A method and system for ammonia injection control of a diesel engine SCR system

Abstract

This invention belongs to the technical field of diesel engine SCR system, and discloses a method and system for controlling ammonia injection in a diesel engine SCR system. The method includes: monitoring and identifying the reaction equilibrium point characterizing the state change of the SCR system; in response to the identification of the reaction equilibrium point, collecting nitrogen oxide (NOx) content at multiple analysis nodes within a measurement segment ending at the reaction equilibrium point, and calculating the rate of change of NOx content as analytical data; based on the analytical data, predicting the future NOx content, and determining the predicted equilibrium point accordingly, at which the predicted NOx content reaches a preset NOx consumption rating; and calculating the required ammonia injection rate based on the predicted equilibrium point and the current NOx consumption. This invention also discloses a system for executing the method. This invention overcomes system response delay by establishing a predictive control mechanism and can adjust the system model based on prediction errors, thereby improving NOx conversion efficiency and control stability.

Description

Technical Field

[0001] This invention belongs to the technical field of diesel engine SCR system, specifically relating to a method and system for controlling ammonia injection in a diesel engine SCR system. Background Technology

[0002] With increasingly stringent requirements for environmental protection and air quality, nitrogen oxides (NOx) in the emissions of internal combustion engines such as diesel engines pose a threat to the atmospheric environment and human health. Therefore, selective catalytic reduction (SCR) technology has become the mainstream technology for diesel engine exhaust after-treatment and NOx emission reduction due to its advantages such as high reduction efficiency, mature technology, and good economy.

[0003] The existing SCR system ammonia injection control model is oversimplified and has a hysteresis problem. The control system relies on a static mapping table of a few parameters such as engine speed and load, or simple proportional feedback control. It cannot accurately describe the complex nonlinear relationship of the instantaneous amount of nitrogen oxides generated by the diesel engine under dynamic conditions. At the same time, it fails to fully consider the physical and chemical delays inside the SCR system, such as the injection, atomization, pyrolysis of urea solution, the generation and transmission of ammonia, and the time lag of the catalytic reaction, resulting in a time difference and quantity difference between the issuance of control commands and the actual demand. Furthermore, due to model simplification and response lag, the system may experience insufficient injection volume when facing rapidly changing nitrogen oxide emissions, resulting in some nitrogen oxides being discharged directly without reduction, leading to excessive emissions. Conversely, the excessive injection strategy adopted to ensure emission compliance can cause the "ammonia escape" problem, where unreacted ammonia enters the atmosphere with the exhaust gas, forming secondary pollution.

[0004] In view of this, the present invention proposes a method and system for controlling ammonia injection in a diesel engine SCR system. Summary of the Invention

[0005] The purpose of this invention is to provide a method for controlling ammonia injection in a diesel engine SCR system, which can dynamically adjust the ammonia injection quantity so that it can always match the nitrogen oxide content in the exhaust gas.

[0006] The technical solution adopted by this invention is as follows: A method for controlling ammonia injection in a diesel engine SCR system, wherein when a reaction equilibrium point characterizing the state change of the SCR system is detected, predictive injection quantity adjustment is performed, the adjustment including: Based on the nitrogen oxide content collected at multiple analytical nodes within the time period before the reaction equilibrium point, the future nitrogen oxide content is predicted; based on the predicted nitrogen oxide content, the predicted equilibrium point is determined, at which the predicted nitrogen oxide content reaches the preset nitrogen oxide consumption rating. Based on the predicted equilibrium point and the current nitrogen oxide consumption, calculate the required ammonia injection rate; compare the current ammonia injection rate of the SCR system with the required ammonia injection rate, and control the amount of liquid urea injected into the SCR system accordingly. The method involves collecting nitrogen oxide content at multiple analytical nodes within a time period prior to the reaction equilibrium point, including: defining a time period of preset duration ending at the time point corresponding to the reaction equilibrium point as a measurement segment; collecting data at least two analytical nodes at preset time intervals within the measurement segment, and obtaining the nitrogen oxide content at each analytical node.

[0007] Preferably, the reaction equilibrium point is the time point at which the nitrogen oxide consumption first exceeds the preset nitrogen oxide consumption rating; and after this time point, the rate of change of nitrogen oxide content is positive.

[0008] Preferably, determining the reaction equilibrium point includes: collecting the nitrogen oxide consumption at various time points within a preset initial stage.

[0009] Preferably, before predicting future nitrogen oxide content, the method further includes: calculating the rate of change of nitrogen oxide content between analysis nodes based on the nitrogen oxide content at each analysis node, and using the rate of change of nitrogen oxide content as analytical data for prediction.

[0010] Preferably, after predicting the future nitrogen oxide content, the method further includes: inputting the analytical data used for prediction into a preset system model to output the predicted nitrogen oxide content corresponding to the future prediction node; At the time point corresponding to the future prediction node, the actual nitrogen oxide content is collected; the difference between the predicted nitrogen oxide content and the actual nitrogen oxide content is calculated to obtain the current error; and based on the current error, the system model is adjusted to predict the nitrogen oxide content of the next future prediction node.

[0011] Preferably, calculating the required ammonia injection amount includes: calculating the time difference between the predicted equilibrium point and the current time point, and defining the time difference as the predicted execution duration; and calculating a correction value for the ammonia injection amount based on the predicted execution duration, and applying the correction value to determine the required ammonia injection amount.

[0012] Preferably, controlling the amount of liquid urea injected into the SCR system includes: increasing the amount of liquid urea injected into the SCR system if the current ammonia injection amount is less than the required ammonia injection amount; and decreasing the amount of liquid urea injected into the SCR system if the current ammonia injection amount is greater than the required ammonia injection amount.

[0013] A diesel engine SCR system ammonia injection control system, comprising: The SCR status monitoring module is used to monitor the nitrogen oxide content in the SCR system chamber and identify the reaction equilibrium point that characterizes the state changes of the SCR system. The ammonia injection rate prediction module is used to predict the future nitrogen oxide content based on the nitrogen oxide content collected at multiple analysis nodes before the reaction equilibrium point in response to the SCR status monitoring module's identification of the reaction equilibrium point. Based on this, the module determines the predicted equilibrium point and calculates the required ammonia injection rate. The urea injection control module is used to control the urea injection system connected to the system to adjust the injection volume of liquid urea based on the required ammonia injection volume calculated by the ammonia injection volume prediction module. The prediction model correction module is used to obtain the predicted nitrogen oxide content output by the ammonia injection quantity prediction module and the subsequent actual nitrogen oxide content from the SCR status monitoring module. Based on the difference between the predicted nitrogen oxide content and the actual nitrogen oxide content, the system model used to predict the nitrogen oxide content is adjusted.

[0014] Preferably, the SCR state monitoring module identifies the reaction equilibrium point in the following manner: The point at which nitrogen oxide consumption first exceeds the preset nitrogen oxide consumption value is defined as the reaction equilibrium point, and the rate of change of nitrogen oxide content is positive after this point.

[0015] Preferably, the ammonia injection rate prediction module is further configured to: calculate the rate of change of nitrogen oxide content between analysis nodes based on the nitrogen oxide content at multiple analysis nodes before predicting the nitrogen oxide content, and use the rate of change of nitrogen oxide content as analytical data for prediction.

[0016] Beneficial effects

[0017] 1. This invention calibrates the measurement section by using the reaction equilibrium point as the boundary, collects and analyzes the nitrogen oxide content at the analysis nodes within the measurement section, and calculates the rate of change of nitrogen oxide content. Based on this rate of change, it predicts the future equilibrium point where the nitrogen oxide consumption will reach the rated value, and determines the required ammonia injection amount. By establishing a predictive control mechanism based on the rate of change, it anticipates the future trend of nitrogen oxide content change rather than responding to the current value, and adjusts the ammonia injection amount in advance. This overcomes the inherent response delay of the SCR system, enabling the ammonia supply to be accurately and timely matched with the nitrogen oxide generation rate.

[0018] 2. After outputting the predicted nitrogen oxide content, this invention collects the actual nitrogen oxide content at future prediction nodes, calculates the current error between the predicted and actual nitrogen oxide content, and corrects the system model based on this current error. Through closed-loop adjustment of error feedback and correction, the system model is endowed with the ability to self-optimize, enabling the control system to automatically compensate for model deviations caused by catalyst aging and changes in operating conditions, ensuring the long-term accuracy and stability of ammonia injection quantity calculation.

[0019] 3. This invention monitors nitrogen oxide consumption in real time and dynamically determines the reaction equilibrium point for dividing the measurement and adjustment segments by the time point when nitrogen oxide consumption first exceeds the temperature-related nitrogen oxide consumption rating. Based on the actual catalytic activity of the system, it selects the precise timing for start-up prediction and adjustment, avoiding intervention in the ineffective range of low load or inactive catalyst, so that the control strategy is executed only during the necessary and effective window period, thereby improving the targeting and response speed of the control. Attached Figure Description

[0020] Figure 1 This is a flowchart of the implementation method of the present invention; Figure 2 This is a block diagram of the implementation system of the present invention. Detailed Implementation

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Example 1

[0023] Please see Figure 1 This embodiment provides a method for controlling ammonia injection in a diesel engine SCR system. Based on the prediction of nitrogen oxide consumption trends, the required ammonia injection quantity is dynamically calculated and matched to ensure that the nitrogen oxide conversion efficiency and injection optimization of the SCR system can be consistent under various operating conditions.

[0024] The method includes the following steps: The process involves obtaining and setting the reaction equilibrium point: The reaction equilibrium point is the critical transition moment in the SCR system from ammonia supply sufficient or balanced to insufficient. This point is precisely captured by establishing the rated nitrogen oxide consumption range as the evaluation benchmark. This process is achieved by collecting temperature control values ​​that are strongly correlated with nitrogen oxide conversion efficiency, such as the exhaust temperature at the SCR catalyst inlet, and inputting the target temperature for calculation. The collected temperature control values ​​are submitted to a preset calculation rule or corresponding table that reflects the thermodynamic response relationship to calculate the target temperature. Then, the target temperature range is defined around the target temperature and combined with a preset offset value; this accommodates the inherent error of the sensor and normal temperature fluctuations in the operating conditions. When the current actual temperature of the SCR catalyst carrier is detected to fall within this target temperature range, it indicates that the system is working in the preset optimal efficiency range, and the measured nitrogen oxide consumption data at this time has high reliability; then, the nitrogen oxide consumption measured under this state is determined as the rated value, and the rated nitrogen oxide consumption range for subsequent judgment is set based on this rated value. After obtaining this range, data on nitrogen oxide consumption downstream of the reactor is continuously collected. The moment when consumption first exceeds the rated range and the downstream nitrogen oxide sensor detects a change in concentration rate from negative to positive is determined as the reaction equilibrium point. Simply put, this means that the ammonia supply is no longer sufficient to completely reduce the incoming nitrogen oxides, causing escape to increase, and the system transitions from an ammonia excess state to an ammonia deficiency state.

[0025] Furthermore, based on the reaction equilibrium point, time intervals are divided. A preset time period before the equilibrium point is defined as the measurement period, and historical data of nitrogen oxides before the critical state are collected. A preset time period after the equilibrium point is defined as the adjustment period, which serves as the control window for predictive injection.

[0026] Furthermore, data sampling is performed within the measurement segment, which is then evenly divided according to a preset period. An analysis node is set at the end of each period to extract the nitrogen oxide content at that moment, forming a time-series data sequence.

[0027] Furthermore, based on the collected nitrogen oxide content sequence, the rate of change of nitrogen oxide content is calculated to quantify the trend of nitrogen oxide content change; the difference of nitrogen oxide content between adjacent nodes is divided by a fixed time interval to obtain the rate of change within that time period; by traversing all adjacent nodes, a set of data reflecting the fluctuation trend of nitrogen oxide content can be obtained.

[0028] Furthermore, predictions and injection volume calculations are performed based on the analyzed data. The predicted nitrogen oxide content corresponding to the future prediction node is calculated based on the most recent rate of change in the analyzed data. When the future prediction node actually arrives in time, the actual nitrogen oxide content at that moment is collected, and the difference between the predicted nitrogen oxide content and the actual nitrogen oxide content is calculated to obtain the current error. This current error will be used as a feedback correction amount to adjust the benchmark for the next prediction calculation. The prediction process is self-corrected by multiplying the current error by a preset correction factor to adjust the initial value or rate of change used for the next calculation. Using this dynamically corrected prediction process, the prediction is continuously pushed forward until a future time point is determined. At this time point, the predicted nitrogen oxide content will reach the preset nitrogen oxide consumption rating set in the aforementioned steps. This time point is defined as the prediction equilibrium point. Then, the time difference between the predicted equilibrium point and the current time point is calculated, and this time difference is defined as the predicted execution time. Based on the predicted execution time and other operating parameters such as the current catalyst temperature and engine load, the total amount of ammonia required to re-reach the equilibrium state is determined by a preset calculation rule, and the required ammonia injection amount is calculated accordingly. The current ammonia injection rate of the SCR system is collected as the current measurement value and compared with the calculated required ammonia injection rate. If the current measurement value is less than the required ammonia injection rate, an instruction is generated to increase the amount of liquid urea injected into the SCR system; otherwise, an instruction is generated to reduce the injection rate. This instruction is transmitted to the urea pump actuator to perform closed-loop regulation of the ammonia injection.

[0029] Example 2

[0030] Please see Figure 2 This embodiment provides an ammonia injection control system for a diesel engine SCR system. By predicting the state of the SCR system, the system can perform predictive injection quantity adjustment in advance before the catalyst ammonia storage reaches saturation and the risk of nitrogen oxide escape occurs, thereby achieving precise closed-loop regulation of ammonia injection.

[0031] In its specific implementation, the system can be integrated into the engine control unit (ECU) of a diesel engine or exist as an independent control unit. It interacts and controls data with sensors in the SCR system, such as upstream and downstream nitrogen oxide sensors, temperature sensors, and actuators such as urea injection pumps and nozzles, through the controller area network (CAN) bus. The system includes the following modules: The SCR status monitoring module continuously monitors the operating status of the SCR system and identifies the reaction equilibrium point that characterizes the key changes in the SCR system status. In the specific execution process, it communicates with nitrogen oxide sensors deployed upstream and downstream of the SCR system catalyst to obtain the nitrogen oxide content in the SCR system chamber in real time. Based on the nitrogen oxide content data of the upstream and downstream, it calculates the nitrogen oxide consumption at the current time point. During the preset initial stage, the nitrogen oxide consumption at each time point in a continuous period is continuously collected to identify the reaction equilibrium point. By comparing the real-time calculated nitrogen oxide consumption with the preset nitrogen oxide consumption value, and also calculating the rate of change of nitrogen oxide content, when the nitrogen oxide consumption is detected to exceed the nitrogen oxide consumption value for the first time, and immediately after that time point, the rate of change of nitrogen oxide content changes from negative or zero to positive, the time point is identified and marked as the reaction equilibrium point. The appearance of this reaction equilibrium point signifies that the catalytic reaction state of the SCR system has transitioned from a highly efficient state to a critical state where it may be about to saturate or its efficiency may decrease. At this point, the module will trigger the ammonia injection quantity prediction module to start predictive adjustment.

[0032] The ammonia injection rate prediction module is activated after receiving a reaction equilibrium point identification signal from the SCR status monitoring module; then it predicts the future system state based on historical data and calculates the required ammonia injection rate to achieve optimal control accordingly. Using the time point corresponding to the reaction equilibrium point as the endpoint, trace back a preset time period and mark this time period as the measurement segment. Within this measurement segment, select at least two analytical nodes at preset time intervals and obtain the nitrogen oxide content at each analytical node. Then, in order to better capture the dynamic trend of the system, the module calculates the rate of change of nitrogen oxide content between analysis nodes based on the nitrogen oxide content at each analysis node, and uses this rate of change as the core analysis data for subsequent prediction. The analysis data is input into a preset system model, which can be a model based on time series analysis, state space model or machine learning, and is trained to simulate the dynamic behavior of the SCR system. After the model runs, it outputs a series of predicted nitrogen oxide contents corresponding to future prediction nodes. Then, it analyzes the prediction sequence to determine the prediction equilibrium point. The prediction equilibrium point is defined as the time point at which the predicted nitrogen oxide content will reach the preset nitrogen oxide consumption limit. Finally, the time difference between the current time point and the predicted equilibrium point is calculated and defined as the predicted execution time. Based on this predicted execution time and the current nitrogen oxide consumption, the correction value for the ammonia injection rate is calculated and applied to determine the final required ammonia injection rate. This required ammonia injection rate represents the amount of ammonia injection that needs to be adjusted in advance to maintain the system's optimal operating state when the predicted equilibrium point arrives in the future.

[0033] The prediction model correction module works in conjunction with the ammonia injection quantity prediction module and the SCR status monitoring module to form an adaptive correction loop, continuously optimizing the prediction accuracy of the system model. During a work cycle, the predicted nitrogen oxide content for a certain future prediction node is obtained from the ammonia injection prediction module; when the time reaches the time point corresponding to the future prediction node, the actual nitrogen oxide content collected at that time point is obtained from the SCR status monitoring module. Then, the difference between the predicted nitrogen oxide content and the actual nitrogen oxide content is calculated to obtain the current error. Based on this current error, the internal parameters of the system model are adjusted by adjusting the algorithm. The system continuously self-corrects based on actual operating data to make its prediction of the nitrogen oxide content of the next future prediction node closer to the real situation, thereby improving the robustness and accuracy of the entire predictive control system.

[0034] The urea injection control module is the bridge connecting predictive decision-making and physical execution. It is responsible for translating the calculated injection strategy into actual control of the urea injection system. It receives the required ammonia injection amount calculated by the ammonia injection amount prediction module and compares the required ammonia injection amount with the current ammonia injection amount of the SCR system. If the comparison results show that the current ammonia injection rate is less than the required ammonia injection rate, it indicates that the ammonia supply needs to be increased in advance to cope with the upcoming changes in operating conditions. At this time, the module will send a command to the urea injection system connected to the system to increase the amount of liquid urea injected into the SCR system. Conversely, if the current ammonia injection rate is greater than the required ammonia injection rate, the module will send an instruction to reduce the liquid urea injection rate; through predictive adjustment, it ensures that the ammonia supply and catalyst demand remain optimally matched throughout the dynamic process.

[0035] Through the coordinated work of the above modules, the system in this embodiment captures the critical point of SCR system state change by constructing a complete feedforward-feedback adaptive control closed loop, and intervenes in advance based on future predictions; thus avoiding excessive nitrogen oxide emissions or unnecessary ammonia escape caused by response delay.

Claims

1. A method for controlling ammonia injection in a diesel engine SCR system, characterized in that, When a reaction equilibrium point characterizing the state change of the SCR system is detected, predictive injection rate adjustment is performed, including: Based on the nitrogen oxide content collected at multiple analytical nodes within the time period before the reaction equilibrium point, the future nitrogen oxide content is predicted; based on the predicted nitrogen oxide content, the predicted equilibrium point is determined, at which the predicted nitrogen oxide content reaches the preset nitrogen oxide consumption rating. Based on the predicted equilibrium point and the current nitrogen oxide consumption, calculate the required ammonia injection rate; compare the current ammonia injection rate of the SCR system with the required ammonia injection rate, and control the amount of liquid urea injected into the SCR system accordingly. The method involves collecting nitrogen oxide content at multiple analytical nodes within a time period prior to the reaction equilibrium point, including: defining a time period of preset duration ending at the time point corresponding to the reaction equilibrium point as a measurement segment; collecting data at least two analytical nodes at preset time intervals within the measurement segment, and obtaining the nitrogen oxide content at each analytical node.

2. The ammonia injection control method for a diesel engine SCR system according to claim 1, characterized in that, The reaction equilibrium point is the time point at which the nitrogen oxide consumption first exceeds the preset nitrogen oxide consumption rating; and after this time point, the rate of change of nitrogen oxide content is positive.

3. A method for controlling ammonia injection in a diesel engine SCR system according to claim 2, characterized in that, Determining the reaction equilibrium point involves collecting nitrogen oxide consumption data at various time points within a predetermined initial phase.

4. A method for controlling ammonia injection in a diesel engine SCR system according to claim 1, characterized in that, Before forecasting future nitrogen oxide levels, further steps include: Based on the nitrogen oxide content at each analysis node, the rate of change of nitrogen oxide content between analysis nodes is calculated, and the rate of change of nitrogen oxide content is used as analytical data for prediction.

5. A method for controlling ammonia injection in a diesel engine SCR system according to claim 1, characterized in that, Following the prediction of future nitrogen oxide levels, further steps include: The analytical data used for prediction is input into a preset system model to output the predicted nitrogen oxide content corresponding to the future prediction node; at the time point corresponding to the future prediction node, the actual nitrogen oxide content is collected. The difference between the predicted nitrogen oxide content and the actual nitrogen oxide content is calculated to obtain the current error; and based on the current error, the system model is adjusted to predict the nitrogen oxide content of the next future prediction node.

6. A method for controlling ammonia injection in a diesel engine SCR system according to claim 1, characterized in that, The calculation of the required ammonia injection amount includes: calculating the time difference between the predicted equilibrium point and the current time point, and defining the time difference as the predicted execution duration; and based on the predicted execution duration, calculating the correction value of the ammonia injection amount, and applying the correction value to determine the required ammonia injection amount.

7. A method for controlling ammonia injection in a diesel engine SCR system according to claim 1, characterized in that, Controlling the amount of liquid urea injected into the SCR system includes: If the current ammonia injection rate is less than the required ammonia injection rate, the amount of liquid urea injected into the SCR system is increased; and if the current ammonia injection rate is greater than the required ammonia injection rate, the amount of liquid urea injected into the SCR system is decreased.

8. An ammonia injection control system for a diesel engine SCR system, characterized in that, include: The SCR status monitoring module is used to monitor the nitrogen oxide content in the SCR system chamber and identify the reaction equilibrium point that characterizes the state changes of the SCR system. The ammonia injection rate prediction module is used to predict the future nitrogen oxide content based on the nitrogen oxide content collected at multiple analysis nodes before the reaction equilibrium point in response to the SCR status monitoring module's identification of the reaction equilibrium point. Based on this, the module determines the predicted equilibrium point and calculates the required ammonia injection rate. The urea injection control module is used to control the urea injection system connected to the system to adjust the injection volume of liquid urea based on the required ammonia injection volume calculated by the ammonia injection volume prediction module. The prediction model correction module is used to obtain the predicted nitrogen oxide content output by the ammonia injection quantity prediction module and the subsequent actual nitrogen oxide content from the SCR status monitoring module. Based on the difference between the predicted nitrogen oxide content and the actual nitrogen oxide content, the system model used to predict the nitrogen oxide content is adjusted.

9. A diesel engine SCR system ammonia injection control system according to claim 8, characterized in that, The SCR status monitoring module identifies the reaction equilibrium point in the following ways: The point at which nitrogen oxide consumption first exceeds the preset nitrogen oxide consumption value is defined as the reaction equilibrium point, and the rate of change of nitrogen oxide content is positive after this point.

10. A diesel engine SCR system ammonia injection control system according to claim 8, characterized in that, The ammonia injection rate prediction module is further used to: calculate the rate of change of nitrogen oxide content between analysis nodes based on the nitrogen oxide content at multiple analysis nodes before predicting the nitrogen oxide content, and use the rate of change of nitrogen oxide content as the analysis data for prediction.

Images