Urea solution concentration automatic adjustment and control system
By combining a multimodal concentration detection and dynamic proportioning execution module with an intelligent control unit, the problems of ammonia storage imbalance and measurement reliability in the urea solution supply system under complex operating conditions were solved, achieving high-precision, adaptive urea solution concentration control and improving NOx conversion efficiency and system reliability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUANENG POWER INT INC DALIAN POWER PLANT
- Filing Date
- 2026-03-11
- Publication Date
- 2026-05-12
AI Technical Summary
The existing urea solution supply system is difficult to adapt to the complex and ever-changing operating conditions of the engine, resulting in an imbalance in ammonia storage, insufficient measurement reliability, and traditional detection methods are easily interfered with. The control system lacks adaptive and learning capabilities, leading to incomplete NOx conversion or excessive ammonia leakage, and high maintenance costs.
By employing a multimodal concentration detection module combined with ultrasonic phase difference and optical refraction detection, a dynamic proportioning execution module, and an intelligent control unit, high-precision automatic adjustment of urea solution concentration is achieved. This includes bidirectional high-precision metering supply, online real-time dynamic mixing, and end-to-end intelligent anti-crystallization heating. Combined with an ammonia storage model, fuzzy PID control, and online self-learning algorithm, it supports human-computer interaction and remote monitoring.
It achieves high-precision, adaptive urea solution concentration detection and control, prevents ammonia storage imbalance, improves NOx conversion efficiency, reduces ammonia leakage, lowers system maintenance costs, and enhances system reliability and energy efficiency.
Smart Images
Figure CN122014388A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of diesel engine exhaust aftertreatment technology, specifically relating to an automatic adjustment and control system for urea solution concentration. Background Technology
[0002] With increasingly stringent regulations on diesel engine emissions, selective catalytic reduction (SCR) technology has become the mainstream method for reducing nitrogen oxide (NOx) emissions. Urea solution (typically 32.5% AdBlue) serves as the reducing agent in the SCR system, and its concentration precision directly affects NOx conversion efficiency, ammonia leakage, and system operational reliability.
[0003] Existing urea solution supply systems mostly use a fixed concentration supply method, which is difficult to adapt to the complex and ever-changing operating conditions of engines. This can easily lead to an imbalance in ammonia storage, resulting in incomplete NOx conversion or excessive ammonia leakage. Furthermore, traditional concentration detection methods (such as single ultrasonic or refractive index methods) are easily affected by temperature, bubbles, and contaminants, resulting in insufficient measurement reliability. The proportioning system is prone to urea crystallization and blockage in low-temperature environments, leading to high maintenance costs. The control system lacks adaptive and learning capabilities and cannot dynamically optimize concentration settings and control parameters according to real-time operating conditions.
[0004] Therefore, there is an urgent need for an automatic adjustment and control system for urea solution concentration that can achieve high precision, high reliability, and adaptive adjustment. Summary of the Invention
[0005] This application provides an automatic adjustment and control system for urea solution concentration, which aims to solve the problems of ammonia storage imbalance and insufficient measurement reliability caused by existing technologies.
[0006] In a first aspect, an automatic adjustment and control system for urea solution concentration includes:
[0007] The multimodal concentration detection module detects the concentration of urea solution in real time. It includes an ultrasonic phase difference detection submodule, an optical refraction detection submodule, and a multimodal information fusion and processing unit for fusing the detection results of the two.
[0008] The dynamic proportioning execution module accurately prepares urea solution according to control instructions. It includes a bidirectional high-precision metering supply submodule that controls the supply of urea stock solution and deionized water respectively, an online real-time dynamic mixing submodule, and a full-link intelligent anti-crystallization heating submodule.
[0009] The intelligent control unit is signal-connected to the multimodal concentration detection module and the dynamic proportioning execution module, and outputs control commands to the dynamic proportioning execution module based on the ammonia storage model, multi-parameter fuzzy PID control and online self-learning algorithm;
[0010] The human-computer interaction and remote monitoring module is used to provide a local operation interface and remote monitoring functions.
[0011] Optionally, in the multimodal concentration detection module:
[0012] The ultrasonic phase difference detection submodule uses multi-frequency ultrasonic phase difference detection technology to calculate the urea solution concentration by calculating the total phase difference across cycles and combining it with the solution temperature.
[0013] The optical refraction detection submodule uses an optical refractive index sensor based on the critical angle principle to detect the concentration of urea solution;
[0014] The multimodal information fusion and processing unit is configured to evaluate the confidence level of the detection results of the two sub-modules, and perform adaptive weighted fusion based on the confidence level to output the final concentration value.
[0015] Optionally, in the dynamic proportioning execution module:
[0016] The bidirectional high-precision metering supply submodule includes two high-precision metering pumps, which are used for the delivery of urea stock solution and deionized water, respectively, and one-way valves are installed at the inlet and outlet of the pumps.
[0017] The online real-time dynamic mixing submodule is a static mixer, installed downstream of the confluence point of the two liquid streams;
[0018] The end-to-end intelligent anti-crystallization heating submodule includes a device for zoned heating and temperature control of the storage tank, pipelines, metering pump head, and static mixer.
[0019] Optionally, the intelligent control unit includes:
[0020] The adaptive setpoint planner dynamically outputs the target concentration value of urea solution based on the NOx sensor signal downstream of SCR, SCR catalyst temperature, engine operating conditions, and real-time estimated ammonia storage.
[0021] A multi-parameter fuzzy PID real-time controller dynamically adjusts PID parameters or outputs control signals to drive the metering pump based on concentration deviation, deviation change rate, solution temperature, and the original NOx level of the engine exhaust.
[0022] The online self-learner uses a reinforcement learning algorithm, taking engine operating conditions and ammonia storage as states, fine-tuning of concentration setpoint and PID parameters as actions, and NOx conversion efficiency, ammonia leakage, and concentration stability as rewards to continuously optimize the control strategy.
[0023] Optionally, the human-computer interaction and remote monitoring module includes:
[0024] The local human-machine interaction submodule has an industrial-grade touch screen for displaying system status, parameter settings, data trends and alarm information, and is equipped with a three-color status light, a buzzer and an emergency stop button.
[0025] The remote monitoring and cloud platform submodule connects to the cloud platform through an IoT gateway to achieve functions such as equipment asset management, global monitoring, data analysis and early warning, report generation, and remote debugging.
[0026] A multi-level alarm and notification management mechanism categorizes alarms into multiple levels and notifies responsible personnel according to level through local sound and light alarms, cloud platform recording, and mobile phone push notifications, forming a closed loop for alarm handling.
[0027] Optionally, the dynamic proportioning execution module supports two operating modes:
[0028] Mixing mode: The control unit calculates and controls the speed of the two metering pumps to achieve precise mixing based on the target concentration and the real-time detected concentration;
[0029] Flushing mode: Before the system is shut down, the deionized water pump is controlled to run independently to flush the mixer and downstream pipelines.
[0030] Optionally, the adaptive setpoint planner is configured to execute at least one of the following rules:
[0031] Under low-temperature and low-ammonia storage conditions, the output target concentration is higher than the standard value;
[0032] Under high load and high ammonia storage conditions, output the standard target concentration;
[0033] Under transient load reduction and ammonia storage supersaturation conditions, the output target concentration is lower than the standard value;
[0034] When an ammonia leak is detected, a zero concentration command is output and an alarm is triggered.
[0035] In a second aspect, a machine-readable storage medium stores a computer program that, when executed by a processor, implements the control logic of the intelligent control unit.
[0036] Compared with the prior art, this application has at least the following beneficial effects:
[0037] This application utilizes a multimodal concentration detection module that integrates ultrasonic phase difference and optical refractive index detection technologies. By combining confidence assessment and adaptive weighted fusion algorithms, it effectively overcomes the measurement bias of a single sensor under complex working conditions, achieving unambiguous and highly reliable concentration detection across the entire range.
[0038] This application employs a bidirectional high-precision metering pump and a static mixer to achieve precise proportioning and instant mixing of urea stock solution and deionized water; the end-to-end intelligent heating system provides zoned temperature control for the storage tank, pipelines, pump head, and mixer, effectively preventing low-temperature crystallization and ensuring stable operation of the system in harsh environments.
[0039] Based on an ammonia storage model, multi-parameter fuzzy PID control, and online self-learning algorithm, this application enables the system to dynamically adjust the target concentration and control parameters of the urea solution according to engine operating conditions, SCR status, and real-time ammonia storage, thereby achieving a balance between efficient NOx conversion and minimizing ammonia leakage, and improving the overall energy efficiency and environmental performance of the system. Attached Figure Description
[0040] Figure 1 This is a schematic diagram of the module connection of an automatic adjustment and control system for urea solution concentration provided in this application. Detailed Implementation
[0041] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments.
[0042] This application provides an automatic adjustment and control system for urea solution concentration, comprising:
[0043] The multimodal concentration detection module uses ultrasonic phase difference method to detect the sound velocity of urea solution, combined with temperature sensor for temperature compensation, and supplemented by optical refraction sensor for cross-validation and error correction, and finally outputs fused concentration signal to improve detection reliability.
[0044] Specifically, the multimodal concentration detection module includes an ultrasonic phase difference detection submodule, an optical refraction detection submodule, and a multimodal information fusion and processing unit;
[0045] The ultrasonic phase difference detection submodule is based on multi-frequency ultrasonic phase difference detection technology, which aims to solve the phase ambiguity problem that occurs when the concentration changes significantly in traditional phase measurement methods. In the specific implementation process, a measurement cavity with a fixed sound path of 50 mm is set in the urea solution pipeline. A high-frequency piezoelectric ultrasonic transmitting transducer and a receiving transducer are installed on both sides of the cavity, respectively. The signal driving unit generates two high-frequency signals with a small frequency difference and combines the two signals into a difference frequency signal as an excitation signal. The excitation signal drives the transmitting transducer to generate ultrasonic waves. After the ultrasonic waves pass through the urea solution to be tested, they are collected by the receiving transducer. The signal processing unit performs detection and demodulation processing on the received signal and extracts the low-frequency envelope signal.
[0046] By comparing the phase difference of the low-frequency envelope signals of the measurement channel and the reference channel, and combining it with the single-cycle phase difference of the high-frequency carrier, the total phase difference across cycles can be obtained using a specific calculation formula. This total phase difference is directly related to the propagation speed of ultrasound in the solution. By monitoring the solution temperature in real time with a temperature sensor, an accurate relationship model between the urea solution concentration, sound velocity, and temperature is established, thereby calculating an accurate concentration value. This method solves the problem that drastic changes in sound velocity due to concentration changes cause the phase difference to exceed the measurement range, and realizes unambiguous concentration measurement across the entire range.
[0047] The optical refraction detection submodule is based on the principle of optical refraction. It utilizes the physical property that changes in the concentration of urea solution lead to changes in refractive index for measurement. This module employs an industrial-grade optical refractive index sensor based on the critical angle principle. Its detection prism is directly immersed in the flowing urea solution. The sensor's internal light source projects light onto the prism interface, and the intensity distribution of the reflected light is detected by a CCD or photodiode array. Changes in solution concentration cause changes in refractive index, resulting in a shift in the critical angle position, which in turn causes a change in the intensity distribution. Finally, an electrical signal corresponding to the concentration is output. The sensor has a built-in temperature sensor that can compensate for the temperature of the measurement results, effectively eliminating the influence of temperature changes on measurement accuracy. This optical measurement method has the characteristics of fast response speed and is not easily affected by tiny bubbles and suspended particles in the solution, forming a good complement to the ultrasonic measurement method.
[0048] The multimodal information fusion and processing unit, as the core of this module, undertakes the important function of intelligently fusing and making decisions on the data from the two sub-modules. This unit uses an embedded microcontroller as its hardware platform and implements data processing through a sophisticated fusion algorithm. The system first synchronously acquires the concentration calculation results and temperature signals from the ultrasonic and optical sub-modules. Then, it evaluates the confidence level of the measurement values of each sensor. The confidence level of ultrasonic measurement is positively correlated with the signal-to-noise ratio of the received signal. When there are a large number of bubbles or impurities in the solution, the attenuation of the ultrasonic signal will lead to a decrease in confidence level. The confidence level of optical measurement is negatively correlated with the degree of contamination of the probe mirror. The system judges the contamination status of the probe by periodically executing a self-cleaning program and comparing the changes in readings before and after cleaning. Abnormal light intensity signals will also lead to a decrease in confidence level.
[0049] In the adaptive weighted fusion stage, the system performs a weighted average of the measured values based on the confidence levels of each sensor, and finally outputs the fused concentration value. Ideally, the system takes the average of the two sensors to suppress random noise; in non-ideal conditions, the system automatically favors the sensor with higher confidence. The system also continuously compares the measurement differences between the two sensors. When the difference exceeds a preset threshold and persists for a certain period of time, a sensor difference alarm is triggered. Combined with the confidence assessment, a faulty sensor can be preliminarily identified, and relevant information is uploaded to the human-machine interaction module to prompt maintenance. The system finally outputs a highly reliable concentration value verified by fusion and a sensor health status indicator, providing an accurate data foundation for subsequent control.
[0050] The dynamic proportioning execution module uses a high-precision metering pump to control the supply of urea stock solution and deionized water separately, and is equipped with an anti-crystallization heating unit to ensure fluidity in low-temperature environments.
[0051] The dynamic proportioning execution module includes a bidirectional high-precision metering supply submodule, an online real-time dynamic mixing submodule, and a full-link intelligent anti-crystallization heating submodule;
[0052] In the bidirectional high-precision metering supply submodule, the system achieves precise supply of urea stock solution and deionized water through independent delivery lines. This part adopts a ceramic pump head metering pump resistant to urea corrosion. Both the urea stock solution metering pump (P1) and the deionized water metering pump (P2) have high-precision control characteristics, with a resolution of 0.1 mL / min and a repeatability of ±0.5%. They also have a speed feedback function to achieve closed-loop control. The two metering pumps are synchronously driven under the coordination of the control unit to ensure the accuracy of the flow ratio.
[0053] The liquid circuit system uses corrosion-resistant engineering plastic or stainless steel pipes. One-way valves are installed at the inlet and outlet of each pump to prevent backflow. Manual or electromagnetic isolation valves are also installed at key locations in the system for easy maintenance. The urea raw material tank and deionized water tank are equipped with liquid level monitoring functions. The deionized water tank is also equipped with a conductivity sensor to monitor water purity in real time. When the conductivity exceeds the set range, a replacement prompt is issued.
[0054] The system supports two operating modes: a mixing mode, in which the control unit dynamically adjusts the speed of the two metering pumps according to the target concentration and the real-time detected concentration using a PID algorithm to ensure that the proportion of urea stock solution flow in the total flow is consistent with the target concentration; and a flushing mode, in which the system automatically switches valves before shutdown and only starts the deionized water pump to flush the mixer and downstream pipelines with pure water, effectively preventing blockages caused by urea crystallization.
[0055] The online real-time dynamic mixing submodule uses a static mixer to achieve uniform mixing of the solution. The mixer consists of a series of fixed spiral blades with no moving parts. When urea stock solution and deionized water are pumped in simultaneously, the fluid is repeatedly divided, rotated and recombined as it passes through these spiral blades, achieving uniform mixing at the molecular level in a short time. This eliminates the lag problem of traditional stirring methods and ensures the consistency of mixing. At the same time, since no dynamic sealing structure is required, the reliability and maintenance-free nature of the system are improved. The mixer is directly installed downstream of the metering pump manifold, allowing the mixed solution to immediately enter the concentration detection module, forming a rapid closed-loop feedback.
[0056] The end-to-end intelligent anti-crystallization heating submodule is responsible for ensuring the reliable operation of the system under various ambient temperatures. This module adopts the strategy of "heating on demand and zoned management" to achieve a balance between energy consumption and antifreeze effect. The storage tank uses enclosed silicone heating pads and is equipped with an insulation layer. When the ambient temperature is below 5℃, the system automatically starts the preheating function to maintain the liquid temperature within the range of 10-15℃. All pipelines transporting urea solution are covered with self-regulating electric heating tapes, whose power automatically decreases as the temperature rises to prevent overheating.
[0057] Special heating devices are installed in key parts, and thermocouple heating elements are integrated into the metering pump head to achieve precise temperature control and ensure smooth pumping; the outer shell of the static mixer and the detection chamber is equipped with a thin film heater to ensure that the mixing and measurement process is carried out within the optimal temperature range of 15-25℃.
[0058] The system collects data through temperature sensors distributed in various locations, establishes a complete temperature field model, and performs independent or collaborative control of each heating unit according to the ambient temperature and system operating status to ensure global temperature balance and achieve energy consumption optimization.
[0059] The intelligent control unit predicts the system’s demand for urea concentration based on the ammonia storage model. It adopts a fuzzy PID control algorithm and dynamically adjusts the ratio according to multiple parameters such as concentration deviation, temperature, and NOx emissions. It has self-learning capabilities and can adapt to the optimal concentration setting under different working conditions.
[0060] The intelligent control unit receives multiple signals from the multimodal detection module, the SCR system CAN bus, and various temperature sensors. Through an intelligent control architecture with a multi-layer structure, a combination of feedforward and feedback, and self-learning capabilities, it sends precise control commands to the dynamic ratio execution module.
[0061] The first-level decision-making of the intelligent control unit is achieved by an adaptive setpoint planner based on an ammonia storage model. This planner dynamically adjusts the target concentration of the urea solution according to the system operating status, rather than using a fixed concentration value of 32.5%. Its model input parameters include: the NOx sensor signal downstream of the SCR reflecting the system purification efficiency, the key state parameter SCR catalyst temperature, engine speed and torque information from the CAN bus (used to predict the original NOx emission trend), and the current ammonia storage amount calculated in real time by the model.
[0062] Regarding the core algorithm, the intelligent control unit estimates changes in ammonia storage through a simplified dynamic model. This model comprehensively considers factors such as urea injection rate, NOx consumption, and ammonia leakage. Based on this model, the system constructs a complete setpoint decision rule: under low-temperature, low-ammonia storage conditions, the target concentration is appropriately increased to 34% to ensure sufficient ammonia molecules participate in the reaction and improve antifreeze performance; under high-load, high-ammonia storage conditions, the standard concentration of 32.5% is maintained to fully utilize the system's existing ammonia storage for efficient conversion; when a transient load reduction leads to ammonia storage oversaturation, the concentration is reduced to 31% to prevent ammonia leakage by reducing urea injection rate; when ammonia leakage is detected, injection is immediately suspended and an alarm signal is issued. Through this mechanism, the system ultimately outputs a target concentration benchmark value that dynamically changes with the operating conditions.
[0063] The second-level decision-making is handled by a multi-parameter fuzzy PID real-time controller. This controller ensures that the system quickly tracks the target value set in the first level in a smooth and stable manner. Its input parameters include the deviation between the target concentration and the actual concentration, the trend of the deviation, the solution temperature, and feedforward factors such as the original NOx emission level of the engine.
[0064] The control system transforms the precise input parameters into fuzzy language descriptions, such as classifying concentration deviations into levels like negative large, negative small, zero, positive small, and positive large; classifying deviation change rates into types like negative fast, negative slow, zero, positive slow, and positive fast; and defining solution temperature as low, medium, and high states. A fuzzy rule base built based on expert experience contains comprehensive control strategies: when the concentration is significantly high and rapidly decreasing, the integral action is automatically reduced to prevent overshoot; when the concentration is slightly high and the solution temperature is low, the proportional action is appropriately enhanced to improve response speed; when a sharp increase in raw NOx is detected, the pump flow output is increased in advance to meet the impending increase in demand. After fuzzy inference, the system transforms the results into precise control quantities, dynamically adjusting PID parameters or directly outputting correction signals, ultimately generating coordinated speed commands for the two metering pumps.
[0065] The third layer of decision-making is completed by an online self-learner based on reinforcement learning. The goal of the self-learner is to find the optimal concentration setpoint and control parameters under different operating conditions. The system state space includes key parameters such as engine speed, torque, catalyst temperature and ammonia storage, while the action space covers the concentration setpoint fine-tuning amount and PID parameter adjustment amount.
[0066] The system has established a comprehensive reward mechanism: a positive reward is given when the NOx conversion efficiency exceeds 95%; a negative penalty is imposed when ammonia leakage is detected to exceed 10 ppm; and corresponding penalty items are also set for situations with excessive concentration fluctuations and excessively frequent actuator actions. By adopting a lightweight reinforcement learning algorithm, the system continuously explores the optimal control strategy under different states during long-term operation and constantly updates the internal experience database. After weeks or months of learning and accumulation, the system can adapt to the working characteristics of specific engines and form a set of tailored optimal control strategies.
[0067] The human-computer interaction and remote monitoring module supports concentration setting, real-time data display, and fault alarm, and can be remotely monitored and exported via a cloud platform.
[0068] The human-computer interaction and remote monitoring module includes a local human-computer interaction submodule and a remote monitoring and cloud platform submodule, wherein:
[0069] The local human-machine interaction submodule provides on-site operators with an intuitive and reliable operating interface. As an auxiliary indicator, the equipment is equipped with a three-color LED status light and a buzzer. Through green normal, yellow warning, and red fault light signals and sound prompts, it realizes long-distance multi-sensory warning of critical alarms. At the same time, an independent emergency stop button is set to ensure that the system power can be quickly cut off in abnormal situations.
[0070] The remote monitoring and cloud platform submodule enables device networking via 4G / 5G or Ethernet industrial IoT gateways, supports standard communication protocols and has the ability to resume transmission after disconnection. The cloud platform provides comprehensive device management functions, creating an independent file for each device and recording basic information such as device number, installation location, and maintenance history. For mobile devices, the platform also provides an electronic fence function, which automatically alarms when the device exceeds the set area. The monitoring center displays the status of all online devices in map form, distinguishing device operating status by color and allowing real-time data viewing. The platform has data analysis capabilities, enabling early warning of sensor performance degradation and metering pump wear by continuously monitoring the stability of concentration control and sensor data fluctuations. At the same time, the platform predicts the replenishment time of urea and deionized water based on the material consumption rate and automatically generates procurement suggestions. When the system encounters complex faults, the platform will automatically push relevant maintenance guidance and handling cases.
[0071] Furthermore, the system establishes a multi-level alarm management mechanism, classifying alarms into three levels based on the severity of the event: Level 1 alarms are minor anomalies, displayed only on the local interface; Level 2 alarms are serious faults, triggering local audible and visual alarms and recording on the cloud platform; Level 3 alarms are emergency situations, in which, in addition to local alarms, a mobile phone notification is sent to the designated responsible person via the cloud platform. The entire alarm handling process forms a complete closed loop, starting from the on-site alarm, through cloud platform recording and push, to the confirmation and handling by maintenance personnel, and finally the system records the handling result and cancels the alarm status, ensuring that every alarm can be handled in a timely and effective manner.
[0072] A machine-readable storage medium storing a computer program that, when executed by a processor, implements the control logic of an intelligent control unit.
[0073] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
Claims
1. An automatic adjustment and control system for urea solution concentration, characterized in that, include: The multimodal concentration detection module detects the concentration of urea solution in real time. It includes an ultrasonic phase difference detection submodule, an optical refraction detection submodule, and a multimodal information fusion and processing unit for fusing the detection results of the two. The dynamic proportioning execution module accurately prepares urea solution according to control instructions. It includes a bidirectional high-precision metering supply submodule that controls the supply of urea stock solution and deionized water respectively, an online real-time dynamic mixing submodule, and a full-link intelligent anti-crystallization heating submodule. The intelligent control unit is signal-connected to the multimodal concentration detection module and the dynamic proportioning execution module, and outputs control commands to the dynamic proportioning execution module based on the ammonia storage model, multi-parameter fuzzy PID control and online self-learning algorithm; The human-computer interaction and remote monitoring module is used to provide a local operation interface and remote monitoring functions.
2. The automatic adjustment and control system for urea solution concentration according to claim 1, characterized in that, In the multimodal concentration detection module: The ultrasonic phase difference detection submodule uses multi-frequency ultrasonic phase difference detection technology to calculate the urea solution concentration by calculating the total phase difference across cycles and combining it with the solution temperature. The optical refraction detection submodule uses an optical refractive index sensor based on the critical angle principle to detect the concentration of urea solution; The multimodal information fusion and processing unit is configured to evaluate the confidence level of the detection results of the two sub-modules, and perform adaptive weighted fusion based on the confidence level to output the final concentration value.
3. The automatic adjustment and control system for urea solution concentration according to claim 1, characterized in that, In the dynamic proportioning execution module: The bidirectional high-precision metering supply submodule includes two high-precision metering pumps, which are used for the delivery of urea stock solution and deionized water, respectively, and one-way valves are installed at the inlet and outlet of the pumps. The online real-time dynamic mixing submodule is a static mixer, installed downstream of the confluence point of the two liquid streams; The end-to-end intelligent anti-crystallization heating submodule includes a device for zoned heating and temperature control of the storage tank, pipelines, metering pump head, and static mixer.
4. The automatic adjustment and control system for urea solution concentration according to claim 1, characterized in that, The intelligent control unit includes: The adaptive setpoint planner dynamically outputs the target concentration value of urea solution based on the NOx sensor signal downstream of SCR, SCR catalyst temperature, engine operating conditions, and real-time estimated ammonia storage. A multi-parameter fuzzy PID real-time controller dynamically adjusts PID parameters or outputs control signals to drive the metering pump based on concentration deviation, deviation change rate, solution temperature, and the original NOx level of the engine exhaust. The online self-learner uses a reinforcement learning algorithm, taking engine operating conditions and ammonia storage as states, fine-tuning of concentration setpoint and PID parameters as actions, and NOx conversion efficiency, ammonia leakage, and concentration stability as rewards to continuously optimize the control strategy.
5. The automatic adjustment and control system for urea solution concentration according to claim 1, characterized in that, The human-computer interaction and remote monitoring module includes: The local human-machine interaction submodule has an industrial-grade touch screen for displaying system status, parameter settings, data trends and alarm information, and is equipped with a three-color status light, a buzzer and an emergency stop button. The remote monitoring and cloud platform submodule connects to the cloud platform through an IoT gateway to achieve functions such as equipment asset management, global monitoring, data analysis and early warning, report generation, and remote debugging. A multi-level alarm and notification management mechanism categorizes alarms into multiple levels and notifies responsible personnel according to level through local sound and light alarms, cloud platform recording, and mobile phone push notifications, forming a closed loop for alarm handling.
6. The automatic adjustment and control system for urea solution concentration according to claim 3, characterized in that, The dynamic proportioning execution module supports two working modes: Mixing mode: The control unit calculates and controls the speed of the two metering pumps to achieve precise mixing based on the target concentration and the real-time detected concentration; Flushing mode: Before the system is shut down, the deionized water pump is controlled to run independently to flush the mixer and downstream pipelines.
7. The automatic adjustment and control system for urea solution concentration according to claim 4, characterized in that, The adaptive setpoint planner is configured to execute at least one of the following rules: Under low-temperature and low-ammonia storage conditions, the output target concentration is higher than the standard value; Under high load and high ammonia storage conditions, output the standard target concentration; Under transient load reduction and ammonia storage supersaturation conditions, the output target concentration is lower than the standard value; When an ammonia leak is detected, a zero concentration command is output and an alarm is triggered.
8. A machine-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the control logic of the intelligent control unit as described in claim 4.