Over-range soft measurement method and feedforward controller

Abstract

The invention relates to an outrange soft measurement method and a feedforward controller, and belongs to the technical field of industrial control methods and devices. According to the technical scheme, the coal gas flow is adjusted in a feed-forward PID incremental mode with an influence degree coefficient, and coal gas pressure serves as feed-forward; when the gas flow is in the normal range, historical data are collected, a regression model is established, the model is used for calculating the following gas flow value, and when the gas flow is in the measuring range, the regression model only updates the numerical value and does not output the numerical value; when the gas flow exceeds an upper limit value, switching to a soft measurement model, and taking the model output as a reference value of the gas flow; and when the gas flow returns to the measuring range again, continuing to update the model parameters, disconnecting model output, and repeating the steps. According to the method, a feedforward PID incremental adjustment mode with an influence degree coefficient is adopted, the real gas flow on site is restored and participates in control, and under the condition of large fluctuation of gas pressure, the gas flow is adjusted rapidly and stably.

Description

Technical Field

[0001] This invention relates to an over-range soft measurement method and a feedforward controller, belonging to the technical field of industrial control methods and devices. Background Technology

[0002] In industrial production, precise control is crucial for ensuring product quality. Feedforward control is a common control strategy that predicts the output of the controlled object and compensates for errors in advance. However, traditional feedforward controllers have certain limitations, such as slow response speed and poor anti-interference ability. Furthermore, overrange soft sensing technology is a method for estimating process parameters that are difficult to measure directly through model prediction and data processing. Traditional overrange soft sensing technology has certain limitations in practical applications, such as low prediction accuracy and poor adaptability. In furnace combustion scenarios with large fluctuations in gas pressure, the actual gas or air flow rate often exceeds the range. Summary of the Invention

[0003] The purpose of this invention is to provide an over-range soft measurement method and a feedforward controller. By adopting a feedforward PID incremental adjustment method with an influence coefficient, the actual gas flow rate on site is reproduced and participates in the control. Under the condition of large fluctuations in gas pressure, the gas flow rate is adjusted quickly and stably. The required equipment is simple, the economic benefits are significant, and the above-mentioned problems existing in the background technology are effectively solved.

[0004] The technical solution of this invention is: an over-range soft measurement method, comprising the following steps:

[0005] (1) The gas flow rate is controlled by feedforward PID incremental regulation with influence coefficient, and the gas pressure is used as the feedforward.

[0006] (2) When the gas flow rate is within the normal range, collect historical data, establish a regression model, and use this model to calculate the next gas flow rate value.

[0007] (3) When the gas flow rate is within the range, the regression model only updates the values ​​and does not output any values;

[0008] (4) When the gas flow rate exceeds the upper limit, switch to the soft measurement model and use the output of this model as the reference value of the gas flow rate.

[0009] (5) When the gas flow rate returns to the range, continue to update the model parameters, disconnect the model output, and repeat this cycle.

[0010] In step (1), the feedforward PID incremental regulation formula for gas flow is...

[0011] feedforward dsv t =P (gas,t) / 0.1*k

[0012] Where k is the influence coefficient, representing the degree of influence of gas pressure fluctuations on gas flow rate; P (gas,t) This represents the current gas pressure.

[0013] The value of k is -0.3; the gas pressure P gas Fluctuation of 10±2 kPa, P (gas,t) / 0.10 corresponds to 100% of the full scale of the gas valve position.

[0014] In step (2), the values ​​of gas flow rate, gas pipe pressure and gas valve opening are recorded within 60 seconds, and a regression model is established based on this data.

[0015] In step (4), the soft measurement model is in the form of: OMQ = f(V,P,A), where OMQ represents the gas flow rate, V is the gas valve opening, P is the gas pipe pressure, and A is the model parameter, which is continuously calculated and updated based on historical data within a 60-second range.

[0016] A feedforward controller for over-range soft measurement includes a prediction unit, a compensation unit, and an output unit. The prediction unit takes a gas pressure and gas flow detection device as its input terminal, is connected to the compensation unit through an over-range flow prediction selector, and the compensation unit is connected to the output unit.

[0017] The beneficial effects of this invention are: by adopting a feedforward PID incremental regulation method with an influence coefficient, the actual gas flow rate on site is restored and participates in the control. Under the condition of large fluctuations in gas pressure, the gas flow rate is regulated quickly and stably. The required equipment is simple and the economic benefits are significant. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the working principle of the present invention;

[0019] Figure 2 This is a diagram showing the predicted gas flow rate when the gas flow rate is within the specified range according to an embodiment of the present invention.

[0020] Figure 3 This is a diagram showing the predicted gas flow rate when the gas flow rate exceeds the upper limit in an embodiment of the present invention.

[0021] In the figure: Model prediction curve 1, Actual gas flow rate curve 2. Detailed Implementation

[0022] To make the purpose, technical solutions, and advantages of the invention's embodiments clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only a small part of the embodiments of the present invention, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0023] An over-range soft measurement method includes the following steps:

[0024] (1) The gas flow rate is controlled by feedforward PID incremental regulation with influence coefficient, and the gas pressure is used as the feedforward.

[0025] (2) When the gas flow rate is within the normal range, collect historical data, establish a regression model, and use this model to calculate the next gas flow rate value.

[0026] (3) When the gas flow rate is within the range, the regression model only updates the values ​​and does not output any values;

[0027] (4) When the gas flow rate exceeds the upper limit, switch to the soft measurement model and use the output of this model as the reference value of the gas flow rate.

[0028] (5) When the gas flow rate returns to the range, continue to update the model parameters, disconnect the model output, and repeat this cycle.

[0029] In step (1), the feedforward PID incremental regulation formula for gas flow is...

[0030] feedforward dsv t =P (gas,t) / 0.1*k

[0031] Where k is the influence coefficient, representing the degree of influence of gas pressure fluctuations on gas flow rate; P (gas,t) This represents the current gas pressure.

[0032] The value of k is -0.3; the gas pressure P gas Fluctuation of 10±2 kPa, P (gas,t) / 0.10 corresponds to 100% of the full scale of the gas valve position.

[0033] In step (2), the values ​​of gas flow rate, gas pipe pressure and gas valve opening are recorded within 60 seconds, and a regression model is established based on this data.

[0034] In step (4), the soft measurement model is in the form of: OMQ = f(V,P,A), where OMQ represents the gas flow rate, V is the gas valve opening, P is the gas pipe pressure, and A is the model parameter, which is continuously calculated and updated based on historical data within a 60-second range.

[0035] A feedforward controller for over-range soft measurement includes a prediction unit, a compensation unit, and an output unit. The prediction unit takes a gas pressure and gas flow detection device as its input terminal, is connected to the compensation unit through an over-range flow prediction selector, and the compensation unit is connected to the output unit.

[0036] In practical applications, the flow rate is adjusted using a feedforward PID incremental regulation method with an influence coefficient, based on the gas pressure P. gas As a feedforward, this feedforward quantity carries an influence coefficient k, which represents the degree of influence of gas pressure fluctuations on gas flow rate. Due to large fluctuations in on-site gas pressure, gas flow rate tracking must be rapid and stable to avoid prolonged deviations from the optimal air-fuel ratio, which could lead to excessive CO content and heat loss.

[0037] feedforward dsv t =P (gas,t) / 0.1*k, where k is -0.3, P (gas,t) Let P be the current gas pressure. gas Fluctuation of 10±2 kPa, P (gas,t) / 0.10 corresponds to 100% of the full scale of the gas valve position.

[0038] Because the on-site gas flow rate range is 150,000 m³ 3 At that time, the gas flow rate exceeded 150,000 m³. 3 If the gas flow meter fails to provide a reliable reference, a soft measurement method will be used here.

[0039] The established model is in the form of: OMQ=f(V,P,A), where OMQ represents the gas flow rate, V is the gas valve opening, P is the gas pipe pressure, and A is the model parameter, which is continuously calculated and updated based on historical data within a 60-second range.

[0040]

[0041]

[0042] A strategy of real-time updating using a soft-sensor model is adopted. When the flow rate is within the normal range, the values ​​of gas flow rate, gas pipe pressure, and gas valve opening are recorded over a 60-second period. A regression model is built using this data to calculate the next gas flow rate value. When the gas flow rate is within the range, this model only updates the values ​​without outputting any data. When the gas flow rate exceeds the upper limit, the system switches to the soft-sensor model, and its output is used as a reference value for the gas flow rate. When the gas flow rate returns to the range, the model parameters are updated again, and the model output is disconnected, thus repeating the cycle.

[0043] Soft measurement model simulation verification: Calculate A using historical data values ​​within a 60-second range, and conduct tests based on the soft measurement model to calculate the maximum error and root mean square error between the soft measurement model and the true value.

[0044] like Figure 2 When the gas flow rate is within the range, the soft measurement model shows good performance.

[0045] Since the soft-sensor model performs well within its measurement range, there is reason to believe that the gas flow rate calculated using this model is reliable when the gas flow rate exceeds the upper limit. A model was built using data from the first 60 seconds before the gas flow rate exceeded its limit, and the flow rate after exceeding the limit was predicted based on this model. The results are as follows. Figure 3 .

[0046] The main advantages of this invention are:

[0047] (1) The changes in the parameters of the controlled object are robust and suitable for controlling complex systems that are difficult to establish mathematical models of the controlled object, such as nonlinear, time-varying, and lag systems.

[0048] (2) It has good control effect, requires simple equipment, and has significant economic benefits.

[0049] This invention uses data simulation to predict and solve the problems of instrument over-range and data errors, providing a more accurate control basis for the feedforward controller, which can adjust the controlled object more quickly and accurately to ensure its stable operation.

Claims

1. A method for over-range soft measurement, characterized in that... Includes the following steps: (1) The gas flow rate is controlled by feedforward PID incremental regulation with influence coefficient, and the gas pressure is used as the feedforward. (2) When the gas flow rate is within the normal range, collect historical data, establish a regression model, and use this model to calculate the next gas flow rate value. (3) When the gas flow rate is within the range, the regression model only updates the values ​​and does not output any values; (4) When the gas flow rate exceeds the upper limit, switch to the soft measurement model and use the output of this model as the reference value of the gas flow rate. (5) When the gas flow rate returns to the range, continue to update the model parameters, disconnect the model output, and repeat this cycle.

2. The over-range soft measurement according to claim 1, characterized in that: In step (1), the feedforward PID incremental regulation formula for gas flow is... feedforward dsv t =P (gas,t) / 0.1*k Where k is the influence coefficient, representing the degree of influence of gas pressure fluctuations on gas flow rate; P (gas,t) This represents the current gas pressure.

3. The over-range soft measurement according to claim 2, characterized in that: The value of k is -0.3; the gas pressure P gas Fluctuation of 10±2 kPa, P (gas,t) / 0.10 corresponds to 100% of the full scale of the gas valve position.

4. The over-range soft measurement according to claim 1, characterized in that: In step (2), the values ​​of gas flow rate, gas pipe pressure and gas valve opening are recorded within 60 seconds, and a regression model is established based on this data.

5. The over-range soft measurement according to claim 1, characterized in that: In step (4), the soft measurement model is in the form of: OMQ = f(V,P,A), where OMQ represents the gas flow rate, V is the gas valve opening, P is the gas pipe pressure, and A is the model parameter, which is continuously calculated and updated based on historical data within a 60-second range.

6. A feedforward controller for over-range soft measurement, characterized in that: It includes a prediction unit, a compensation unit, and an output unit. The prediction unit uses a gas pressure and gas flow detection device as its input terminal, and is connected to the compensation unit through an over-range flow prediction selector. The compensation unit is connected to the output unit.

Images