54 results about "Multivariable predictive control" patented technology

An apparatus and method is disclosed for detecting, identifying, and classifying faults occurring in sensors measuring a process. A variety of process models can be used such as first principles models, dynamic multivariable predictive control models, from data using statistical methods such as partial least squares (PLS) or principal component analysis. If faults are identified in one or more sensors, the apparatus and method provide replacement values for the faulty sensors so that any process controllers and process monitoring systems that use these sensors can remain in operation during the fault period. The identification of faulty sensors is achieved through the use of a set of structured residual transforms that are uniquely designed to be insensitive to specific subsets of sensors, while being maximally sensitive to sensors not in the subset. Identified faults are classified into one of the types Complete Failure, Bias, Drift, Precision Loss, or Unknown.

An integrated multivariable predictive controller (MPC) and tester is disclosed. The invention system provides optimal control and step testing of a multivariable dynamic process using a small amplitude step for model identification purposes, without moving too far from optimal control targets. A tunable parameter specifies the trade-off between optimal process operation and minimum movement of process variables, establishing a middle ground between running a MPC on the Minimum Cost setting and the Minimum Move setting. Exploiting this middle ground, embodiments carry out low amplitude step testing near the optimal steady state solution, such that the data is suitable for modeling purposes. The new system decides when the MPC should run in optimization mode and when it can run in constrained step testing mode. The invention system determines when and how big the superimposed step testing signals can be, such that the temporary optimization give-away is constrained to an acceptable range.

The invention discloses a reheated gas temperature optimization control method based on multi-variable predictive control. According to the method, the whole hot steam temperature system is regarded as a two-in one-out multi-variable object, and the flow rate of reheated de-superheating jet water and the opening of a flue gas baffle plate are simultaneously controlled by adopting a multi-variable predictive control method; moreover, optimization of economical efficiency of the system is realized by adding a steady state target value of the opening of a reheated de-superheating jet water adjusting valve into optimization indexes of the conventional predictive control. By adopting the multi-variable predictive control method, coordination of side adjustment of flue gas and side adjustment of steam can be better realized, and the dynamic adjusting quality of the reheated steam temperature is further improved; and meanwhile, by adding the steady state target value of the de-superheating water adjusting valve into the conventional predictive control performance indexes, optimization of the de-superheating water jet quantity is realized, so that the circulating efficiency of a unit can be effectively improved.

The invention discloses a method for model mismatching detection and positioning of a multivariate predictive control system in a chemical process. By aiming at the typical hierarchical control system in the traditional process industry, the model mismatching detection and positioning is carried out respectively on a multivariate system and all univariate subsystems in the multivariate system. The method comprises the steps of: acquiring residual errors of the multivariate system in a predicative control layer on the basis of an orthogonal projection method, constructing a monitoring statistic computing engine, and acquiring the residual errors, sensitive to process dynamic characteristic variation only, of the univariate system on the basis of an auxiliary variable identification method.In the invention, closed-loop operation data is adopted directly and additional information of the system is not needed; rapid detection and positioning can be made just by utilizing a small amount of data so that fewer disturbances are made to a working condition; and the mismatching information of process models of all univariate subsystems in the multivariate system can be positioned so as to eliminate the influence of the disturbances. By means of the method disclosed by the invention, the accurate model mismatching information can be obtained rapidly at lower cost, the maintenance cost of the predictive control system can be reduced, the service life of the predicative control system can be prolonged and the gain can be improved.

A system and method for determining and controlling for cure status of binder on a fibrous product are disclosed. Cure status is monitored by measuring one or more control variables and attempting to keep them within known control limits. Exemplary control variables include oven temperatures at various locations and color values of sections of the fibrous product. Sensors such as thermocouples and image capture systems sense these variables continuously online and provide input signals for a MPC processor-optimizer. The MPC optimizers balances the constraints according to a programmed optimization function and priority ranking of control variables and solves for optimal control setting on manipulatable variables, such as oven fan speed, oven setpoint temperatures and coolant water flow rate.

This invention relates to a method for controlling the temperature of continuous annealing heating oven. The entrance temperature of the oven, belt steel flow rate of a second and oven temperature are used as the input to calculate the optimal gas flow increase, and the oven temperature can be comtrolled by adjusting the gas flow increase. This invention considers the influences of belt steel thickness, width, flow rate and the entrance temperature to the oven temperature, and increases the accuracy and dynamic responsivity of the oven temperature control.

The invention discloses a multi-variable interval constrained estimation-based SCR denitration optimized control system, which comprises an NOx concentration target value setting unit, a multi-variable predictive control unit, a proportional control unit A, an ammonia spraying valve A, an SCR reactor A, a transition flue between the reactor A to a chimney inlet, a proportional control unit B, an ammonia spraying valve B, an SCR reactor B, a flue between the reactor B and the chimney inlet, and the chimney inlet. The invention also discloses a control method of the above system. According to the technical scheme of the invention, the response speed of the ammonia spraying amount control to the unit load change is increased. Meanwhile, the influence on the performances of the system, caused by the saturation of an actuator, can be avoided. The actuating number of the actuator is reduced, and the service life thereof is prolonged. The economical efficiency of the denitration system is improved.

This invention relates to one dye machine module formula multi-variable prediction robust bar H infinity control method, which determines one dye machine mode formula multiple step prediction robust H infinity control method through data collection, process resolution, prediction control, robust H infinity control technique; using the control method to reduce uncertain factors color impact.

The invention discloses a method for realizing reheat steam temperature multivariable coordinate predictive control through performance indexes. According to the method, the reheat steam temperature system of a thermal power generating unit is regarded as a 2*1 multivariable system, a multivariable predictive controller is designed on the basis of novel performance indexes, and the opening degree of a reheat steam temperature reducing water-spray valve and the opening degree of a smoke baffle are controlled simultaneously. In the performance indexes, on one hand, an algorithm is simplified and rapid control is realized on the basis of requirements on a single-step prediction error and a prediction error change rate, and the control performance of a control system can also be effectively adjusted by adjusting a predictive length and a predictive error change rate weight; on the other hand, dynamic limit on the opening degree of the temperature reducing water-spray valve is added in the indexes, the part of the performance parameters is self-adaptive to changes during a control process, and the amount of temperature reducing water sprayed in reheat steam can be reduced while a reheat steam control quality is improved, thus effectively increasing the cycle efficiency of the unit.

The invention relates to an air-cooled condenser unit with optimized operation, comprising: an air-cooled condenser, which has a heat exchange tube bundle and fan equipment, the heat exchange tube bundle is configured to receive exhaust steam from a steam turbine, and the fan equipment is matched with the heat exchange tube bundle arranged to cool the exhaust steam flowing in the heat exchange tube bank into condensed water; a data measurement and acquisition system configured to measure and acquire at least part of the controlled variables and disturbance variables of the air-cooled condenser; a multivariable predictive controller, It is based on pre-set modular units, and automatically controls the control variables of the fan equipment according to the values ​​of at least part of the controlled variables and disturbance variables from the data measurement and acquisition system to optimize the performance of the air-cooled condenser while preventing the back pressure of the steam turbine from being lower than the set value The range of pressure protection limits and prevent the condensate temperature from falling below the predetermined limit of condensate temperature.

The invention discloses a deep diagnostic method of performance reduction of a predictive control model. The deep diagnostic method of the performance reduction of the predictive control model includes: using a production run data calculation process disturbance signal, and then using production data and a step response coefficient of a design stage of a predicting controller to calculate model prediction error, and judging overall performance advantages and disadvantages of the predictive control model through a model performance index built through the production data and the step response coefficient of the design stage of the predicting controller. The deep diagnostic method of the performance reduction of the predictive control model further includes: for the model deteriorative in performance, using a method of removing input variables one by one to calculate a new model performance index, judging sub model performance corresponding the removed input variables through model performance index change situations, and thereby achieving monitoring for the performance of each sub module. The deep diagnostic method of the performance reduction of the predictive control model only uses the production process data and design data, and thereby not only can evaluate the overall performance of the multivariable predictive control model, but also can further evaluate the performance of all the sub modules corresponding to the input variables, provides a suggestion for an engineer to maintain a control system, and can substantially reduce maintenance cost of the predictive controller.

For the characteristics that a middle-discharging type cement raw material mill system is provided with two control loops of a coarse powder bin and a fine powder bin, the coupling action exists between the coarse powder bin and the fine powder bin and the load of the two bins needs to be coordinated and optimized, the invention adopts a double-layer-structure predication control method with two layers of optimization and control to carry out independent optimization and control on a control system. A steady optimization layer realizes the load optimization on the coarse powder bin and the fine powder bin according to the economic optimization index; a dynamic control layer selects milling noise and the negative pressure of an inlet of the fine powder bin as output variables, selects the feeding amount and the openness of a distributing valve as input variables, and selects the rotating speed of a powder selecting machine as a feed-forward variable so as to form a two-input two-output one-feed-forward multi-variable predication control system. The double-layer-structure predication control method has the advantages that the total load level of a raw material mill is effectively improved, the mutual coupling action when the coarse powder bin and the fine powder bin are respectively controlled is avoided, the influence of lagging of the feeding amount is reduced, the disturbance action of quality adjustment on a product on the control system is inhibited, the automation degree is improved, the yield of raw materials is increased, the continuous operation rate of the control system is increased and the economic benefit is increased for enterprises.

A cryogenic air separation system that optimizes the recovery of argon includes an air intake, a high-pressure distillation column, a low-pressure distillation column, a crude argon distillation column, and a controller that automatically controls the composition of the raw argon stream to decrease the concentration of oxygen in the stream while preventing the concentration of nitrogen in the crude argon stream from exceeding a selected value. In addition, the controller also controls the composition of the crude argon stream until an oxygen concentration of the crude argon stream reaches a selected value. The controller may include a multivariable predictive controller.

An SCR denitration device ammonia injection flow adjusting method based on model prediction comprises: firstly, training and obtaining an industrial furnace combustion prediction model and a multivariable prediction control model, then obtaining a predicted value of the nitrogen oxide concentration of an inlet of an SCR reactor through prediction, and then determining the initial ammonia injectionamount; then, according to the corrected prediction value of the concentration of the nitrogen oxide at the inlet of the SCR reactor, multivariable parameters of the SCR reactor and the multivariableprediction control model, predicting the concentration of nitric oxide at the outlet of the SCR reactor, constructing a performance index function according to the predicted value, the measured value, the set value and the ammonia spraying amount of the concentration of nitric oxide at the outlet of the SCR reactor, and then conducting preliminary optimization of the ammonia spraying amount; andfinally, updating the actual ammonia injection amount of the SCR reactor by using the preliminary optimization result of the ammonia injection amount, taking the actual ammonia injection amount as theoptimal position initial value of the next PSO algorithm, predicting the concentration of nitrogen oxide at the outlet of the SCR reactor, returning to the step 7, continuing to optimize the performance index by using the PSO algorithm, and repeating the iteration in this way.

The invention relates to a split-ranging control realization method for multi-variable model predictive control. The method comprises: carrying out off-line initialization on a multi-input and multi-output control system; and carrying out on-line split-ranging controlling on the system based on a model prediction algorithm. Prediction models are established according to different action combinations of all execution mechanisms in all linear intervals in split-ranging control; during each control period, a prediction model is selected by comparing an output control signal value of a controller with a corresponding intermediate threshold; and a control increment is calculated by using a dynamic matrix control algorithm and the calculated control increment is applied to the corresponding execution mechanism for action implementation. Therefore, a design problem of a multi-variable model predictive controller for a split-ranging control need during a control process can be solved; and thus the application range of the prediction controller can be extended.

The invention relates to a quick multivariable predication control method for an ethylene cracking furnace. The method has an offline link and an online link. In the offline link, a dynamic mathematic model and a static mathematic model of a production process of the ethylene cracking furnace are obtained through an analysis or recognition method and control parameters are set. In the online link, a prediction value is calculated and output according to a measurement value of an ethylene cracking furnace control system (DCS), a predication error is calculated based on the difference between the predication value and the measurement value and accordingly a static output predication value is corrected; and then a static control input predication value is calculated through static optimization or a direct inversion method; and finally a control input value of a current moment is calculated according to a given control parameter, the dynamic performance of the control input value is corrected through a lead-lag link and the corrected value is then transmitted to the DCS. According to the invention, the calculation complexity for multivariable prediction control is simplified and the control period of multivariable prediction control is reduced distinctively, and control problems in quick reaction processes of the ethylene cracking furnace and the like can be solved effectively.

The invention relates to a method for the continuous preparation of acetic acid and/or methyl acetate by carbonylation using carbon monoxide, methanol or a methanol derivative that can be carbonylated, in liquid phase, in the presence of water and a catalytic system. According to the invention, in order to improve the production of acetic acid and/or methyl acetate, the temperature of the reactor and the inlet flow of methanol, or the methanol derivative that can be carbonylated, into said reactor are controlled by the inlet flow of carbon monoxide and by at least one of the parameters that define the composition of the reaction medium and/or events, said automatic control being performed preferably by means of a predictive multi-variable regulator. The inventive method is suitable for use in Monsanto-type preparation methods and low water content methods.

Process control system for detecting abnormal events in a process having one or more independent variables and one or more dependent variables. The system includes a device for measuring values of the one or more independent and dependent variables, a process controller having a predictive model for calculating predicted values of the one or more dependent variables from the measured values of the one or more independent variables, a calculator for calculating residual values for the one or more dependent variables from the difference between the predicted and measured values of the one or more dependent variables, and an analyzer for performing a principal component analysis on the residual values. The process controller is a multivariable predictive control means and the principal component analysis results in the output of one or more scores values, T2 values and Q values.

The present invention describes a control system for units for the production of 1-butene by the oligomerization of ethylene in the presence of a homogeneous catalyst, in the liquid phase and at the bubble point, which uses a multivariable predictive controller and provides the units with better stability.

A method is presented for adjusting the steady-state gains of a multivariable predictive control, planning or optimization model with uncertainty. The user selects a desired matrix relative gain criteria for the predictive model or sub-model. This is used to calculate a base number. Model gains are extracted from the predictive model and the magnitudes are modified to be rounded number powers of the calculated base number.