Design method of automobile exhaust SCR control system based on improved nonlinear active disturbance rejection

Abstract

The present invention discloses a method for designing an automobile exhaust SCR control system based on improved nonlinear self-anti-disturbance control, which belongs to the field of control engineering technology. The method of the present invention comprises: establishing a nonlinear self-anti-disturbance control system for automobile exhaust SCR; designing an ammonia injection pretreatment module to adjust the initial ammonia input; designing a control object monitoring module, setting the control object to a single nitrogen oxide emission, introducing the nitrogen oxide input value change rate to select the ammonia emission monitoring interval; designing a temperature control module to monitor ammonia emissions regularly, and increasing the temperature of the SCR reaction device to reduce ammonia emissions each time the ammonia emissions exceed the standard. The present invention solves the problems of poor control performance and multi-control object control of automobile exhaust SCR control systems, and can accurately and quickly control the pollutant emissions of automobile exhaust SCR systems under complex system disturbances, thereby improving the anti-disturbance and real-time performance of the SCR control system.

Description

Technical Field

[0001] The present invention belongs to the technical field of control engineering, relates to the emission control of nitrogen oxides and ammonia, and particularly relates to a design method of an automobile exhaust SCR control system based on improved nonlinear auto-disturbance rejection. Background Art

[0002] Nitrogen oxide pollution refers to the environmental pollution caused by the release of nitrogen monoxide and nitrogen dioxide into the atmosphere as byproducts of combustion. Once released into the atmosphere, nitrogen oxides can form acid rain and photochemical smog, depleting atmospheric ozone and causing further damage. Therefore, controlling nitrogen oxide emissions is crucial for air pollution prevention and control.

[0003] Nitrogen oxide pollution is primarily generated by industries with high demand for nitrogen-containing fossil fuels, such as thermal power generation, automobile exhaust, and cement manufacturing. The mainstream solution for nitrogen oxide control is post-combustion denitrification. Selective catalytic reduction (SCR) technology is recognized as the most efficient solution in the automobile exhaust treatment industry. The core of SCR technology lies in the injection of ammonia to neutralize nitrogen oxides. This technology relies on precise control of the ammonia injection rate to achieve efficient and economical nitrogen oxide control. However, ammonia escape into the atmosphere can also cause air pollution. When the SCR system is stable, the amount of ammonia escape and nitrogen oxide emissions may not simultaneously meet emission targets.

[0004] In SCR control system research, traditional control systems such as model predictive control systems, adaptive feedforward control systems, and fuzzy PID control systems have been improved based on the SCR system model to enhance control performance and achieve rapid and accurate control of ammonia injection. However, in automotive exhaust SCR control systems, external conditions such as high-speed and drastic fluctuations in engine load cause significant disturbances to the SCR control system. When these disturbances cannot be modeled, existing control systems have a poor ability to quickly offset complex system disturbances. With the rapid development of artificial intelligence, data-driven ammonia injection control algorithms have also gradually emerged. Machine learning algorithms such as random forest algorithms and support vector machines predict and analyze ammonia injection based on SCR system input and output data. However, data-driven control algorithms have long training cycles and face the problem of overfitting. Summary of the Invention

[0005] To address the poor control performance and multiple control objects of existing automotive exhaust SCR control systems, this paper proposes a design method for an automotive exhaust SCR control system based on an improved nonlinear active disturbance rejection (ADRC) system. This method controls ammonia injection using a nonlinear active disturbance rejection control (ADRC) system. It designs an ammonia injection preprocessing module based on the SCR model to rationally set the initial system input value. It also designs a control object monitoring module to determine the range of control object changes and analyze whether control object target values ​​conflict. Finally, it designs a temperature control module to regulate the SCR device temperature to avoid control object conflicts. Ultimately, this method achieves a design for an automotive exhaust SCR ADRC control system with high interference immunity, no overshoot, and sufficient control of multiple control objects.

[0006] The present invention proposes a design method for an automobile exhaust SCR control system based on improved nonlinear auto-disturbance rejection, which includes the following four steps:

[0007] Step 1: Establish the automobile exhaust SCR reaction model and nonlinear active disturbance rejection control system;

[0008] The SCR reaction system model is derived based on the internal chemical reaction equation, kinetic equation, ideal gas equation, and mass conservation equation of the SCR system. An automobile exhaust SCR control system is established based on a nonlinear active disturbance rejection system and the system parameters are adjusted.

[0009] Step 2: Design an ammonia injection preprocessing module of the nonlinear active disturbance rejection control system to adjust the initial ammonia input;

[0010] Considering the chemical reaction characteristics of the SCR system, which sees nitrogen oxide and ammonia concentrations surge initially and then rapidly slow, ammonia injection pre-processing is implemented based on the SCR system reaction model, using the nitrogen oxide concentration change rate. The ammonia injection pre-processing module calculates the ammonia injection increment based on the nitrogen oxide concentration change rate, and also calculates the accumulated ammonia injection output value and the rate of change of the ammonia injection output value over time. When the rate of change of the ammonia injection output value meets a set stop signal, the module stops accumulating the ammonia injection output value and outputs the current accumulated ammonia injection output value, increasing it to the initial ammonia input value.

[0011] Step 3: Design a control object monitoring module for the nonlinear active disturbance rejection control system;

[0012] To address the multiple control objects in an SCR system, the control object monitoring module sets the control object as a single NOx emission target while simultaneously calculating ammonia emissions. The control object monitoring module uses the rate of change of the NOx input value to select the ammonia emissions monitoring interval. When the NOx input value stabilizes, ammonia emissions monitoring is performed periodically.

[0013] Step 4: Design the temperature control module of the nonlinear active disturbance rejection control system.

[0014] When the nitrogen oxide input value is stable, the temperature control module regularly detects ammonia emissions and increases the temperature of the SCR reaction device to reduce ammonia emissions every time it detects that the ammonia emissions exceed the standard.

[0015] Compared with the prior art, the advantages and positive effects of the method of the present invention are:

[0016] (1) This paper proposes a design method for an automobile exhaust SCR control system based on improved nonlinear self-disturbance rejection, which can accurately and quickly control the pollutant emissions of the automobile exhaust SCR system under complex system disturbances, thereby improving the anti-disturbance and real-time performance of the control system.

[0017] (2) The present invention designs an ammonia injection pretreatment module based on the SCR reaction model, which is beneficial to solving the problem of large overshoot and long rise time in the initial control stage of the nonlinear self-disturbance rejection system, and improves the efficiency and speed of the automobile exhaust SCR control system.

[0018] (3) The present invention uses nitrogen oxide emissions as the control object and monitors ammonia emissions. It introduces the SCR device temperature as a control variable for the first time. When the nitrogen oxide emission target conflicts with the ammonia emission target, the temperature control method is used to ensure that both nitrogen oxide and ammonia emissions meet the emission indicators, thereby improving the accuracy of the automobile exhaust SCR control system. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a flow chart of the design method of the automobile exhaust SCR control system based on the improved nonlinear auto-disturbance rejection system of the present invention;

[0020] Figure 2 This is a structural diagram of the improved nonlinear automobile exhaust SCR self-disturbance rejection control system in the present invention;

[0021] Figure 3 This is a comparison of the anti-interference performance of the automobile exhaust SCR control system in the embodiment;

[0022] Figure 4 This is a comparison of the control performance of the improved active disturbance rejection control system and the traditional active disturbance rejection control system in the embodiment;

[0023] Figure 5 The embodiment improves the performance of the control object monitoring module of the active disturbance rejection control system;

[0024] Figure 6 This is the control effect of the temperature control module of the improved active disturbance rejection control system in the embodiment. DETAILED DESCRIPTION

[0025] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0026] The present invention addresses the issue of complex system disturbances affecting control performance in automobile exhaust SCR control systems and designs an automobile exhaust SCR control system based on a nonlinear auto-disturbance rejection control system. Taking into account the SCR reaction characteristics, namely, the fact that the concentrations of nitrogen oxides and ammonia surge in the initial stage of the reaction and then rapidly slow down, the traditional nonlinear auto-disturbance rejection system has a large control overshoot and a long rise time in the initial stage. Therefore, an ammonia injection amount preprocessing module is designed to reasonably set the initial value of the control system input. Considering that the SCR system needs to meet the requirement that both nitrogen oxide emissions and ammonia emissions do not exceed the standard, nitrogen oxide emissions are used as the control object and ammonia emissions are monitored. A control object monitoring module and a temperature control module are designed to adjust the SCR device temperature when the emissions of the two gases cannot simultaneously meet the requirements, thereby avoiding control target conflicts. Ultimately, a design of an automobile exhaust SCR auto-disturbance rejection control system with high anti-disturbance performance, no overshoot, and the ability to control multiple control objects is achieved.

[0027] like Figure 1 As shown, the design method of the automobile exhaust SCR control system based on improved nonlinear auto-disturbance rejection implemented in the embodiment of the present invention includes the following four steps, each of which is described below.

[0028] Step 1: Establish the automobile exhaust SCR reaction model and nonlinear active disturbance rejection control system.

[0029] The SCR reaction system model can be deduced based on the internal chemical reaction equation, kinetic equation, ideal gas equation and mass conservation equation of the SCR system.

[0030]

[0031] in is the nitrogen oxide emission, C NO,IN is the nitrogen oxide input, is the ammonia emission, is the ammonia input, EF v is the ratio of exhaust volume flow rate to SCR system volume, θ max is the maximum ammonia adsorption capacity of the catalyst, θ is the ammonia adsorption capacity of the catalyst, T is the temperature of the SCR device, and R1 to R5 are temperature parameters with the following values:

[0032]

[0033] For a second-order control system, we have

[0034]

[0035]

[0036] Where t is the system time, ω(t) is the external disturbance of the system, is the total disturbance of the system, y is the state variable, are the first and second derivatives of y, respectively, and b is the control gain. Let the state variable Formula (3) can be transformed into

[0037]

[0038] In order to eliminate the disturbance in the SCR control system A nonlinear controller is introduced to calculate the ammonia input, and a linear extended state observer is introduced to observe the state variables of the SCR system, such as Figure 2 shown.

[0039]

[0040] Where z1, z2, z3 are x1, x2, The state variable is, e is the observed difference of nitrogen oxide emissions, e1 is the error between the nitrogen oxide emission value and the target value, e2 is the error between the rate of change of nitrogen oxide emissions and the rate of change of the target value, is the initial ammonia input, b is the control gain, k1, k2, a1, a2, δ1, δ2 are the observer adjustable gains, β1, β2, β3 are the observer adjustable parameters, and their values ​​are obtained according to the bandwidth method, as shown in formula (6). The function fal(x, a, δ) is shown in formula (7).

[0041]

[0042] Where ω0 is the system bandwidth.

[0043]

[0044] Where sgn is the sign function.

[0045] Figure 2 Where r is the nitrogen oxide input value, C0 is the adjustment value of the initial ammonia input amount calculated by the ammonia injection pretreatment module, v1 is the nitrogen oxide emission target tracking value, and v2 is the nitrogen oxide emission target tracking rate.

[0046] Step 2: Design the ammonia injection amount preprocessing module of the nonlinear active disturbance rejection control system.

[0047] According to the SCR reaction system model established in step 1, the nitrogen oxide concentration change rate is introduced To calculate the increase in ammonia injection amount f1, as shown in formula (8):

[0048]

[0049] Where e3 is the error between nitrogen oxide emissions and the target value, a3 and δ3 are the module adjustable gains, and s is used to indicate the rate of change of nitrogen oxide concentration. Whether to increase.

[0050] The ammonia injection amount increase value is accumulated to calculate the ammonia injection amount output value f2(t), as shown in formula (9):

[0051]

[0052] Calculate the rate of change of the ammonia injection output value f3(t), as shown in formula (10):

[0053]

[0054] Among them, f2(t) and f2(t-1) are the output values ​​of the ammonia injection amount at time t and the previous moment respectively.

[0055] Determine the accumulation stop signal f4(t), as shown in formula (11):

[0056]

[0057] Where μ1 is the stop signal parameter.

[0058] When f4(t)=0, stop calculating the ammonia injection output value f2(t), and substitute f2(t) back into formula (5), as follows:

[0059]

[0060] Step 3: Design the control object monitoring module of the nonlinear active disturbance rejection control system.

[0061] The control object is set to control nitrogen oxide emissions and ammonia emissions at the same time. The nitrogen oxide input value change rate is introduced to monitor the change range of the control object, as shown in formulas (13), (14), and (15):

[0062]

[0063] Where r(t) and r(t-1) are the nitrogen oxide input values ​​at time t and time t-1, respectively. start The delay start time, such as 0.05 seconds, is used to prevent the monitoring module from misoperation. When the nitrogen oxide input value changes, f5(t) is 1, otherwise it is 0. The nitrogen oxide input value change signal f5(t) is accumulated and its changes are detected regularly, and the following is obtained:

[0064]

[0065] where t setThis is the detection interval for the sensor to detect whether the ammonia emission exceeds the standard.

[0066]

[0067] When the nitrogen oxide input value is stable, f7(t) is 1, and ammonia emissions are monitored.

[0068] Step 4: Design the temperature control module of the nonlinear active disturbance rejection control system.

[0069] The ammonia emission f8(t) is detected periodically within the range where the nitrogen oxide input value is constant, as follows:

[0070]

[0071] Where u1 is the current ammonia emission detected under the nitrogen oxide emission standard.

[0072] Determine whether the ammonia emissions meet the standards, as shown in formula (17):

[0073]

[0074] where u tar is the ammonia emission limit. A value of 0 for f9(t) indicates that the standard is met, and a value of 1 indicates that the standard is not met.

[0075] The temperature control module sets the adjustment temperature f of the SCR reaction device according to f9(t) 10 (t), as follows:

[0076]

[0077] T1 is the temperature value that increases each time when the ammonia emission exceeds the standard.

[0078] Each time the ammonia emission exceeds the standard, the temperature of the SCR reactor is increased to reduce the ammonia emission, completing the design of the temperature control module. After N times of temperature adjustment, the temperature of the SCR reactor is T SCR for:

[0079]

[0080] Where T0 is the initial temperature of the SCR reaction device.

[0081] Example

[0082] Step 1: Establish the automobile exhaust SCR reaction model and nonlinear active disturbance rejection control system;

[0083] The simulation experiment of this embodiment is carried out in the MATLAB / Simulink simulation environment, and a diesel engine vehicle exhaust SCR system model is established. The SCR system parameters are shown in Table 1.

[0084] Table 1 Diesel engine exhaust SCR system parameters

[0085]

[0086]

[0087] The parameters of the nonlinear active disturbance rejection control system are shown in Table 2.

[0088] Table 2 Parameters of nonlinear active disturbance rejection control system

[0089] Parameter Symbol Parameter name Parameter value <![CDATA[ω0]]> System bandwidth 4.8 b Control gain 0.12 <![CDATA[a1]]> Observer adjustable gain 0.1 <![CDATA[a2]]> Observer adjustable gain 0.3 <![CDATA[δ1]]> Observer adjustable gain 0.6 <![CDATA[δ2]]> Observer adjustable gain 0.8 <![CDATA[k1]]> Observer adjustable gain 1 <![CDATA[k2]]> Observer adjustable gain 1

[0090] The target value for nitrogen oxide emissions is set at 0.2 mg / m 3 And add Gaussian white noise external disturbance to the nonlinear active disturbance rejection control system (ADRC), and compare the control error with the control error of the traditional PID control system. The results are as follows Figure 3 As shown. Figure 3 It can be seen that compared with the PID control method, the nonlinear active disturbance rejection control system has a stronger ability to suppress complex external disturbances.

[0091] Step 2: Design an ammonia injection preprocessing module for the nonlinear active disturbance rejection control system;

[0092] According to the SCR reaction system model established in step 1, the nitrogen oxide concentration change rate is introduced, and the ammonia injection amount is accumulated before the cumulative stop signal output is 0. The parameters of the ammonia injection amount preprocessing module are shown in Table 3.

[0093] Table 3 Nonlinear active disturbance rejection control system parameters

[0094] Parameter Symbol Parameter name Parameter value <![CDATA[a3]]> Module adjustable gain 0.4 <![CDATA[δ3]]> Module adjustable gain 0.9 <![CDATA[μ1]]> Stop signal parameters 0.5

[0095] After the ammonia injection pretreatment system was introduced, the nitrogen oxide control target values ​​were set to 0.3 mg / m at 0s, 35s, and 70s respectively. 3 , 0.4mg / m 3 , 0.6mg / m 3 , the control effects of the traditional nonlinear active disturbance rejection control system and the nonlinear active disturbance rejection system with ammonia injection pretreatment module are compared. The results are as follows Figure 4 As shown. Figure 4 It can be seen that compared with the traditional nonlinear active disturbance rejection control system, the improved active disturbance rejection control system containing the ammonia injection amount pretreatment module of the present invention can significantly reduce system overshoot and achieve zero overshoot control.

[0096] Step 3: Design a control object monitoring module for the nonlinear active disturbance rejection control system;

[0097] The control object is set to control nitrogen oxide emissions and calculate ammonia emissions at the same time. The change rate of nitrogen oxide input value is introduced to select the monitoring interval of ammonia emissions. The detection time interval t set The nitrogen oxide input value r(t) changes at 18s, 30s-40s, and 85s, and the delayed start time value t start The monitoring module identifies the change interval of the nitrogen oxide input value (i.e., the ammonia emission monitoring interval) and the results are as follows: Figure 5 shown.

[0098] Step 4: Design the temperature control module of the nonlinear active disturbance rejection control system.

[0099] In the range where the nitrogen oxide input value is constant, the internal ammonia emission is regularly checked to see if it exceeds the limit. tar 20.6 mg / m 3 , the temperature increase value T1 is 10k. The temperature control module controls the temperature of the SCR device as follows Figure 6 As shown. Figure 6 It can be seen that the method of the present invention can monitor and control the ammonia emission index while controlling the nitrogen oxide emission.

[0100] The results show that the method of the present invention can better solve the problem of poor control performance of the nonlinear auto-disturbance rejection automobile exhaust SCR control system and meet the control requirements of multiple control objects.

[0101] Except for the technical features described in the specification, all other technical features are known to those skilled in the art. The present invention omits the description of known technologies. The implementation methods described in the above embodiments do not represent all implementation methods consistent with the present application. Based on the technical solution of the present invention, various modifications or variations that can be made by those skilled in the art without creative effort are still within the scope of protection of the present invention.

Claims

1. A design method for an automobile exhaust SCR control system based on improved nonlinear auto-disturbance rejection, characterized in that: The steps include: Step 1: Establish the automobile exhaust SCR reaction model and nonlinear active disturbance rejection control system; The established automobile exhaust SCR reaction model is as follows: in, is the nitrogen oxide emission, C NO,IN is the nitrogen oxide input, is the ammonia emission, is the ammonia input, EF v is the ratio of exhaust volume flow to SCR system volume, θ is the ammonia adsorption capacity of the catalyst, and θ max is the maximum ammonia adsorption capacity of the catalyst, T is the temperature of the SCR device; R1 to R5 are parameters related to temperature T; In order to eliminate the disturbance in the SCR control system, a nonlinear controller is introduced to calculate the ammonia input, and a linear extended state observer is introduced to observe the state variables of the SCR system. The state variables include in The nonlinear auto-disturbance rejection control system of automobile exhaust SCR is established as follows: Among them, z1, z2, z3 are The observed value is, e is the observed difference of nitrogen oxide emissions, e1 is the error between the nitrogen oxide emission value and the target value, e2 is the error between the rate of change of nitrogen oxide emissions and the rate of change of the target value, is the initial ammonia input, β1, β2, β3 are the observer adjustable parameters, k1, k2, a1, a2, δ1, δ2 are the observer adjustable gains, and b is the control gain; function The function is used to calculate fal(e1,a1,δ1) and fal(e2,a2,δ2), where sgn is the sign function; Step 2: Design an ammonia injection preprocessing module of the nonlinear active disturbance rejection control system to adjust the initial ammonia input; The ammonia injection amount preprocessing module calculates the ammonia injection amount increase value according to the nitrogen oxide concentration change rate, and calculates the ammonia injection amount output value accumulated over time and the ammonia injection amount output value change rate. When the ammonia injection amount output value change rate meets the set stop signal, the ammonia injection amount output value is stopped from being accumulated, and the current accumulated ammonia injection amount output value is output, which is added to the initial ammonia input amount. Step 3: Design a control object monitoring module for the nonlinear active disturbance rejection control system. The control object monitoring module sets the control object as a single nitrogen oxide emission, introduces the rate of change of the nitrogen oxide input value for monitoring, and regularly monitors ammonia emissions when the nitrogen oxide input value is stable. Step 4: Design a temperature control module for the nonlinear active disturbance rejection control system. When the nitrogen oxide input value is stable, the temperature control module regularly detects ammonia emissions. Whenever it detects that the ammonia emissions exceed the standard, it increases the temperature of the SCR reactor to reduce ammonia emissions.

2. The method according to claim 1, characterized in that In the step 2, the implementation of the ammonia injection amount pretreatment module includes: First, the nitrogen oxide concentration change rate is introduced Calculate the increase in ammonia injection amount f1 as follows: Among them, the identification s is based on The value is determined by whether it is greater than 0, e3 is the error between the nitrogen oxide emission and the nitrogen oxide target value, a3 and δ3 are the adjustable gains of the ammonia injection pretreatment module; Then, the increase in the amount of ammonia injected is accumulated and the output value of the amount of ammonia injected, f2(t), is calculated as follows: Where, t represents time t; Calculate the rate of change of the ammonia injection output value f3(t)=f2(t)-f2(t-1); The accumulation stop signal f4(t) is determined according to f3(t) as follows: Where μ1 is the stop signal parameter; When f4(t)=0, stop calculating the ammonia injection output value f2(t), and add f2(t) to the initial ammonia input amount.

3. The method according to claim 1, characterized in that In step 3, the control object monitoring module detects whether the nitrogen oxide input value is stable in the following manner: First, the NOx input value change signal f5(t) at the current time t is calculated as follows: Among them, t start is the delayed start time, r(t) and r(t-1) are the nitrogen oxide input values ​​at time t and time t-1 respectively; The nitrogen oxide input value change signal f5(t) is accumulated and its change is detected regularly to obtain the change parameter f6(t) as follows: Among them, t set The time interval for detecting whether ammonia emissions exceed the standard; Then the nitrogen oxide input value is further detected to be stable, which is expressed by the parameter f7(t) as follows: When the parameter f7(t) takes the value of 1, it indicates that the nitrogen oxide input value is stable, and ammonia emissions are monitored at this time.

4. The method according to claim 1 or 3, characterized in that In step 4, the temperature control module detects the amount of ammonia emissions in the following manner: First, the ammonia emission is detected regularly, and the ammonia emission at time t is detected as f8(t), as follows: Where u1 is the detected ammonia emission, t set The time interval for detecting whether ammonia emissions exceed the standard; Secondly, determine whether the ammonia emissions meet the standards and use the parameter f9(t) to represent it: where u target is the limit value of ammonia emission; f9(t) takes a value of 0 to indicate that the standard is met, and a value of 1 to indicate that the standard is not met; The temperature control module sets the adjustment temperature f of the automobile exhaust SCR reaction device 10 (t) are as follows: Where T1 is the temperature value that increases each time when the ammonia emission exceeds the standard; Each time the ammonia emission exceeds the standard, the temperature of the automobile exhaust SCR reaction device is increased to reduce the ammonia emission. Suppose that after N times of temperature control, the temperature of the automobile exhaust SCR reaction device is Where T0 is the initial temperature of the SCR reaction device.

Images