Search

CN-116186993-B - Flotation optimization control method based on online quality monitoring

CN116186993BCN 116186993 BCN116186993 BCN 116186993BCN-116186993-B

Abstract

The invention relates to a flotation optimization control method based on online quality monitoring, which comprises the steps of collecting data collected by a plurality of links in the whole flotation process, processing the collected data to obtain a data set, extracting causal characteristics of the data set by adopting a convergence cross mapping method to obtain a dynamic relation model of key operation variables and production index variables, designing an optimization target based on Lyapunovo Barrier functions to realize single-step optimization control, searching for optimal actions by adopting a limited time domain optimization strategy with forward rolling time to realize a cyclic forward optimization control task, and repeating the steps 5, 6 and 7 to realize online optimization control of the flotation process. The invention has the advantages of effectively avoiding the dislocation of a fixed global optimization target and actual production, reasonably optimizing target variables, and having better effects of reducing the tail floating grade and stabilizing the floating quality level.

Inventors

  • YANG ZHENYU
  • YUAN QINGBO
  • ZHANG WENHUI
  • SUN BIN
  • BAI YUANSHENG
  • ZHANG GUOLIANG
  • HU JIAN
  • FAN LIPENG

Assignees

  • 鞍钢集团矿业有限公司

Dates

Publication Date
20260512
Application Date
20221220

Claims (9)

  1. 1. The flotation optimization control method based on-line quality monitoring is characterized by comprising the following steps of: step 1, collecting data collected by a plurality of links in the whole flotation process in real time Mainly comprises real-time monitoring data collected by a sensor Image characteristic data extracted by foam analyzer Control feedback value data Adding the data into the subsequent data modeling analysis; Step2, real-time monitoring data Control feedback data Respectively carrying out normalization processing to obtain dimensionless real-time monitoring data And dimensionless control feedback data For image characteristic data Dummy variable processing is carried out to obtain pseudo-coded image characteristic data ; Step 3, adopting Kalman filtering to monitor dimensionless real-time data Smoothing to obtain smoothed dimensionless real-time monitoring data Real-time monitoring data without dimension after smoothing Dimensionless control feedback data And pseudo-encoded image characteristic data Combining to obtain a data set ; Step 4, extracting the data set by adopting a convergence cross mapping method Is a causal feature of (a) to the manipulated variable Sensor monitoring variable And foam state variable Is analyzed for causality of key operation variable set Confirming production index variable according to actual demand Last set data set Removing key operating variables And production index variable Other than variables being other state variables ; Step 5, using key operation variables And other state variables As input, training the expansion time sequence convolution network for production index variable Predicting to obtain key operation variables And production index variable Is a dynamic relationship model of (1); step 6, designing an optimization target for realizing single-step optimization control based on Lyapunovo Barrier functions; step 7, searching for an optimal action by adopting a time forward rolling type finite time domain optimization strategy, and realizing a circular forward optimization control task; And 8, repeating the steps 5, 6 and 7, and optimally controlling the flotation process on line.
  2. 2. The method for optimizing control of flotation based on-line quality monitoring according to claim 1, wherein the real-time monitoring data Includes the underflow speed and air flow rate, liquid level, temperature and PH value of each pump pool, and image characteristic data extracted by foam analyzer RGB values including the image's own point location, foam size and color identified from the image, control feedback data Including cone valve opening, dosage type and quantity, and dosing pump frequency.
  3. 3. The flotation optimization control method based on-line quality monitoring according to claim 1, wherein the collected real-time monitoring data And control feedback data Normalizing by Min-Max normalization method to obtain mapping range of [ -1,1], and processing the image characteristic data Processing by means of dummy variables, converting into m numerical characteristics with 0,1 values, wherein the formula of the Min-Max standardization method is as follows: In the middle of For real-time monitoring data collected And control feedback data Is defined as the entire data for each of the feature columns, And Is respectively to The maximum and minimum values are found.
  4. 4. The method for optimizing and controlling flotation based on-line quality monitoring according to claim 1, wherein the step 3 uses Kalman filtering to monitor dimensionless real-time data Smoothing is performed, wherein the Kalman filter time update formula is as follows: the Kalman filter state update formula is as follows: In the middle of And The a posteriori state estimates at times k-1 and k respectively, Is the prior state estimation value at the moment k, and is also the output of Kalman filtering update, and is dimensionless real-time monitoring data after smoothing ; And The a posteriori estimated covariance at k-1 and k times respectively, Is the prior estimated covariance at time k, H is the state variable to measured conversion matrix; Is the filtered input, here dimensionless real-time monitoring data ; Is the filter gain matrix, a is the state transition matrix, Q is the process excitation noise covariance, also the covariance of the system process, R is the measurement noise covariance, B is the matrix that converts the input into state, Is the residual of the actual observation and the prior state estimation.
  5. 5. The flotation optimization control method based on-line quality monitoring according to claim 1, wherein the step 4 of screening key operation variables comprises the following steps of; step 4-1, determining production index variable according to actual production And based on production index variables And a data set Constructing a dataset at time t Simultaneously constructing the operation variables Sensor monitoring variable And foam state variable Excluding production index variables therein And data set Forming a t-time dataset And then based on And Construction And Corresponding shadow manifold 、 The formula is as follows: Wherein E is the optimal embedding dimension; step 4-2, calculating the distance between any two points of the shadow manifold of X and the distance between any two points of the shadow manifold of Y by adopting Euclidean distance, and performing the calculation on the shadow manifold And Finding E+1 neighbor nodes from each point in the manifold Obtained by cross mapping The cross-map formula is as follows: Wherein the method comprises the steps of On a representative manifold And (3) with Euclidean distance between them; step 4-3, calculating And (3) with The correlation coefficient r of (2) is as follows: as the length L of the input data sequence increases, Gradually converge to I.e. the correlation coefficient r converges to a value greater than 0, then the decision is made from To the point of And judging the causal relation between the operation variable and the monitoring variable according to the value of the correlation coefficient r, and further mining the monitoring variable and the control priority thereof which need to be regulated to enable the monitoring variable to reach the target value.
  6. 6. The flotation optimization control method based on-line quality monitoring according to claim 5, wherein the searching process of the optimal embedding dimension E comprises the following steps: the optimal embedding dimension is judged by using a red pool information criterion, wherein the red pool information criterion is based on an entropy concept and is used for balancing the complexity of the process of the step 4 and the superiority of fitting data, the complexity of the process is defined as n, a loss function V is defined, the number of samples is K, and the red pool information criterion is defined as follows: By passing through The autoregressive model of (a) to determine the optimal embedding dimension is represented by the following formula: Loss function Model complexity n=e, when And taking the minimum value, wherein the corresponding n is the optimal embedding dimension E.
  7. 7. The method for optimizing control of flotation based on-line quality monitoring according to claim 1, wherein in step 5, the key operation variables are used With other state variables As input, training the expansion time sequence convolution network for production index variable The training expansion time sequence convolution network predicts output and comprises the following steps: Step 5-1, setting the number of convolution kernels, the size of the convolution kernels and step length parameters, setting an activation function of a convolution layer as relu functions, and establishing a unidirectional causal convolution network of M layers; step 5-2, applying an expansion convolution, the convolution kernel being in step size Skipping part of input, wherein f is an expansion factor, c is an expansion coefficient, and l is a layer where a convolution kernel is located; Step 5-3, adding the jump layer connection sum of residual convolution One residual block contains two layers of expanded convolution and relu nonlinear mapping, and WeightNorm and Dropout are added after each hole convolution to realize regularization.
  8. 8. The method for optimizing and controlling flotation based on-line quality monitoring according to claim 1, wherein the optimization objective based on Lyapunovo Barrier functions is designed in the step 6, The CLBF model based on Lyapunovo Barrier functions is determined by introducing Lyapunovo Barrier functions on the basis of model predictive control strategies, and the optimization problem based on the CLBF-MPC design is as follows: Wherein the method comprises the steps of Is a set of piecewise constant functions of predicted state trajectories, S+father, with a period of father, To predict the number of sampling periods in the range, Is used to represent Cost function Satisfy the following requirements , , So as to obtain the minimum value of the cost function under the condition of stable system, thereby Is that the system optimization function is in the prediction range The optimal solution above includes Lyapunovo Barrier function conditions in addition to the constraints described above.
  9. 9. The method for optimizing control of flotation based on-line quality monitoring according to claim 1, wherein the step 7 searches for an optimal action by using a time-forward rolling finite time domain optimization strategy; the rolling optimization is carried out by adopting iLQR algorithm, compared with LQR algorithm, the iLQR has nonlinear dynamic function, so that the approximate expansion of Taylor is needed to be carried out on the local part of the function on the basis of LQR algorithm, and the local dynamic characteristic of a complex nonlinear function is estimated through Taylor display And cost function The specific formula is as follows: Wherein, the The key operation variables screened in the step 4 are the state parameters at the current moment, The key operation variables screened in step 4 are referred to herein as current time control parameters, Respectively an actual sampling value at the current moment and a state parameter estimated value at the current moment And control parameter estimation value Is the difference between (1); Dynamic function according to correlation definition The method comprises the following steps: Cost function The method comprises the following steps: Obtaining a current time control law by LQR reverse pushing And control constant Performing the following cyclic actions until convergence; 。

Description

Flotation optimization control method based on online quality monitoring Technical Field The invention belongs to the technical field of intelligent control of mineral separation processes, and particularly relates to an operation optimization control method for a flotation process. Background In recent years, with the development of science and technology, the demand for mineral resources has been increasing. Enterprises have to improve the economic benefit by ensuring the product quality, improving the production efficiency and reducing the production cost. Froth flotation is a highly efficient mineral separation technique that is commonly used in the modern mineral separation industry. The traditional flotation process control is usually realized by on-site technicians adjusting parameters such as the dosage, the aeration quantity, the cone valve opening degree of a flotation machine and the like according to the past production experience. The control mode through subjective judgment has strong randomness, and the on-site process environment is bad, the process fluctuation is frequent, so that the product quality is poor and the stability is poor. Since froth flotation is a complex process with multiple inputs and outputs and coupling and is affected by numerous parameters, the current methods have difficulty in establishing an accurate process model for the flotation process, and thus the automated control of the froth flotation process is difficult to implement. The current automatic control research of flotation is mainly focused on the research of single links or single variable control strategies, but aiming at the flotation process with long production period, high internal coupling degree and multiple parameters, a plurality of data source data in the whole flotation process are required to be fully added into data modeling analysis, and a full-process operation optimization control model is established. Disclosure of Invention The invention aims to provide a flotation optimization control method based on online quality monitoring, which is characterized in that real-time flow data acquired by a sensor and an image characteristic value extracted by a foam analyzer are comprehensively modeled, a reliable predictive control model is established by considering causal characteristics among variables, and based on real-time prediction of monitoring variables and indexes, the operation variables such as the dosage, the aeration quantity, the cone valve opening degree and the like in the flotation process are optimally controlled. The invention discloses a flotation optimization control method based on online quality monitoring, which is characterized by comprising the following steps of: step 1, collecting data collected by a plurality of links in the whole flotation process in real time Mainly comprises real-time monitoring data collected by a sensorImage characteristic data extracted by foam analyzerControl feedback dataAdding the data into the subsequent data modeling analysis; Step2, real-time monitoring data Control feedback dataRespectively carrying out normalization processing to obtain dimensionless real-time monitoring dataAnd dimensionless control feedback dataFor image characteristic dataDummy variable processing is carried out to obtain pseudo-coded image characteristic data; Step 3, adopting Kalman filtering to monitor dimensionless real-time dataSmoothing to obtain smoothed dimensionless real-time monitoring dataReal-time monitoring data without dimension after smoothingDimensionless control feedback dataPseudo-encoded image characteristic dataCombining to obtain a data set; Step 4, extracting the data set by adopting a convergence cross mapping methodIs a causal feature of (a) to the manipulated variableSensor monitoring variableAnd foam state variableIs analyzed for causality of key operation variable setConfirming production index variable according to actual demandLast set data setRemoving key operating variablesAnd production index variableOther than variables being other state variables; Step 5, using key operation variablesAnd other state variablesAs input, build expansion time sequence convolution network, for production index variablePredicting to obtain key operation variablesAnd production index variableIs a dynamic relationship model of (1); step 6, designing an optimization target for realizing single-step optimization control based on Lyapunovo Barrier functions; step 7, searching for an optimal action by adopting a time forward rolling type finite time domain optimization strategy, and realizing a circular forward optimization control task; And 8, repeating the steps 5, 6 and 7, and optimally controlling the flotation process on line. Preferably, said real-time monitoring dataComprises underflow speed and air flow rate of each flow path, liquid level of each pump pool, temperature and PH value, and image characteristic value extracted by a foam analyzerRGB values including