equipment. Relates to the technical field of industrial control. The method comprises the following steps: when to-be-processed time series data output by an industrial system is obtained, performing feature extraction on the time series data through a small model to obtain a first feature matrix of the time series data, and determining a first confidence score of the first feature matrix; when the first confidence score is smaller than a preset threshold value, performing feature extraction on the time sequence data through a large model to obtain a second feature matrix of the time sequence data, and determining a second confidence score of the second feature matrix; when the second confidence score is greater than the first confidence score, taking a reasoning result obtained by the large model based on the second feature matrix as a target reasoning result; and when the second confidence score is smaller than or equal to the first confidence score, determining a target reasoning result through a small model. According to the method, the reliability of the model reasoning result can be improved.

The application relates to the motor winding technology field and provides a motor winding process automatic control method and system.The method comprises the following steps: obtaining motion state information, stress state information and material characteristic information of a flat wire; inputting the motion state information, the stress state information and the material characteristic information into a preset flat wire mechanical simulation model to obtain a predicted torsion angle of the flat wire in a bending process output by the preset flat wire mechanical simulation model; obtaining an actual torsion angle of the flat wire; adjusting a parameter value of an equivalent shear modulus of the flat wire in the preset flat wire mechanical simulation model according to the predicted torsion angle and the actual torsion angle to obtain an adjusted preset flat wire mechanical simulation model; and performing winding control on the flat wire based on the adjusted preset flat wire mechanical simulation model. The accuracy of motor winding can be controlled.

The invention relates to the technical field of industrial control, in particular to a self-adaptive cutting control method based on a continuous feeding scene of a cutting machine, which comprises the following steps: acquiring process input parameters of a material sample of the cutting machine, process constraint parameters under different processing tracks and corresponding actual process output parameters; respectively constructing a first training data set and a second training data set, and training to obtain a constraint process model and a state process model; a process input parameter flow of a to-be-processed material area is obtained online and input to the constraint process model, and a dynamic process constraint is generated; creating a candidate control trajectory based on the dynamic process constraint; inputting the candidate control trajectory into the state process model, outputting a prediction performance index, and selecting and determining an optimal control trajectory; and executing the optimal control track to realize self-adaptive cutting control, and carrying out online updating on the constraint process model and the state process model to realize optimal control on the machining quality and efficiency of the cutting machine.

The present application relates to the technical field of industrial process dynamic characteristic monitoring, and discloses a kind of injection molding process real-time monitoring method based on multi-sensor data fusion, comprising: firstly, the dynamic response characteristics of drive source itself are identified and compensated to obtain corrected process output signal;Further, process input signal and the corrected process output signal are cooperatively processed to generate process gain and time constant capable of overall representing process dynamic transfer characteristics, and process stability is judged accordingly, the present application separates the disturbance source of drive system performance fluctuation from the monitoring signal, realizes the direct quantitative characterization of the dynamic transfer process connecting input and output itself, converts a complex industrial process stability problem into the monitoring of process gain and time constant two-dimensional coordinate point, and improves the reliability of monitoring.

The present invention discloses a model predictive control method based on a dynamic output feedback strategy, which is applied to polyhedral uncertain systems. When physical devices in the polyhedral uncertain system are operating, data is transmitted through sensors. The model predictive control process is as follows: a dynamic output feedback strategy including a controller and an observer is established. Based on the model predictive control problem, the polyhedral uncertain system, the controller, and the observer are described using state equations. The observer is used to constrain the data transmitted by the sensor and process measurement wild values in the data. The controller is used to perform model predictive control on the processed output data to obtain the control input at the next moment, and the control input is applied to the physical device through an actuator. The present invention is targeted at polyhedral uncertain systems with measurement wild values. The study uses a saturation function to study an observer-based dynamic output feedback strategy, mitigate the impact of measurement wild values, and ensure the robustness and asymptotic stability of the system.

The invention discloses a chemical process fault diagnosis method. A special calculation model matched with an actual device is constructed and corrected, and a baseline data set is generated; generating a simulation data set containing normal and various fault working conditions by using the model; on the basis of the data set, a reverse diagnosis reasoning model with process output deviation as input and fault types as output is established; the model is applied to factory real-time data, and automatic online diagnosis is achieved. According to the method, the diagnosis problem caused by sparse fault samples in a traditional diagnosis method is solved, and rapid and accurate fault positioning of complex processes such as double-effect rectification is realized through combination of mechanism simulation and data driving.

A method includes receiving one or more process inputs and one or more process outputs associated with one or more operations of one or more components of an industrial automation system. The method also includes determining a model representative of the one or more components based on the one or more process inputs and the one or more process outputs; identifying a boundary response time threshold associated with an unmeasured operating parameter of the one or more components based on the one or more process outputs; determining one or more inputs to modify the one or more operations of the one or more components based on the boundary response time threshold; and receiving one or more measured outputs of the one or more components after providing the one or more inputs to the model, the one or more measured outputs are acquired before the boundary response time threshold.

The invention discloses a control method of wire harness production equipment, and relates to the technical field of intelligent manufacturing. The method comprises the following steps: acquiring preset process parameter curves of all N procedures in a complete wire harness production process; inputting all the process parameter curves and the estimated environment parameters into a virtual production system integrated with a defect-process correlation model; executing continuous production simulation from the first process to the last process in the virtual system, and generating virtual product quality evaluation based on the output of the last process; and according to an evaluation result, if the evaluation result is qualified, issuing all the process parameter curves to physical equipment to execute production, and if the evaluation result is not qualified, adjusting the process parameter curves and re-simulating until the evaluation result is qualified. According to the invention, through full-process virtual simulation and beforehand quality prediction, conversion from post-process detection to beforehand prevention is realized, the quality and efficiency of wire harness production are effectively improved, and meanwhile, the production trial and error cost is remarkably reduced through a closed-loop optimization mechanism.

The present application relates to the field of data processing technology, and specifically to a feature extraction model training method, device, and computer equipment; by training the feature extraction model, the key features in the electrolytic aluminum process are acquired and the operating status is characterized based on the key features. The specific process first uses the hidden layer of the denoising autoencoder as the input layer, and configures a new hidden layer and output layer after the input layer to form a three-layer BP artificial neural network, and then adds the loss function obtained from the process output layer to the loss function of the denoising autoencoder reconstruction layer; finally, the model structure is trained with the constructed distinguishing composite error function as the target criterion, which is expected to extract more robust feature information.

A model predictive control (MPC) device includes an input interface configured to receive an industrial process input associated with at least one component of a process automation plant; an output interface configured to transmit control instructions to control the component; a memory configured to store a first MPC process model and a second MPC process model corresponding to different states; and a processor configured to: identify a current state parameter of the industrial process; and predicting a future industrial process output using the first MPC process model or the second MPC process model based on the current state parameter being associated with the first MPC process model or the second MPC process model. The processor is configured to calculate a target operating point from a predicted future industrial process output, determine a control signal to drive the industrial process to the calculated target operating point, and output the determined control signal to control operation of the components of the industrial process plant.

To provide, with respect to predictive modeling, a method, system, and storage medium for predicting optimal input parameters for a manufacturing process using a set of models.SOLUTION: A method in a system architecture includes: receiving, by a processing device, expected output data for a manufacturing process defining attributes of an output of the manufacturing process; accessing a plurality of machine learning models that model the manufacturing process; determining, using a first machine learning model, input data for the manufacturing process including a value for the first input and a value for the second input, based on expected output data for the manufacturing process; and combining the input data determined using the first machine learning model and the input data determined using the second machine learning model to create a set of inputs for the manufacturing process including candidate values for the first input and candidate values for the second input.SELECTED DRAWING: Figure 4

A grinding system includes a pressure system, a substrate carrier including a film, a first sensor, and a control system. The first compartment of the membrane is fluidly coupled to a pressure system. The first sensor is configured to monitor the pressure system and generate a first output based on a condition detected in the pressure system. A control system is coupled to the first sensor and configured to process the first output to produce a first processed output, the control system configured to compare the first processed output to a threshold to detect the presence of fluid in the pressure system.

The invention relates to the field of furniture manufacturing, in particular to a rapid modeling system for digital pattern making of a sofa soft bed. The system comprises a modeling platform module, a three-dimensional modeling and two-dimensional expansion module, a component library and material database, a structure disassembly and consumable statistics module and a process output module. Parametric modeling is carried out on parts such as a seat bag, a backrest, armrests and a bottom frame, a three-dimensional structure is automatically generated, a complex curved surface is unfolded through an energy optimization algorithm, and leather cloth version data are output in combination with splicing optimization. Calling the standard component library by the system, matching material parameters, calculating the consumption of consumables and a loss coefficient, and generating a consumable list; and forming a structure decomposition graph based on the assembly relationship, automatically adding a process label, and finally outputting a digital process document which can be directly used for production. According to the invention, full automation of the plate making process is realized, the development period is shortened, the cost is reduced, and the consistency and production efficiency are improved.

The application provides a production cycle determination method, device, equipment, medium and program product, and relates to the oil and gas well mining process technical field.The method comprises the following steps: obtaining target production time sequence data of a plunger gas lift process; the target production time sequence data comprises tubing pressure data and casing pressure data corresponding to a plurality of time points respectively; inputting the target production time sequence data into a recognition model, recognizing the target production time sequence data through the recognition model, and obtaining a running cycle recognition result of the plunger gas lift process output by the recognition model; and determining a production cycle corresponding to the plunger gas lift process based on the running cycle recognition result.The production cycle determination method, device, equipment, medium and program product provided by the application improve the recognition efficiency and recognition accuracy of the production cycle of the plunger gas lift process, reduce the labor cost, and improve the automation level and production efficiency of the plunger gas lift process.

To configure the control agent (POL), pre-given training data is read in, specifying the state dataset (S), action dataset (A), and the obtained performance values ​​(R) of the technical system (TS). Based on this training data, a data-driven dynamic model (NN) is trained to reproduce the obtained performance values ​​(R) based on the state dataset (S) and action dataset (A). Furthermore, a action evaluation process (VAE) is trained to reproduce the action dataset (A) based on the state dataset (S) and action dataset (A) after information reduction, where reproduction errors (DR, D0, D1) are determined. To train the control agent (POL), training data is fed to the trained dynamic model (NN), the trained action evaluation process (VAE), and the control agent (POL). Here, the performance values ​​(R1, R2) output by the trained dynamic model (NN) are fed into a pre-given performance function (P). Additionally, the reproduction errors (D0, D1) output by the trained action evaluation process (VAE) are fed into the performance function (P) as parameters affecting performance reduction. In this way, the control agent (POL) is trained to output an action dataset (A) that optimizes the performance function (P) based on the state dataset (S).

Machine learning systems and methods for extreme weather event modeling using generative diffusion models are provided. The system includes a weather modeling processor and a weather modeling engine executed by the processor. The weather modeling engine causes the processor to: receive a dataset including a plurality of vorticity samples; process the dataset using a deterministic mean model having a temporal attention unit to model spatial, cross-channel, and temporal dependencies using dynamical attention units; and process output of the deterministic mean model using a reverse diffusion model to capture stochastic fine scale features and to generate a denoised output. A downscaling pipeline can also be executed by the weather modeling engine to downscale outputs of the system.

Systems and methods for input evaluations through model inversion are described herein. In certain embodiments, a system includes a memory configured to store a model of a physical process, wherein the model receives input values and provide output values, wherein the input values represent potential input parameters for the physical process and the output values represent potential measures of process outputs. The system also includes an interface capable of receiving desired output values for the physical process. Further, the system includes processors executing computer executable instructions associated with an application that cause the processors to receive the desired output values; test a plurality of values for at least one of the input values; identify a combination of the input values that is associated with the desired output values; and provide the combination for output through the interface.

The present invention provides an auxiliary recommendation system, method, device and medium for earth observation mission processes, including a data preparation unit, a process link generation module, a process node parameter generation module, a recommended process output unit and a manual interpretation unit; the process link generation module includes a data preprocessing unit, a process mining unit and a process pruning unit; the process node parameter generation module includes a data preprocessing unit, a knowledge reasoning unit and an entity linking unit. The present invention separates the recommendation of process links from the recommendation of process node parameters, simplifies the process of earth observation mission process recommendation; and can mine process models from historical data of earth observation missions to achieve autonomous updating of process models; it can also obtain a potential list of available parameters in real time, and provide alternative solutions for process node parameter conflicts, failures and other problems.

The invention relates to a system for modelling the dependency of process output variables on process variables of a process in an automation system, comprising: a server platform (SP) with a computer unit (CE) and a control unit (SE); first measuring devices (M1) for detecting measured values of the process variables (W1); second measuring devices (M2) for detecting measured values of the process output variables (W2); a closed communication network (KN) for connecting the measuring devices (M1, M2) and server platform (SP); wherein the first measuring devices (M1) detect measured values of the process variables (W1) and transmit them as input values (I) to the server platform (SP) by means of the closed communication network (KN); wherein the second measuring devices (M2) detect measured values of the process output variables (W2) and transmit them as output values (O) to the server platform (SP) by means of the closed communication network (KN); wherein the server platform (SP) stores transmitted input values (I) and output values (O) at least partially as time series (ZR); wherein the server platform (SP) trains an algorithm (ML) with stored time series (ZR) by means of computer unit (CE); wherein the server platform (SP) calculates values (WO) for the process output variables by means of the trained algorithm (ML) on the basis of the transmitted input values (I).

Machine learning systems and methods for extreme weather event modeling using generative diffusion models are provided. The system includes a weather modeling processor and a weather modeling engine executed by the processor. The weather modeling engine causes the processor to: receive a dataset including a plurality of vorticity samples; process the dataset using a deterministic mean model having a temporal attention unit to model spatial, cross-channel, and temporal dependencies using dynamical attention units; and process output of the deterministic mean model using a reverse diffusion model to capture stochastic fine scale features and to generate a denoised output. A downscaling pipeline can also be executed by the weather modeling engine to downscale outputs of the system.

Embodiments of this present disclosure include a non-transitory computer readable medium storing instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations including determining a predicted value for a process output representative of controlled operation of one or more industrial automation devices within an industrial automation system, receiving an actual value for the process output from the industrial automation devices, and determining a deviation between the predicted value and the actual value for the process output. Additionally, the operations include generating a first portion of a reference trajectory based on a closed-loop response performance to a change in a setpoint, generating a second portion of the reference trajectory based on a calculated effect from a disturbance variable on the actual value for the process output, and generating a third portion of the reference trajectory based on an estimation of an unmeasured disturbance variable.

Computing an input relevance measure of process inputs by obtaining from a sensor array sensor measurements that relate process inputs to a process output, computing from the sensor measurements, a number of data partitions, building for each data partition of the number of data partitions a corresponding stochastic gradient boosting model, and computing for each process input, a number of partial dependency plots with each partial dependency plot being based on the corresponding stochastic gradient boosting model. The input relevance measure is then computed for each process input based on the number of partial dependency plots of the process input to estimate a degree of change in the process output obtained by varying the process input.

A model predictive control (MPC) device includes an input interface configured to receive an industrial process input associated with at least one component of a process automation plant, an output interface configured to transmit a control instruction to control the component, memory configured to store first and second MPC process models corresponding to different states, and a processor configured to identify a current state parameter of an industrial process, and predict a future industrial process output using the first or second MPC process model, based on the current state parameter being associated with the first or second MPC process model. The processor is configured to calculate a target operating point according to the predicted future industrial process output, determine a control signal to drive the industrial process to the calculated target operating point, and output the determined control signal to control operation of the component of the industrial process plant.

The present invention provides a non-extracting rotor detection control method and system, which relates to the technical field of generator detection, and the method includes: obtaining the control instructions of the robot in real time, and at the same time, obtaining the motion data of the robot in real time through a preset sensor, and outputting the initial data; parsing the initial data to obtain the parsed data, and performing data analysis on the parsed data to obtain the target current data corresponding to the electromagnet set in the robot; based on the target current data, and in combination with the acquired time constant, determining the control data of the electromagnet, and outputting the electromagnet control data; based on the electromagnet control data, controlling the operating parameters of the electromagnet, synchronously outputting the control process data, evaluating the effect of the control process, outputting the control evaluation result, and optimizing the control process. The present invention can monitor the operating state of the robot in the generator in real time, and improve the operating efficiency and operating accuracy of the robot in the generator, so as to realize the precise control of the robot.

The invention provides a troubleshooting process determination method and device, equipment and a storage medium, and the method comprises the steps: obtaining production problem information; matching is carried out based on the production problem information and a knowledge graph, service system information is determined, and the service system information comprises service process information and system architecture information; the knowledge graph comprises a corresponding relation between the production problem information and the business system information; and inputting the business system information into the troubleshooting process determination model to obtain a target troubleshooting process output by the troubleshooting process determination model. By means of the method, when a production problem occurs, the troubleshooting process can be automatically generated, the efficiency and accuracy of determining the troubleshooting process are improved, the manual analysis and judgment time is shortened, and the efficiency and accuracy of troubleshooting are improved.

This invention discloses a method and system for designing robust operating conditions for industrial production processes. Considering various uncertainties in industrial production, and addressing scenarios where process mechanism models are unknown or simulation costs are high, this method utilizes only the input and output data of the production process. By constructing a robust Bayesian optimization model, it can determine nominal operating condition setpoints with strong disturbance resistance for the production process. The implementation process is as follows: sampling and testing within the operable range of the operational variables to obtain process output data; defining constraints, constructing the robust operating condition design problem as an optimization proposition that minimizes the maximum constraint value; establishing a Gaussian process regression surrogate model based on the sampled data, and constructing a sampling function that integrates the predicted mean, uncertainty, and constraint satisfaction probability; employing a two-stage Bayesian optimization framework, iteratively sampling and updating the model until convergence is achieved to obtain the robust operating condition setpoints.

System for modeling the dependence of process output variables on process variables of a process in an automation plant, comprising: server platform (SP) with computer unit (CE) and control unit (SE); first measuring devices (M1) for acquiring measured values ​​of the process variables (W1); second measuring devices (M2) for acquiring measured values ​​of the process output variables (W2); closed communication network (KN) for connecting measuring devices (M1, M2) and server platform (SP); wherein first measuring devices (M1) acquire measured values ​​of the process variables (W1) and transmit them as input values ​​(I) to the server platform (SP) via a closed communication network (KN); wherein second measuring devices (M2) acquire measured values ​​of the process output variables (W2) and transmit them as output values ​​(O) to the server platform (SP) via a closed communication network (KN);wherein the server platform (SP) stores transmitted input values ​​(I) and output values ​​(O) at least partially as time series (ZR); wherein the server platform (SP) trains an algorithm (ML) with stored time series (ZR) using a computer unit (CE); wherein the server platform (SP) uses the trained algorithm (ML) to calculate values ​​(WO) for the process output variables based on the transmitted input values ​​(I).