35 results about "Soft sensing" patented technology

The invention relates to the field of soft measurement, discloses a soft measurement method based on an optimization algorithm, and solves the industrial pain points of fast precision attenuation, poor mechanism adaptation and high maintenance cost of a traditional soft measurement model. Comprising the steps of obtaining historical production data and a configuration file, and processing and integrating the historical production data and the configuration file into a complemented data file; determining a candidate state variable set according to the complemented data file, and searching an optimal independent variable set through optimization algorithms such as a genetic algorithm; training a data driving model by using the optimal independent variable set to generate a soft measurement model; an unmeasurable input variable is used as a genetic algorithm optimization target, a fitness function is constructed through an average absolute error of a predicted value and an actual value, and an optimal solution is output through iterative evolution. By comprehensively applying various optimization algorithms, the precision and the reliability of the soft measurement model are effectively improved, powerful support is provided for real-time optimization of an industrial control system, and the production efficiency and the product quality can be improved.

The present application relates to the technical field of thermal power plants, and discloses an intelligent operation method and system for a thermal power plant, a device, and a storage medium. The method comprises: measuring operation parameters of a thermal power plant in real time by means of advanced detection and soft sensing techniques; establishing a fault monitoring model to monitor the operation process of the thermal power plant on the basis of the real-time detected operation parameters, and generating an early warning signal when a fault occurs; performing fault diagnosis, and generating a fault self-healing instruction; and evaluating the operation status of the thermal power plant by means of a performance evaluation model, and then obtaining optimal operation modes and optimal operation parameter setpoints by means of a multi-objective optimization algorithm. The present application achieves high-performance real-time intelligent closed-loop control, improving the adaptability to complex working conditions and the disturbance rejection capability of a thermal power plant production process control system, and the overall performance of a thermal power plant production process; and efficient human-computer interaction is achieved by means of artificial intelligence-based early warning diagnosis and closed-loop self-healing.

The invention relates to the technical field of particle size measurement in the ore grinding production process. The ore grinding granularity soft measurement method comprises the steps of analyzing a time sequence characteristic matrix based on a step-by-step regularization characteristic sorting method, forming an optimized input characteristic matrix, obtaining a linear prediction value based on the optimized input characteristic matrix, and calculating the ore grinding granularity according to the linear prediction value. Performing error calculation on the linear predicted value and the actual measurement value to obtain a nonlinear predicted value, combining the linear predicted value and the nonlinear predicted value according to the weight to generate a preliminary soft measurement model, performing global optimization on the combined weight of the preliminary soft measurement model through a multi-objective optimization algorithm to generate an optimized soft measurement model, and performing soft measurement on the optimal soft measurement model. When the statistical property exceeds a set threshold value, a dynamic correction mechanism is triggered, and an updated soft measurement model is obtained according to the dynamic correction mechanism. The method has the effect of meeting the high-precision prediction requirement under the dynamic working condition.

The invention discloses a coal-fired unit hearth temperature soft measurement method based on data fusion. The method comprises the following steps: acquiring a hearth combustion image and operation data in a coal-fired unit distributed control system (DCS) at the same moment; preprocessing the combustion image; establishing a feature extraction network mixed with deconvolution and interpolation upsampling; adding the structural similarity index loss into a network training loss function; training the feature extraction network to obtain a feature extraction model; extracting an M * N two-dimensional feature map obtained by the last layer of encoder; dividing the two-dimensional feature map into S same areas, and adding feature values in the single areas to serve as representative feature values of the areas; dCS operation parameter characteristics are analyzed and selected, and Q characteristics with the highest correlation degree with the hearth temperature are selected; performing data fusion on the S combustion image features and the Q DCS operation parameters; and establishing a bidirectional long-short-term memory neural network prediction model to predict the hearth temperature. According to the invention, the precision and robustness of the hearth temperature soft measurement model can be improved.

The invention relates to an industrial process soft measurement method, in particular to an industrial process soft measurement method based on a hierarchical space-time enhanced quality related network. Performing normalization processing on the original data; a multi-head hierarchical time enhancement module (MHTE) is designed, local time characteristics are extracted through cross convolution, global time characteristic extraction is guided, and local and global characteristics are fused; a stacked encoder framework is constructed; introducing a mixing quality regularization mechanism combining a maximum information coefficient (MIC) and a Jaccard similarity coefficient (JSC); and optimizing the model through layer-by-layer pre-training and regression fine tuning. And the trained soft measurement model is utilized to carry out online quality prediction, and real-time monitoring and production strategy adjustment are carried out, so that the product quality and the production efficiency are improved, and green production is promoted. Simulation experiments prove that the method can efficiently extract dynamic spatio-temporal characteristics, and has high prediction precision and stability.

The invention relates to an industrial process soft measurement method, in particular to a soft measurement method of a space-time fused hierarchical enhancement denoising auto-encoder network, which is familiar with an actual industrial process and determines a process variable and a quality variable; industrial data are collected and divided; performing data preprocessing; noise is added into the processed data, the processed data is input into an enhanced de-noising auto-encoder network (HEDAE) to learn effective information, and global enhanced features are extracted; and inputting global enhancement features extracted by the HEDAE to the GRU and the CNN at the same time, designing an interactive gating mechanism (TSIF), fusing time and space features, constructing a hybrid model to realize quality variable prediction, and verifying model prediction performance according to an evaluation index. On-line prediction is carried out through a soft measurement model, an industrial product quality variable prediction value is obtained, a production operation process is monitored and controlled in real time, and a production strategy is adjusted to ensure production efficiency, improve product quality and realize green production.

The invention discloses a deep extension VAE soft measurement method based on instant learning, and belongs to the technical field of industrial process control. In the modeling process, DE-VAE is formed by stacking multiple layers of E-VAE, the output of the E-VAE comprises reconstructed original input data, expected output data and hidden variables of the previous E-VAE, correlation between extracted features and the original input and output is enhanced by multiplexing prior information, meanwhile, quality variables are introduced in the pre-training process, and the quality of the original input and output is improved. And the extraction capability of the model on target related features is enhanced. In the online prediction stage, the model is updated by using instant learning, a novel modeling method different from the existing soft measurement modeling method is designed, and the prediction precision of the soft measurement model is well improved.

The invention discloses a crystal size distribution soft measurement method integrated with mechanism constraints. The method comprises the following steps: 1) data acquisition and integration; 2) modeling training and model testing: constructing a crystal size distribution soft measurement model integrated with mechanism constraint, training a training data set, injecting crystallization process dynamics knowledge into the Kolmogorov-Arnod network in a loss function form, improving the nonlinear modeling capability and credibility of the model, and performing model testing on the Kolmogorov-Arnod network; and verifying test set data by using the trained model. According to the method, the knowledge of the crystallization dynamic process is injected into the Kolmogorov-Arnod network in the form of the loss function, so that the prediction capability of the model on the crystal size distribution in the crystallization process and the credibility of the model are effectively improved.

The application provides a process industry soft measurement method based on feature extraction and incremental learning. The method introduces an end-to-end neural network soft measurement model composed of multiple parallel time series feature extraction self-encoder networks and a regressor network connected in front and back, extracts time series features from multi-process variable time series data, and uses the time series features for performance index soft measurement modeling. In addition, based on the time series features extracted by the model, the method proposes an incremental learning method based on time series features. The soft measurement method provided by the application can effectively and accurately perform soft measurement on some performance indexes that are difficult to directly measure or have high measurement cost in the process industry, and through the incremental learning method, the soft measurement model can long-term maintain good soft measurement performance in the industrial data stream with strong real-time performance, thereby providing reference and guidance for monitoring, measurement, optimization and control of various parameters in the process industry production process.

The invention discloses an industrial process soft measurement method based on graph attention, and belongs to the field of industrial process soft measurement. The method comprises the following steps: firstly, determining a secondary variable and a quality variable, taking a sampling point of each batch of experiment as a graph structure node, encoding different batches, and performing joint encoding with a production batch dimension as a heterogeneous node; in the same batch, bidirectional edges with different weights are established according to the current sampling time and the next sampling time, different batches are connected through cosine similarity, three cross-batch nodes with the highest similarity with each node are taken as edges, and the weights are taken as similarity values; and secondly, constructing a hierarchical attention mechanism, establishing node-level attention and batch-level attention, and completing attention fusion through residual connection. And the prediction precision of the quality variables among different batches is effectively improved.

The present invention discloses a carbon dioxide concentration prediction method based on a semi-supervised deep probability model for on-line detection of the carbon dioxide content in a carbon dioxide absorption tower. Based on the conventional probabilistic principal component regression model, the present invention simultaneously introduces the ideas of semi-supervised learning and deep learning, expands the basic probabilistic principal component regression model into the structure of a semi-supervised deep learning model, that is, proposes a new soft sensing method based on a semi-supervised deep probabilistic principal component regression model for on-line detection of the carbon dioxide concentration in a carbon dioxide absorption tower. Compared with the conventional principal component regression model, the method of the present invention can not only effectively utilize a large amount of cheap unlabeled sample information, but also deeply extract the information hidden in the process data, thereby improving the actual prediction effect of the carbon dioxide concentration.

A fermentation process soft-sensing method based on a topologically guided temporal convolutional network belongs to the technical field of fermentation process soft-sensing. The method includes the following steps: 1) Data acquisition and integration: Using a simulation platform to acquire penicillin fermentation processes under different operating conditions, the acquired data is divided, collected, and integrated; 2) Data selection: Selecting data, removing redundant and useless data, and establishing a causal graph between variables; 3) Constructing and training a TGTCN model; and 4) Model prediction. The method adopts a fermentation process soft-sensing method based on a topologically guided temporal convolutional network. It utilizes a graph attention network and a temporal convolutional network to extract data in both temporal and spatial dimensions, increasing the generalizability of the model and enabling accurate measurement of key product quality across different fermentation processes.

The invention discloses an industrial soft measurement method of sparse ONLSTM based on non-negative notching, and belongs to the field of soft measurement modeling and application in the modern industrial process. According to the method, the problems of modeling difficulty increase and model performance reduction caused by nonlinearity, dynamics and variable redundancy of a complex industrial process are considered, an NNG algorithm is embedded into an input layer weight matrix and a hidden layer weight matrix of an ONLSTM network, and a DNNG-ONLSTM neural network prediction model is constructed to realize prediction of key quality variables of the complex industrial process; on the premise of ensuring the prediction capability of the model, redundant input data is eliminated. And meanwhile, an NNG algorithm is added to a hidden layer, so that the calculation amount of the model is reduced, redundant nodes of the network model are eliminated, the sparsification of the model is realized, the complexity and training difficulty of the model are reduced, and the generalization performance of the model is improved.

The invention discloses a rock-fill dam concrete panel strain soft sensing method and system in the technical field of structural safety monitoring, and the method comprises the steps: obtaining the monitoring data of a rock-fill deformation sensor, inputting the monitoring data into a pre-constructed mapping model, and calculating the monitoring data of a concrete panel strain virtual sensor; construction of a mapping model: establishing a finite element simulation model according to an actual rock-fill dam project; simulating conditions of rock-fill deformation and concrete panel strain under an environmental load by using a finite element simulation model to obtain simulation result data; the method comprises the following steps of: establishing a mapping relation between rockfill deformation and concrete panel strain through actually acquired rockfill deformation monitoring data, concrete panel strain monitoring data and simulation result data; and constructing a final mapping model based on the mapping relation. The rock-fill dam concrete panel strain soft sensing method can provide more comprehensive panel strain virtual monitoring data under the condition of low monitoring cost.

The invention discloses an anaerobic fermentation soft measurement modeling method based on EWOA-LSSVM. The method is used for accurately predicting the methane concentration which is difficult to directly measure in the anaerobic fermentation process. Comprising the following steps: 1) performing normalization processing after data acquisition, and eliminating the influence of units and dimensions of different variables; 2) initializing the population number, the inertia weight and the number of iterations of the whale optimization algorithm (EWOA); 3) establishing an EWOA-LSSVM soft measurement model for an anaerobic fermentation process, and optimizing parameters of the LSSVM model by using a whale optimization algorithm; and 4) substituting the obtained parameter optimal value into the LSSVM, extracting the characteristics of the anaerobic fermentation process data by using the LSSVM, and carrying out modeling. The modeling method provided by the invention has the advantages of good estimation performance, high accuracy and strong generalization ability, and is also suitable for soft measurement modeling in other complex chemical reaction processes.

This invention relates to the field of online measurement technology, and more particularly to a method and system for soft measurement modeling of industrial processes for incomplete data. The steps include: acquiring historical sample data containing missing values ​​in the industrial process and constructing a BGPLVM model; extracting sample features based on the historical sample data and using them as prior constraints to initialize the hyperparameters of the BGPLVM model; acquiring query sample data and obtaining the latent variable variational distribution of the query sample data based on the BGPLVM model; extracting historical sample data most similar to the latent variable variational distribution to construct a local sample dataset; constructing a Gaussian process regression model and determining the parameters of the Gaussian process regression model by maximizing the log-likelihood; and substituting the latent variable variational distribution of the query sample data into the Gaussian process regression model to predict the output variable corresponding to the query sample data.

The present invention proposes a soft measurement method for effluent ammonia nitrogen based on ADw-CLPSO radial basis emotion neural network, which belongs to the field of sewage treatment. The main design process of this method is as follows: first, a nonlinear function is constructed to improve the CLPSO algorithm, so that the inertia weight of each particle can be adaptively and dynamically adjusted, and then based on the improved ADw-CLPSO algorithm, a particle variable dimension learning mechanism is used to simultaneously realize parameter updating and structural adjustment in the radial basis emotion network training process. The soft measurement method for effluent ammonia nitrogen based on ADw-CLPSO radial basis emotion neural network composed of the above steps belongs to the protection scope of the present invention. The present invention uses the ADw-CLPSO algorithm to simultaneously optimize the parameters and structure of the radial basis emotion network, which can improve the prediction accuracy of effluent ammonia nitrogen and improve the generalization performance of the model.

The present invention discloses a soft measurement method, system, electronic device and medium for sewage water quality indicators, which relates to the technical field of industrial process control. The method includes: obtaining two initial regression models according to two labeled data subsets by using the PLS algorithm and the ELM algorithm; obtaining two labeled prediction error values according to the two models and the two labeled data subsets; obtaining two Plus labeled data subsets according to the two models and the unlabeled data set; obtaining two labeled prediction error values according to the two initial regression models and the two Plus labeled data subsets; obtaining a confidence level according to the four error values; obtaining two new labeled data subsets according to the confidence level; and obtaining a final prediction model according to the two new labeled data subsets by using RPLS and RELM. The present invention can improve the measurement accuracy and overcome the problem of prediction model degradation.

The invention discloses a filter cake moisture soft measurement method based on a double-regularization width learning system, and the method comprises the steps: obtaining key process parameter data in a filter cake forming process, and taking the key process parameter data as input characteristics; preprocessing the key process parameter data, removing the influence of an abnormal value by using an IQR quartile distance method, removing the influence of a missing value by using a dropna function, and removing the influence among different dimensions by using Z-score standardization to obtain a robust data set; training a double-regularization width learning system model by using the robust data set, performing constraint optimization on feature nodes of the double-regularization width learning system by combining a gradient descent algorithm and a cross validation method, and determining an optimal regularization parameter to obtain a prediction model; and inputting to-be-measured data into the prediction model to obtain a predicted value of the water content of the filter cake. According to the method, the moisture prediction precision can be improved, the calculation complexity is reduced, and rapid and stable online measurement is realized.

The invention discloses a catalytic cracking unit raw material property soft measurement modeling method based on an integrated learning algorithm, and relates to the technical field of flow industrial process soft measurement. On the basis of a Stacking ensemble learning algorithm, a distillation curve model and a data driving model are connected in series, the soft measurement modeling method for the properties of the catalytic cracking raw oil is provided, and the method has the advantages that the distillation range point, the specific gravity, the carbon residue value, the PONA value and other properties of raw materials of a catalytic cracking device can be comprehensively and accurately predicted in real time; the defect that an existing soft measurement model is single in raw material property prediction result is overcome, and the requirement for industrial optimization control can be better met. The embodiment shows that compared with a traditional machine learning method, the soft measurement model has the advantages that the prediction result is more accurate, the average absolute percentage error of each distillation range point is smaller than 2%, and the soft measurement model can accurately and comprehensively predict the properties of the raw materials of the catalytic cracking device in real time.

The present invention relates to a method for filling missing data for industrial soft sensing, including determining auxiliary variables of a target variable, generating a first predicted value according to the time correlation of the existing data in the auxiliary variables, and filling the missing data in the auxiliary variables by using the first predicted value; and predicting the target variable based on the spatial dependence relationship between the filled auxiliary variables and the target variable. This application can use the existing data set to predict and fill missing values, ensure the integrity and accuracy of the data while guaranteeing the data generation rate and precision, and at the same time can meet the requirements of high-precision downstream soft sensing and realize customized missing data generation tasks.

The invention discloses an asphalt mixing station drying roller energy-saving temperature control method and related equipment, and the method comprises the steps: firstly building a soft sensing model for predicting the discharge temperature based on the historical operation data of a drying roller, covering the main variable of the discharge temperature, the opening degree of a gas valve and other auxiliary variables; and the model is used as an interaction environment for reinforcement learning agent training, and the agent is trained by taking stable control of the discharging temperature and minimization of the gas usage amount as targets. The trained intelligent agent can output actions according to the current state; and finally, the intelligent agent is deployed to a drying roller control system, the state is collected and input in real time, and the intelligent agent outputs actions to adjust a gas valve and an air blower in real time. The invention aims to realize real-time accurate control of the heating process of the drying roller, stabilize the aggregate discharge temperature, ensure the quality of the finished product material and effectively control the energy consumption.

The present invention provides a soft measurement method for the CO content of a blast furnace top based on collaborative analysis of multi-perspective data, and relates to the technical field of blast furnace ironmaking process. First, image data and related physical variables of the blast furnace tuyere and furnace top are collected, and image matrices and physical variable matrices of the tuyere and furnace top are obtained through grayscale processing and normalization processing; then, collaborative correlation analysis is performed on the above data to obtain a correlation information index matrix for prediction; then, the index matrix is ​​further processed to further process information for calculating and removing associated data; a prediction model for the CO content of the blast furnace top is established using the information after removing associated data, and parameters for predicting the CO content of the blast furnace top are obtained; finally, after grayscale processing and removing associated data information on the data acquired in real time, the CO content of the blast furnace top is predicted according to the parameters of the blast furnace top CO content prediction model. The present invention can improve the accuracy of CO measurement of the blast furnace top and reduce its measurement difficulty and time lag.

The invention discloses an equipment fault diagnosis system based on a large model, and relates to the technical field of industrial equipment.The system constructs a soft sensing sign vector set only by relying on existing electrical and process signals under the cooperation of a data access module and a preprocessing module, and through the secondary characterization and normalization of the soft sensing sign vector set, the fault diagnosis accuracy is improved. And explainable physical sign energy E is synthesized, so that amplification characterization of slow abnormality of 0.2-3Hz is realized. The robust comparison of a same-group set G is introduced by means of a software feature comparison initial evaluation module, a comparison software feature anomaly score Csoft is output, an early quantitative prompt of relative anomaly can be given before hard threshold alarm, and the installation difference, the seasonal difference and the calibration difference are naturally counteracted. Therefore, on the premise that no hardware is newly added, early recognition of slight looseness, alignment deviation and the like of the coupler is achieved, and non-planned shutdown and invalid overhaul caused by misjudgment are effectively reduced.

Fouling is formation of deposits on the heat exchanger surfaces that adversely affects operation of heat exchanger. Fouling can be approximated through a set of estimated heat exchanger parameters, which may not be accurate, leading to uncertainty in operation / maintenance decisions and hence the losses. A system and a method for identification and forecasting fouling of a plurality of heat exchangers in a refinery has been provided. The system comprises a digital replica of the heat exchanger network. The digital replica is configured to receive real-time sensor data from a plurality of data sources and provides real-time soft sensing of key parameters. The system is also configured to diagnose the reasons behind a specific condition of fouling. Further, an advisory is provided, that alerts and recommends corrective actions. The system provides estimate for the remaining useful life (RUL) of the heat exchangers and suggests the cleaning schedule.

Pulverizers are very critical equipment in overall functioning of a plant. They need to be controlled and monitored properly for the optimized operation of the pulverizers. A system and method for performance and health monitoring to optimize operation of a pulverizer is provided. The system comprises a digital twin that can mimic the performance of the pulverizer in real-time and assist the operators in decision making related to operation, maintenance and scheduling. The digital twin is configured to receives real-time sensor data from a plurality of data sources and provides real-time soft sensing of key health and performance parameters of the pulverizer. One more key aspect of the solution is the advisory system that alerts and recommends corrective actions in terms of parameters controlled through other equipment or changes in operation or design or changes in cleaning schedule.

The soft measurement method for nitrogen oxides in the urban solid waste incineration process based on a modular neural network belongs to the field of solid waste treatment. The real-time detection of the toxic gas - nitrogen oxides (NOx) generated in MSWI can effectively control the emission of NOx. In industrial sites, a high-precision instrument - the continuous emission monitoring system for flue gas is used to detect the NOx concentration in flue gas emissions. The measurement results are greatly affected by the environment, and the equipment maintenance cost is high. First of all, the fuzzy c-means algorithm is used for task decomposition, decomposing the task into different subtasks; secondly, for different subtasks, a radial basis function neural network is used to design soft measurement sub-models respectively, establishing the non-linear relationship between the characteristic variables and NOx; finally, the output of the sub-networks is integrated through a cascaded neural network. The effectiveness of the proposed method is verified by benchmark experiments and the actual data of a certain MSWI plant.

This invention discloses a method for monitoring and evaluating shipborne wastewater treatment based on high-resolution full-spectrum soft sensing, relating to the fields of water quality monitoring and marine environmental protection. It aims to address the technical problems of existing shipborne water quality monitoring systems, which rely on hardware sensors, have weak anti-interference capabilities, poor model generalization, and cannot accurately analyze the status of each stage of the wastewater treatment system in real time, resulting in an incomplete evaluation system. This method uses high-resolution full-spectrum data as a foundation, integrates soft sensing technology, and constructs a water quality parameter prediction model based on machine learning and self-learning to achieve real-time and accurate measurement of key water quality parameters of shipborne wastewater. Simultaneously, it establishes a comprehensive evaluation index system for the shipborne wastewater treatment system, combining monitored water quality parameters and operational data from each stage to complete status analysis and fault prediction for core stages such as sedimentation, oxidation, and biological treatment.

The embodiment of the invention discloses a method, device, equipment and system for determining the parameter relation of a control valve and a medium. The method comprises the following steps: determining an actual value of a mechanical structure parameter and an actual value of a controlled pressure difference parameter of the pressure-independent control valve; the actual value of the mechanical structure parameter and the actual value of the controlled pressure difference parameter are input into a calibration model, and the calibration model is adapted to determine the incidence relation between the flow parameter and the pressure difference parameter of the pressure-independent control valve based on the mechanical structure parameter and the controlled pressure difference parameter; an association relationship between the flow parameter and the differential pressure parameter is received from the calibration model. The incidence relation between the flow parameter and the pressure difference parameter of the pressure-independent control valve is determined through various models such as an artificial intelligence model, soft sensing and accurate control of flow can be achieved without installing a flow meter, and the equipment cost and the installation expenditure are reduced. Moreover, lightweight processing is performed on the traffic model, so that field deployment and application are facilitated.

The invention discloses an industrial soft measurement method based on a variable rate complementary cycle unit network, and belongs to the technical field of soft measurement. The method comprises the following steps: designing a rate self-adaptive regulation gate in a GRU architecture, performing nonlinear exponential normalization transformation on the rate self-adaptive regulation gate and an update gate to form a variable rate forgetting gate and a variable rate input gate, embedding the variable rate forgetting gate and the variable rate input gate into a GRU hidden state update function, constructing an SVCRU neural network prediction model, and predicting the hidden state of the GRU. According to the model, the inherent linear constant rate regulation constraint of the GRU is replaced by a nonlinear variable rate regulation mechanism, so that the model training convergence speed is increased, and the prediction precision is improved.