Cement production control system and method for predicting disturbance compensation based on time sequence data
By introducing a multivariate time series data prediction and disturbance compensation mechanism in cement production control, dynamically predicting and compensating measurable disturbances, the problem of MPC-dependent model accuracy is solved, the control accuracy and adaptability are improved, and better control effects are achieved and Productivity.
Patent Information
- Application Number
- CN202411371886.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-29
- Publication Date
- 2025-01-10
AI Technical Summary
In the existing cement production control technology, model prediction control (MPC) depends on the accuracy of the system model, and the measured disturbance changes in traditional algorithms cannot be effectively considered, resulting in limited control effects.
Adopt the disturbance compensation mechanism based on multivariate timing data prediction, and by establishing a timing data prediction model to dynamically predict and compensate measurable disturbances, the prediction accuracy and control adaptability are improved, and the control effect is optimized.
It improves the model prediction accuracy, enhances the robustness and stability of the control system, reduces control errors, achieves better control effects and production efficiency, and reduces energy consumption and environmental pollution.
Smart Images
Figure CN119292056A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of automatic control, and in particular, relates to a cement production control method based on time series data prediction disturbance compensation. Background Art
[0002] In the existing cement production control technology, experience-based control strategies such as PID control are usually used. Although these methods are simple and easy to use, they are often difficult to adapt to the complex changes and uncertainties in the production process. In recent years, with the development of computer technology, model predictive control (MPC) technology has received widespread attention because it can handle complex problems such as multivariables, nonlinearity and constraints.
[0003] MPC is an advanced control strategy that builds a mathematical model of the process, predicts the process behavior over a period of time in the future, and calculates control inputs based on these predictions. MPC can take into account future constraints, such as equipment limitations, safety thresholds, etc., to achieve better control results.
[0004] Although the model predictive control (MPC) algorithm has been applied to cement production to a certain extent, the existing technology still has some limitations, which are mainly reflected in the following aspects:
[0005] 1. Strong dependence on model accuracy: The effect of MPC depends largely on the accuracy of the system model. If the model cannot accurately reflect the dynamic characteristics of the actual system, the control effect will be greatly reduced.
[0006] 2. In the traditional model predictive control algorithm, the measurable disturbance uses the initial input value in all prediction steps. However, in the actual system, the measurable disturbance may change, which may cause a large deviation between the predicted model state and the actual model, thereby reducing the control effect of the controller. Summary of the invention
[0007] The present invention aims to overcome the shortcomings of the prior art and proposes a cement production control system and method based on time series data prediction disturbance compensation to achieve the following objectives: by introducing a disturbance compensation mechanism based on multivariable time series data prediction, measurable disturbances can be dynamically predicted and compensated, thereby improving prediction accuracy, enhancing control adaptability, reducing control errors, optimizing control effects, and improving the stability and robustness of the cement production control system.
[0008] In order to achieve the above object, the technical solution adopted by the present invention is:
[0009] A cement production control system based on time series data prediction disturbance compensation, the control system includes a control target setting module, an MPC controller, a disturbance prediction module, a cement production system, and a feedback module, wherein the control target setting module is used to set an ideal control target for the cement production system; the disturbance prediction module is used to predict measurable disturbances and output them to the MPC controller; the MPC controller is used to generate a control quantity based on the ideal control target, the measurable disturbance and the feedback value of the feedback module and output it to the cement production system; the cement production system is used to input the control quantity to act on the corresponding controlled quantity; the feedback module is used to measure the changes in the controlled quantity of the cement production system and feed back to the MPC controller.
[0010] Furthermore, the disturbance prediction module predicts the measurable disturbance by establishing a time series data prediction model, and the steps include:
[0011] S11. Collect historical data of cement production system;
[0012] S12. Use statistical analysis methods to establish a time series data prediction model.
[0013] Furthermore, the historical data includes cement production line parameters and measurable disturbances. The cement production line parameters include the temperature, pressure, and flow rate of the production line; the measurable disturbances include changes in cement raw materials and changes in the cement production line environment.
[0014] Furthermore, the time series data prediction model is expressed by the following formula:
[0015] [DV pred (t+1|t),DV pred (t+2|t),…,DV pred (t+Np|t)
[0016] =f(X(t),X(t-1),…,X(tn))
[0017] Among them, Np represents the prediction step size; DV pred (t+i|t) represents the measurable disturbance value predicted at time t+i within the prediction step, i ranges from 1 to Np; n represents the reference step; X(tj) represents the cement production line parameter value selected at time tj within the reference step; j ranges from 0 to n; f represents the prediction relationship between the predicted measurable disturbance value and the selected cement production line parameter value based on time series.
[0018] Further, the MPC controller is used to predict the influence of the measurable disturbance output by the disturbance prediction module on the controlled quantity of the cement production system, which is expressed by the following formula:
[0019] [y d(t+1|t),y d (t+2|t),...,y d (t+Np|t)]
[0020] =f d (DV pred (t+1|t),DV pred (t+2|t),...,DV pred (t+Np|t)
[0021] Among them, y d (t+i|t) represents the influence of the measurable disturbance predicted at time t+i on the controlled quantity, f d Represents the relationship model between measurable disturbance and controlled quantity.
[0022] Furthermore, the MPC controller is also used to calculate the loss function of the entire cement production control system, and the loss function is expressed by the following formula:
[0023]
[0024] Where y(t+i|t) represents the predicted output of the MPC controller control quantity at time t+i; Δu(t+j) represents the control quantity to be solved at time t+j; y ref represents the expected output of the MPC controller control variable; Nc represents the control step size.
[0025] Furthermore, the MPC controller is also used to solve the optimal control quantity sequence with the minimum value of the loss function as a constraint, and apply the solution at the first moment to the cement production system.
[0026] At the same time, the present application also proposes a cement production control method based on time series data prediction disturbance compensation, the method is applied to the above-mentioned cement production control system based on time series data prediction disturbance compensation, the method comprises the following steps:
[0027] S1. Establish a time series data prediction model to predict measurable disturbances;
[0028] S2. predicting the influence of the measurable disturbance on the controlled quantity of the cement production system;
[0029] S3, calculate the loss function of the entire control system;
[0030] S4. Solving the optimal cement production system control quantity sequence according to the loss function.
[0031] The technical effects of the present invention are:
[0032] 1. Improve prediction accuracy: By using a multivariate time series data prediction model, the present invention can dynamically predict the measurable disturbances within the future control step, rather than relying on the static values initially passed in. This dynamic prediction method can more accurately reflect the changes in disturbances in the actual system, thereby improving the accuracy of model prediction.
[0033] 2. Enhanced control adaptability: Since the present invention can predict changes in disturbances, the control strategy can be adjusted in real time to adapt to these changes. This adaptability enables the control system to more flexibly respond to uncertainties and disturbances in the production process, improving the robustness of the system.
[0034] 3. Reduce control error: By predicting the impact of disturbances on the controlled quantity and correcting the controlled quantity accordingly, the present invention can reduce the control error caused by disturbances. This correction mechanism helps to reduce the deviation between the predicted model state and the actual model state, thereby improving control accuracy.
[0035] 4. Optimize control effect: The present invention can more effectively achieve accurate control of the controlled quantity by predicting disturbances and compensating for them. This can not only improve production efficiency, but also reduce energy consumption and environmental pollution, thereby achieving better control effects.
[0036] 5. Improve system stability: By dynamically predicting and compensating for disturbances in real time, the present invention helps reduce fluctuations in the system caused by disturbances, thereby improving system stability. This is particularly important for continuous production processes such as cement production lines, as stability is directly related to production continuity and product quality. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 This is a structural diagram of a cement production control system based on time series data prediction and disturbance compensation according to an embodiment of the present invention. DETAILED DESCRIPTION
[0038] The following is a further detailed description of the specific implementation of the present invention by describing the embodiments with reference to the accompanying drawings, with the purpose of helping those skilled in the art to have a more complete, accurate and in-depth understanding of the inventive concept and technical solution of the present invention and to facilitate its implementation. In order to make the technical solution of the present invention clearer, the present invention is explained by the following embodiments.
[0039] Cement production control system based on time series data prediction disturbance compensation, such as Figure 1As shown, the control system includes a control target setting module, an MPC controller, a disturbance prediction module, a cement production system, and a feedback module, wherein the control target setting module is used to set an ideal control target for the cement production system; the disturbance prediction module is used to predict a measurable disturbance and output it to the MPC controller; the MPC controller is used to generate a control quantity according to the ideal control target, the measurable disturbance and the feedback value of the feedback module and output it to the cement production system. In this embodiment, the feedback module is set to negative feedback, that is, the difference between the ideal control target and the feedback value is used as the input of the MPC controller; the cement production system is used to apply the control quantity input to the corresponding controlled quantity; the feedback module is used to measure the change of the controlled quantity of the cement production system and feed it back to the MPC controller.
[0040] In the cement production process, measurable disturbances are often dynamically changing, such as changes in cement raw materials, changes in the cement production line environment, etc., which may lead to large deviations between the predicted model state and the actual model. For this reason, this embodiment introduces a disturbance compensation mechanism based on multivariable time series data prediction, which can dynamically predict and compensate for measurable disturbances and support the processing of multiple variables at the same time, that is, it can simultaneously consider multiple key parameters in related cement production processes, such as temperature, pressure, flow, etc., thereby improving the accuracy and reliability of the prediction. This is not common in traditional model predictive control. Specifically, the disturbance prediction module of this embodiment predicts measurable disturbances by establishing a time series data prediction model, and its steps include:
[0041] S11. Collect historical data of the cement production system, the historical data including cement production line parameters and measurable disturbances. The cement production line parameters include the temperature, pressure, flow rate, etc. of the production line; the measurable disturbances include changes in cement raw materials, changes in the cement production line environment, etc.
[0042] S12. Establish a time series data prediction model using statistical analysis methods. The statistical analysis methods include time series analysis, machine learning algorithms, etc. The statistical analysis methods described in this embodiment are all commonly used by those skilled in the art, and there are mature statistical models or algorithm models in the prior art. The unified core principle is to collect existing measurable disturbances and cement production line parameter data through big data, and calibrate the relationship between them. The calibrated data can constitute a statistical model for prediction on the one hand; on the other hand, it can be used as training and test data for machine learning algorithms, so as to build an algorithm model for prediction.
[0043] In this embodiment, the time series data prediction model is expressed by the following formula:
[0044] [DV pred (t+1|t),DV pred(t+2|t),…,DV pred (t+Np|t)
[0045] =f(X(t),X(t-1),…,X(tn))
[0046] Among them, Np represents the prediction step size; DV pred (t+i|t) represents the measurable disturbance value predicted at time t+i within the prediction step, i ranges from 1 to Np; n represents the reference step; X(tj) represents the cement production line parameter value selected at time tj within the reference step; j ranges from 0 to n; f represents the prediction relationship between the predicted measurable disturbance value and the selected cement production line parameter value based on the time series. The time series data prediction model of this embodiment realizes the dynamic prediction of changes in measurable disturbances instead of using static initial values. This mechanism allows the control system to adjust the prediction model according to real-time data, thereby more accurately predicting future disturbances and enhancing control adaptability. This adaptability enables the control system to more flexibly respond to uncertainties and disturbances in the production process, thereby improving the robustness of the system.
[0047] After obtaining the predicted measurable disturbance, in order to further improve the control accuracy of the MPC controller, this embodiment predicts the influence of the disturbance on the controlled quantity and corrects the control quantity output by the MPC controller accordingly, thereby reducing the control error caused by the disturbance. Specifically, the MPC controller is used to predict the influence of the measurable disturbance output by the disturbance prediction module on the controlled quantity of the cement production system, which is expressed by the following formula:
[0048] [y d (t+1|t),y d (t+2|t),...,y d (t+Np|t)]
[0049] =f d (DV pred (t+1|t),DV pred (t+2|t),...,DV pred (t+Np|t)
[0050] Among them, y d (t+i|t) represents the influence of the measurable disturbance predicted at time t+i on the controlled quantity, f d This correction mechanism helps to reduce the deviation between the predicted model state and the actual model state, thereby improving the control accuracy.
[0051] After obtaining the influence of the measurable disturbance on the controlled quantity, the MPC controller also incorporates compensation for the influence when calculating the loss function of the entire cement production control system. The loss function is expressed by the following formula:
[0052]
[0053] Where y(t+i|t) represents the predicted output of the MPC controller control quantity at time t+i; Δu(t+j) represents the control quantity to be solved at time t+j; y ref represents the expected output of the MPC controller control quantity; Nc represents the control step size. The present invention can more effectively achieve accurate control of the controlled quantity by predicting the disturbance and compensating it in the loss function. This can not only improve production efficiency, but also reduce energy consumption and environmental pollution, thereby achieving a better control effect.
[0054] After obtaining the loss function, the MPC controller solves the optimal control quantity sequence Δu(t+j) with the minimum value of the loss function as a constraint, and applies the solution at the first moment to the cement production system to achieve control of the cement production system.
[0055] Overall, the present invention can dynamically predict and compensate for measurable disturbances by introducing a disturbance compensation mechanism based on multivariable time series data prediction, which helps reduce the fluctuations of the control system caused by disturbances, thereby improving the stability and robustness of the control system. This is particularly important for continuous production processes such as cement production lines, because stability is directly related to production continuity and product quality.
[0056] At the same time, the present invention also proposes a cement production control method based on time series data prediction disturbance compensation, the method is applied to the above-mentioned cement production control system based on time series data prediction disturbance compensation, the method comprises the following steps:
[0057] S1. Establish a time series data prediction model to predict measurable disturbances;
[0058] S2. predicting the influence of the measurable disturbance on the controlled quantity of the cement production system;
[0059] S3, calculate the loss function of the entire control system;
[0060] S4. Solving the optimal cement production system control quantity sequence according to the loss function.
[0061] The present invention is described above by way of example in conjunction with the accompanying drawings. Obviously, the specific implementation of the present invention is not limited to the above-mentioned methods. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention; or the above-mentioned concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.
Claims
1. A cement production control system based on time series data prediction disturbance compensation, characterized by: The control system includes a control target setting module, an MPC controller, a disturbance prediction module, a cement production system, and a feedback module, wherein the control target setting module is used to set an ideal control target for the cement production system; the disturbance prediction module is used to predict measurable disturbances and output them to the MPC controller; the MPC controller is used to generate a control quantity based on the ideal control target, the measurable disturbance and the feedback value of the feedback module and output it to the cement production system; the cement production system is used to input the control quantity to act on the corresponding controlled quantity; and the feedback module is used to measure changes in the controlled quantity of the cement production system and feed back the changes to the MPC controller.
2. The cement production control system based on time series data prediction disturbance compensation according to claim 1 is characterized in that: The disturbance prediction module predicts the measurable disturbance by establishing a time series data prediction model, and the steps include: S11. Collect historical data of cement production system; S12. Use statistical analysis methods to establish a time series data prediction model.
3. The cement production control system based on time series data prediction disturbance compensation according to claim 2 is characterized in that: The historical data includes cement production line parameters and measurable disturbances.
4. The cement production control system based on time series data prediction disturbance compensation according to claim 3 is characterized in that: The cement production line parameters include the temperature, pressure and flow of the production line; the measurable disturbances include changes in cement raw materials and changes in the cement production line environment.
5. The cement production control system based on time series data prediction disturbance compensation according to claim 3 or 4, characterized in that: The time series data prediction model is expressed by the following formula: [DV pred (t+1∣t),DV pred (t+2∣t),…,DV pred (t+Np∣t) =f(X(t),X(t-1),…,X(tn)) Where Np represents the prediction step length; DV pred (t+i|t) represents the measurable disturbance value predicted at time t+i within the prediction step, i ranges from 1 to Np; n represents the reference step; X(tj) represents the cement production line parameter value selected at time tj within the reference step; j ranges from 0 to n; f represents the prediction relationship between the predicted measurable disturbance value and the selected cement production line parameter value based on time series.
6. The cement production control system based on time series data prediction disturbance compensation according to claim 5 is characterized in that: The MPC controller is used to predict the influence of the measurable disturbance output by the disturbance prediction module on the controlled quantity of the cement production system, which is expressed by the following formula: [y d (t+1|t),y d (t+2|t),...,y d (t+Np|t)] =f d (DV pred (t+1|t),DV pred (t+2|t),...,DV pred (t+Np|t) Among them, y d (t+i|t) represents the influence of the measurable disturbance predicted at time t+i on the controlled quantity, f d Represents the relationship model between measurable disturbance and controlled quantity.
7. The cement production control system based on time series data prediction disturbance compensation according to claim 6 is characterized in that: The MPC controller is also used to calculate the loss function of the entire cement production control system, and the loss function is expressed by the following formula: Where y(t+i|t) represents the predicted output of the MPC controller control quantity at time t+i; Δu(t+j) represents the control quantity to be solved at time t+j; y ref represents the expected output of the MPC controller control variable; Nc represents the control step size.
8. The cement production control system based on time series data prediction disturbance compensation according to claim 7 is characterized in that: The MPC controller is also used to solve the optimal control quantity sequence with the minimum value of the loss function as a constraint, and apply the solution at the first moment to the cement production system.
9. A cement production control method based on time series data prediction disturbance compensation, characterized in that: The method is applied to a cement production control system based on time series data prediction disturbance compensation according to any one of claims 1 to 8, and the method comprises the following steps: S1. Establish a time series data prediction model to predict measurable disturbances; S2. predicting the influence of the measurable disturbance on the controlled quantity of the cement production system; S3, calculate the loss function of the entire control system; S4. Solving the optimal cement production system control quantity sequence according to the loss function.