Differential pressure sensor control method for measuring engine aftertreatment differential pressure

Abstract

The present invention relates to the technical field related to sensor measurement and control, and specifically includes a pressure difference sensor control method for measuring the after-treatment pressure difference of an engine, including: connecting an exhaust after-treatment device; monitoring and obtaining an engine operating condition data set; using a blocking probability model to analyze and determine a blocking probability index for the engine operating condition data set, inputting the blocking probability index into a gain amplification control module for matching, obtaining a matching sensing range, calculating with the current sensing range of the pressure difference sensor and the matching sensing range, and determining a signal gain multiple to control the sensing range of the pressure difference sensor. The method solves the technical problem that the use of a fixed gain amplifier cannot adapt to changes in engine operating conditions, resulting in insufficient measurement accuracy of the pressure difference sensor, realizes real-time monitoring of engine operating condition data, sets the sensing range of the pressure difference sensor according to the blocking probability, dynamically adjusts the signal gain multiple of the pressure difference sensor to adapt to measurement requirements under different operating conditions, and improves the technical effect of the accuracy of pressure difference measurement.

Description

Technical Field

[0001] The present invention relates to the technical field related to sensor measurement control, and in particular to a pressure difference sensor control method for measuring engine aftertreatment pressure difference. Background Art

[0002] Engine aftertreatment is an important part of reducing emissions in modern internal combustion engines. As the use time increases, the aftertreatment device will become clogged, resulting in increased exhaust resistance and reduced emission performance, affecting the engine's operating efficiency and emission control effects. Commonly, the exhaust aftertreatment device monitoring system uses a fixed-gain differential pressure sensor to measure the system's pressure difference, but in the face of different operating conditions, there is a lack of an effective dynamic adjustment mechanism, making it difficult to adjust the sensor's gain according to actual operating conditions, which affects the accurate judgment of the clog status.

[0003] In summary, the prior art has a technical problem that the fixed gain amplifier cannot adapt to changes in engine operating conditions, resulting in insufficient measurement accuracy of the differential pressure sensor. Summary of the invention

[0004] The present application provides a pressure difference sensor control method for measuring the engine aftertreatment pressure difference, aiming to solve the technical problem in the prior art that the fixed gain amplifier cannot adapt to changes in engine operating conditions, resulting in insufficient measurement accuracy of the pressure difference sensor.

[0005] In view of the above problems, the technical solution to implement this application is:

[0006] The present application provides a differential pressure sensor control method for measuring an engine aftertreatment differential pressure, wherein the method comprises: connecting an exhaust aftertreatment device of an engine aftertreatment system, wherein the exhaust aftertreatment device is provided with a differential pressure sensor, the differential pressure sensor is connected to a variable gain amplifier, and the variable gain amplifier is used to adjust a control signal of the differential pressure sensor;

[0007] Acquiring an engine operating condition data set by monitoring the real-time operating condition of the engine;

[0008] Analyzing the engine operating condition data set using a blocking probability model to determine a blocking probability index, wherein the blocking probability index represents a probability of blocking of the exhaust gas after-treatment device;

[0009] The blocking probability index is input into the gain amplification control module for matching, a matching sensing range is obtained, and a signal gain multiple is determined by calculating the current sensing range of the differential pressure sensor and the matching sensing range;

[0010] The variable gain amplifier controls the sensing range of the differential pressure sensor through the signal gain multiple.

[0011] In summary, one or more technical solutions provided in the present application solve the technical problem that the use of a fixed gain amplifier cannot adapt to changes in engine operating conditions, resulting in insufficient measurement accuracy of the differential pressure sensor, and realizes real-time monitoring of engine operating condition data, sets the sensing range of the differential pressure sensor according to the blockage probability, and dynamically adjusts the signal gain multiples of the differential pressure sensor to adapt to measurement requirements under different operating conditions, thereby improving the accuracy of differential pressure measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A flow chart of a pressure differential sensor control method for measuring an engine aftertreatment pressure differential is provided for this application;

[0013] Figure 2 A flow chart of generating a label indicating the size of a blockage probability in a pressure difference sensor control method for measuring an engine aftertreatment pressure difference is provided for this application. DETAILED DESCRIPTION

[0014] The present application is described in detail below with reference to the accompanying drawings. Figure 1 As shown, the present application provides a pressure difference sensor control method for measuring the engine aftertreatment pressure difference, wherein the method includes:

[0015] S1: An exhaust after-treatment device connected to an engine after-treatment system, wherein the exhaust after-treatment device is provided with a differential pressure sensor, the differential pressure sensor is connected to a variable gain amplifier, and the variable gain amplifier is used to adjust a control signal of the differential pressure sensor; S2: Acquire an engine operating condition data set by monitoring the real-time operating condition of the engine.

[0016] Specifically, the differential pressure sensor is used to measure the pressure difference between two points in the engine aftertreatment system, and is usually used to monitor the blockage of the filter in the exhaust aftertreatment device; the variable gain amplifier is mostly used to dynamically adjust the output signal gain. In the present application, the variable gain amplifier is used to amplify or reduce the control signal of the differential pressure sensor, and adaptively adjust the amplitude of the output signal to ensure that the control signal is within the measurable range; the engine operating condition data set is a real-time data set obtained by monitoring the operating status of the engine, including information such as engine speed, exhaust temperature, and particulate matter accumulation.

[0017] Execution steps: The differential pressure sensor is installed on the exhaust after-treatment device of the engine after-treatment system to ensure an effective electrical connection with the variable gain amplifier. The exhaust after-treatment device includes filters and other equipment to reduce particulate matter in the exhaust. When the filter is blocked, there will be a difference in the front and rear pressures of the exhaust after-treatment device of the engine after-treatment system. The differential pressure sensor is used to measure this pressure difference and monitor the blockage. The differential pressure sensor transmits the measured differential pressure signal to the variable gain amplifier, and dynamically adjusts the gain according to the strength of the control signal of the differential pressure sensor to ensure that the sensor signal is within an effective measurement range. For example, when the blockage is severe, the differential pressure signal is large, and the amplifier will reduce the gain to avoid overload; when the blockage is slight, the gain is increased to ensure the accuracy of the signal.

[0018] When the engine is started, the real-time working condition of the engine is monitored through the sensor network to collect the engine working condition data set, which includes the working state of the engine (such as speed, exhaust temperature, load condition, etc.) and the accumulation of particulate matter. For example, when the vehicle is driving on a congested road in the city, due to frequent starts and stops, there will be more particulate matter deposited. At this time, the differential pressure sensor will detect a higher pressure difference; on the contrary, when driving at a constant speed on the highway, due to good combustion conditions, less particulate matter is generated, and the pressure difference measured by the differential pressure sensor is smaller. The real-time working condition of the engine is monitored to provide a basis for subsequent analysis.

[0019] S3: Use a blockage probability model to analyze the engine operating condition data set to determine a blockage probability index, which represents the probability of blockage of the exhaust after-treatment device; S4: Input the blockage probability index into the gain amplification control module for matching, obtain the matching sensing range, calculate with the current sensing range of the differential pressure sensor and the matching sensing range, and determine the signal gain multiple; S5: The variable gain amplifier controls the sensing range of the differential pressure sensor through the signal gain multiple.

[0020] Specifically, the blockage probability model is used to analyze engine operating data and predict the possibility of blockage in the exhaust after-treatment device. The blockage probability model obtains the probability of blockage by performing machine learning analysis on engine operating parameters (such as speed, exhaust temperature, particulate matter accumulation, etc.); the gain amplification control module is used to receive the blockage probability index, dynamically adjust the gain range of the signal, and match the sensing range of the differential pressure sensor to decide whether to amplify or reduce the signal, ensuring that the sensor output signal is within a reasonable range; the differential pressure sensor is used to monitor the pressure difference in the exhaust after-treatment device, and is used to determine the blockage condition of the filter or after-treatment equipment. Generally speaking, the larger the pressure difference, the more serious the blockage.

[0021] Execution steps: By real-time monitoring of various operating parameters of the engine, such as engine speed, exhaust temperature, particulate matter accumulation, etc., an engine operating condition data set containing various operating parameters is formed. The engine operating condition data set is sent to the trained congestion probability model for analysis. The congestion probability model uses the previously accumulated learning experience to estimate a congestion probability index based on the current operating condition data. The congestion probability index quantifies the possibility of congestion in the post-processing device. For example, if it is monitored that the engine is in a low-speed running state for a long time and particulate matter accumulates more, the congestion probability model will infer a higher congestion probability.

[0022] The obtained blockage probability index will be input into the gain amplification control module, which selects a matching sensing range from the preset sensing range. The matching sensing range refers to the measurement interval in which the sensor can work effectively. For example, the higher the blockage probability, the sensor's sensing range needs to be appropriately expanded in order to more accurately monitor subtle pressure difference changes; conversely, when the blockage probability is low, a smaller sensing range can be used; the matching sensing range determines the sensitivity of the differential pressure sensor. The gain amplification control module compares the current sensing range with the matching sensing range and calculates the signal gain multiple based on the difference between the two. The signal gain multiple is used to adjust the output signal strength of the sensor. Generally speaking, the larger the gain multiple, the more the sensor's output signal is amplified to adapt to a wider pressure range, ensuring that the sensor can accurately detect blockages.

[0023] Through the calculated signal gain multiple, the variable gain amplifier will adjust the output signal of the differential pressure sensor, adjust its own output signal strength according to this gain multiple, and then control the sensitivity of the differential pressure sensor, ensuring that the differential pressure sensor can accurately measure the pressure difference in the exhaust aftertreatment device to avoid measurement distortion. The amplifier will also ensure that the output signal of the sensor remains within the maximum measurement capability of the sensor. In this way, it can dynamically adapt to different working conditions of the engine, ensure that the differential pressure sensor is always in the best working state, and improve the accuracy and reliability of detection.

[0024] Furthermore, the present application method also includes:

[0025] An engine operating condition training set is obtained, wherein the engine operating condition training set includes engine speed, exhaust temperature, particulate matter accumulation, and a label representing the size of a jam probability; an LSTM network structure is constructed, wherein the LSTM network structure includes multiple LSTM units and is used to capture the dependency between the engine operating condition accumulation and time; a loss function is introduced to train the LSTM network structure with the engine operating condition training set to obtain a converged jam probability model.

[0026] Specifically, the engine operating condition training set includes the engine's operating data, such as engine speed, exhaust temperature, particulate matter accumulation, etc., which reflects the state of the engine under different operating conditions and is used to train the machine learning model so that it can predict the blockage probability of the exhaust after-treatment device; the LSTM network structure (long short-term memory network) is a recursive neural network (RNN) structure that is good at processing sequence data and can capture the dependencies of data that change over time. Through the memory and forgetting mechanism, it can effectively learn and predict long-term dependencies in time series data; the loss function is used to measure the gap between the model output and the actual label. During the model training process, the prediction results of the blockage probability model are made more accurate by minimizing the value of the loss function.

[0027] Execution steps: Collect engine operating data from the engine sensor network, including engine speed, exhaust temperature, particulate matter accumulation and other indicators, to form an engine operating condition training set, which is used to characterize the engine's operating conditions at different time points. Each piece of data also carries a label that characterizes the size of the jam probability; construct an LSTM network structure. The LSTM network consists of multiple LSTM units, and each LSTM unit can remember and forget previous operating data. Through this structure, the LSTM network structure can capture the dependency of engine operating conditions over time, that is, the correlation between the historical performance of engine operating conditions and jam probability; the input of the LSTM network is the engine operating condition data sequence, such as the time series of engine speed or the time series of exhaust temperature. The data point of each time step is input into the LSTM unit, and the LSTM unit makes predictions based on past inputs.

[0028] After the LSTM network is constructed, the LSTM network structure is trained with the engine operating condition training set. The training goal is to let the model learn to accurately predict the jam probability. For this purpose, a loss function is introduced. The loss function is used to measure the error between the model's prediction result and the actual jam probability label. The commonly used loss function is the mean square error (MSE), and the calculation formula is as follows: ,in, is the congestion probability corresponding to the label, is the predicted value of the jam probability model, and n is the number of data samples. After receiving each batch of operating condition data, the LSTM network structure calculates the output jam probability prediction value, and then compares it with the actual jam probability label to calculate the loss value. The back propagation algorithm adjusts the model parameters according to the loss value, so that the loss value gradually decreases and the model prediction becomes more and more accurate. With the continuous training, the prediction accuracy of the LSTM network gradually improves. When the value of the loss function reaches a certain convergence standard, that is, the error of the model is small enough and stable, it means that the jam probability model has been trained. At this time, the converged jam probability model can more accurately predict the jam probability of the engine under different operating conditions.

[0029] Furthermore, if Figure 2 As shown, a label of the size of the congestion probability is generated, and the method of the present application includes:

[0030] Model the filter of the differential pressure sensor to obtain a filtration simulation model; call the filtration simulation model to simulate and measure the material blockage of the upstream pressure interface of the differential pressure sensor to obtain a blockage simulation data set; extract blockage features based on the blockage simulation data set, and output blockage thickness, blockage uniformity and blockage area index; generate a label of the blockage probability size based on the blockage thickness, blockage uniformity and blockage area index.

[0031] Specifically, the differential pressure sensor is usually equipped with a filter to prevent impurities or particles from entering the measuring area of ​​the sensor and affecting the accuracy of the differential pressure sensor. The performance of the filter directly affects the long-term stability of the sensor. The filtration simulation model is used to simulate the working conditions of the filter under different working conditions. The filtration simulation model can be used to understand the efficiency of the filter in blocking particles, the accumulation of particles in the filter, and the possible blockage. The blockage simulation data set refers to a simulation data set generated by the filtration simulation model, which contains the accumulation of particles in the filter under different pressures, flow rates and other conditions. The blockage simulation data set includes information such as the degree of blockage and the location of the blockage. The key indicators of the blockage feature extracted from the blockage simulation data include blockage thickness, uniformity and blockage area, among which thickness refers to the cumulative depth of particles on the filter, uniformity is the balance of particle distribution, and area refers to the filter area covered by particles. The blockage probability label is used to identify the probability of filter blockage under current conditions.

[0032] Execution steps: mathematically model the filter of the differential pressure sensor in order to accurately describe the behavior of the filter under different working conditions. The filtration simulation model usually takes into account the size, shape, mass of the particles, the pore size of the filter, and the pressure change when the airflow passes through the filter; after determining the filtration simulation model, call the filtration simulation model to simulate the material blockage of the upstream pressure interface of the differential pressure sensor. By changing the operating conditions, such as pressure, flow rate and the number of particles, the filtration simulation model can generate a large amount of data. At the same time, the dynamic changes of material blockage are recorded to obtain a blockage simulation data set.

[0033] The blockage simulation data set records the blockage of the filter at different time points and under different conditions. Three key features are extracted from the blockage simulation data set: blockage thickness, blockage uniformity, and blockage area, which are used to better describe the blockage state of the filter. Among them, blockage thickness is used to describe the thickness of particles accumulated on the filter surface; blockage uniformity is used to describe the balance of particle distribution on the filter. If a large number of particles accumulate in a certain area, but not in other areas, the uniformity is low; blockage area is used to describe the percentage of the filter surface area covered by particles; the blockage thickness, blockage uniformity, and blockage area indicators are converted into a probability value to identify the risk of filter blockage. Through the above steps, the filtration simulation model is used to generate blockage simulation data, extract blockage features, and calculate the probability of filter blockage, providing data support for differential pressure sensor control.

[0034] Furthermore, the application method includes:

[0035] The output end of the variable gain amplifier is connected to the input end of the pressure difference sensor, and the output end of the pressure difference sensor is feedback connected to the input end of the variable gain amplifier; when the variable gain amplifier receives the control signal of the pressure difference sensor, it performs signal gain based on the signal gain multiple to determine the gain control signal; the pressure difference sensor performs pressure difference sensing on the exhaust after-treatment device according to the gain control signal.

[0036] Specifically, the differential pressure sensor is used to measure the pressure difference between two points and convert the pressure difference into an electrical signal for analysis. It is widely used to monitor the exhaust system of the engine. The health of the exhaust system is evaluated by measuring the pressure difference in the exhaust pipe, especially the blockage of the filter device; the gain control signal is the signal output by the variable gain amplifier, which is used to control the sensing range of the differential pressure sensor. The gain control signal will automatically adjust according to the current working status to ensure that the differential pressure sensor operates within the optimal sensitivity range.

[0037] Execution steps: The differential pressure sensor and the variable gain amplifier are connected to each other through a signal interface, and the input end of the differential pressure sensor is connected to the output end of the variable gain amplifier, which means that the differential pressure sensor receives a gain control signal from the amplifier, and the output end of the differential pressure sensor is fed back to the input end of the variable gain amplifier, so that the amplifier can dynamically adjust the gain multiple according to the feedback signal of the sensor; when the variable gain amplifier receives the initial signal from the differential pressure sensor, the signal is amplified by the calculated signal gain multiple, and the signal gain multiple is based on the current sensing environment and the range of the differential pressure signal to ensure that the output signal is within an appropriate range. Specifically, a drastic pressure difference fluctuation may occur in the exhaust pipe. For example, when the load of the engine suddenly increases, the variable gain amplifier will detect that the output signal of the sensor has weakened, and then automatically adjust the signal gain multiple (for example, from 1.5 times to 2.5 times) to make the signal reach a level more suitable for processing.

[0038] After calculating the signal gain multiple, the variable gain amplifier converts it into a gain control signal and outputs it to the differential pressure sensor. The gain control signal is the basis for adjusting the working range of the sensor. Through the control signal, the sensor can sensitively detect pressure difference changes under different pressure conditions and adapt to complex working conditions. After the variable gain amplifier sends the gain control signal, the differential pressure sensor will measure and adjust the pressure difference according to the requirements of the control signal. The measurement range and sensitivity of the differential pressure sensor will be adjusted accordingly based on the signal of the variable gain amplifier to ensure that accuracy can be maintained under complex working conditions.

[0039] Furthermore, the jam probability index is input into the gain amplification control module for matching to obtain a matching sensing range. The method of the present application further includes:

[0040] A step control granularity is set, wherein the step control granularity corresponds to a plurality of sensing ranges, and the radius of each sensing range is different; the gain amplification control module determines the range to which the jam probability index belongs based on the step control granularity, and obtains a matching sensing range according to the range corresponding to the jam probability index.

[0041] Specifically, the gain amplification control module is a control system component used to adjust the gain setting of the variable gain amplifier to ensure that the amplification effect of the sensor signal adapts to different working environments, and calculates the required gain setting based on the input blocking probability index to match the working range of the sensor; the blocking probability index reflects the possibility of blockage of the exhaust after-treatment device, which is obtained by analyzing the engine operating condition data set and characterizes the potential blockage situation in the exhaust after-treatment device; the step control granularity refers to the fineness or level between multiple different sensing ranges set in the control system, which are divided in a stepped manner, each step corresponds to a different sensing range, and each range has a different radius.

[0042] Execution steps: Set the step control granularity in the gain amplification control module. The step control granularity refers to the degree of detail of dividing different sensing ranges. The radius of each sensing range is set according to different needs to form a step control strategy. For example, the sensing range is divided into several levels. The first level corresponds to a low jam probability, the second level corresponds to a medium jam probability, and the third level corresponds to a high jam probability. Each level corresponds to a different gain setting. Correspondingly, if the jam probability index falls within the first level range, then the first level sensing range that matches it is selected; if the jam probability rises and exceeds the first level range, it automatically switches to the more sensitive second level sensing range to By analogy, when the engine operating condition changes are monitored, the jam probability index is predicted through the LSTM network structure, and the gain amplification control module inputs the jam probability index. The gain amplification control module will adjust the gain multiple of the differential pressure sensor based on the predicted jam probability index, and match the jam probability index with the predefined sensing range to identify the most suitable gain setting. At the same time, the input jam probability index and the set step control granularity; based on the range to which the jam probability index belongs, the gain amplification control module will calculate and determine a matching sensing range, which will reflect the current gain requirement and be used to adjust the working settings of the sensor to optimize the measurement effect. Through the above steps, it is ensured that the gain amplification control module can accurately adjust the gain setting of the sensor according to the jam probability index, thereby optimizing the monitoring effect of the exhaust aftertreatment device.

[0043] Furthermore, to obtain the initialization gain multiple of the gain amplifier, the method of the present application includes:

[0044] Collect historical signal gain multiples of the gain amplification control module; perform frequency identification based on the historical signal gain multiples, and output a characteristic signal gain multiple, wherein the characteristic signal gain multiple has the highest frequency of occurrence; and arrange the characteristic signal gain multiple as the initialization gain multiple of the gain amplifier.

[0045] Specifically, the initialization gain multiplier of the gain amplifier refers to the initial gain multiplier set when the gain amplifier system is started or reset, which determines the baseline level of signal amplification to ensure that the sensor can correctly amplify the signal during the startup phase; the historical signal gain multiplier, the recorded gain multiplier of the gain amplifier, is used to help determine the most commonly used gain setting in actual applications; the characteristic signal gain multiplier refers to the most frequently occurring gain multiplier determined after frequency analysis of the historical signal gain multiplier, which reflects the most common gain setting in actual use and is usually used as the baseline value during initialization.

[0046] Execution steps: Collect historically recorded signal gain multiple data, usually from the operation records of the gain amplifier control module at different time points, and obtain these historical data through the data acquisition system or log management tool; use statistical analysis tools to perform frequency identification on the collected historical gain multiples. The statistical analysis tool can calculate the frequency of occurrence of each gain multiple value and determine the most frequently occurring gain multiple, which usually involves the use of frequency analysis algorithms such as histogram analysis or frequency distribution analysis; determine the characteristic signal gain multiple from the frequency analysis results, that is, the gain multiple with the highest frequency, and use it as the initialization gain multiple of the gain amplifier to ensure that the variable gain amplifier can provide a reasonable signal amplification level; set the determined characteristic signal gain multiple as the initialization gain multiple of the gain amplifier, which can be achieved through the control panel or configuration file. Through the above steps, ensure that the gain multiple used by the gain amplifier at startup is the optimal setting based on actual operation data.

[0047] Furthermore, the signal gain multiple is determined by calculating the current sensing range of the differential pressure sensor and the matching sensing range, and the calculation formula is as follows:

[0048] ;in, It is the signal gain multiple, which reflects the degree of signal amplification of the sensor and ensures that the signal output by the sensor is within the appropriate measurement range. is the matching sensing range of the differential pressure sensor, is the current sensing range of the differential pressure sensor, , is the differential pressure signal based on the sensor time t output Signal correction factor.

[0049] Furthermore, the application method includes:

[0050] The differential pressure sensor monitors the differential pressure signal of the exhaust aftertreatment device through a sensor connection pipeline, and the differential pressure signal is transmitted to the variable gain amplifier through a sensor interface, and the variable gain amplifier adjusts the signal gain through a communication interface.

[0051] Specifically, the sensor connecting pipeline refers to the pipeline used to connect the differential pressure sensor and the exhaust after-treatment device, ensuring that the differential pressure sensor can accurately measure the pressure difference in the engine after-treatment system; the differential pressure signal refers to the electrical signal output by the differential pressure sensor, which represents the pressure difference between two measuring points and is used to evaluate the performance of the exhaust after-treatment device; the sensor interface is used to connect the differential pressure sensor and the interface of the variable gain amplifier, which is responsible for transmitting the differential pressure signal to the amplifier; the communication interface is used to transmit control instructions and data between the variable gain amplifier and other system components for gain adjustment.

[0052] Execution steps: Monitor the differential pressure signal of the engine aftertreatment system through the differential pressure sensor, and determine the signal gain multiple according to the current sensing range and the matching sensing range, so as to optimize the signal strength and keep the output signal of the sensor within the optimal measurement range. The specific calculation formula is as follows: ,in, It is the signal gain multiple, which reflects the degree of signal amplification of the sensor and ensures that the signal output by the sensor is within the appropriate measurement range. is the matching sensing range of the differential pressure sensor, that is, the sensing range suitable for the current blockage probability, is the current sensing range of the differential pressure sensor, , is the differential pressure signal based on the sensor time t output The signal correction factor is used to adjust the signal to compensate for the drift or error of the sensor. The calculation formula for setting the signal gain multiple ensures that the signal output by the differential pressure sensor is within the appropriate measurement range.

[0053] Execution steps: Install the differential pressure sensor at an appropriate position of the exhaust aftertreatment device, usually between the inlet and outlet of the device, and ensure that the sensor's connecting pipe (such as a hose or a rigid pipe) can correctly connect the measuring point to the sensor; the differential pressure sensor continuously monitors the pressure difference in the exhaust aftertreatment device through the corresponding connecting pipe, and converts the measurement result into an electrical signal; the electrical signal output by the differential pressure sensor is transmitted to the variable gain amplifier through the sensor interface, and the sensor interface can be a cable connection or other signal transmission methods; after receiving the differential pressure signal, the variable gain amplifier receives the gain adjustment command through its communication interface; according to the gain adjustment command, the amplifier adjusts the gain multiple of the signal; during the gain adjustment process, the variable gain amplifier will amplify the differential pressure signal according to the set gain multiple, and the result of the gain adjustment is to optimize the signal strength so that it can be accurately processed by the downstream modules of the engine aftertreatment system (such as a data recording system or a control system) to adapt to the optimal working conditions of the sensor under different working conditions. Through the above steps, the pressure difference of the exhaust aftertreatment device is monitored and adjusted in real time to ensure that the equipment operates in the best state and improve the overall performance and emission control efficiency of the engine.

[0054] In summary, the beneficial effects of the embodiments of the present application are:

[0055] 1. Use variable gain amplifiers to dynamically adjust signal gain based on real-time operating data and jam probability models to improve measurement accuracy.

[0056] 2. Based on the jam probability model and real-time operating data, adjust the sensor’s sensing range to ensure that the sensor is always in the optimal measurement state.

[0057] 3. By calculating and matching the sensing range, ensure that the sensor output signal is within the appropriate measurement range, and improve the overall performance and response speed of the engine aftertreatment system.

[0058] 4. The historical signal gain multiples of the gain amplifier control module are collected; frequency recognition is performed based on the historical signal gain multiples, and the characteristic signal gain multiples are output, and the characteristic signal gain multiples have the highest frequency of occurrence; the characteristic signal gain multiples are used as the initialization gain multiples of the gain amplifier. Through the above steps, it is ensured that the gain multiples used by the gain amplifier at startup are the optimal settings based on the actual operation data.

[0059] In summary, any step can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor, and no unnecessary restrictions are made here.

[0060] Furthermore, the above-mentioned technical scheme only reflects the preferred technical scheme of the technical scheme of the embodiment of the present application. Some changes that may be made to some parts thereof by technicians in this technical field all reflect the novel principles of the embodiment of the present application. Obviously, technicians in this field can make various changes and modifications to the present application without departing from the scope of the present application.

Claims

1. A pressure differential sensor control method for measuring an engine aftertreatment pressure differential, characterized in that: The method comprises: An exhaust gas after-treatment device connected to the engine after-treatment system, wherein the exhaust gas after-treatment device is provided with a differential pressure sensor, the differential pressure sensor is connected to a variable gain amplifier, and the variable gain amplifier is used to adjust a control signal of the differential pressure sensor; Acquiring an engine operating condition data set by monitoring the real-time operating condition of the engine; Analyzing the engine operating condition data set using a blocking probability model to determine a blocking probability index, wherein the blocking probability index represents a probability of blocking of the exhaust gas after-treatment device; The blocking probability index is input into the gain amplification control module for matching, a matching sensing range is obtained, and a signal gain multiple is determined by calculating the current sensing range of the differential pressure sensor and the matching sensing range; The variable gain amplifier controls the sensing range of the differential pressure sensor through the signal gain multiple; The method further comprises: Acquire an engine operating condition training set, wherein the engine operating condition training set includes engine speed, exhaust temperature, particulate matter accumulation, and a label representing a blockage probability; Constructing an LSTM network structure, wherein the LSTM network structure includes a plurality of LSTM units for capturing the dependency between the accumulated engine operating conditions and time; A loss function is introduced to train the LSTM network structure with the engine operating condition training set to obtain a converged jam probability model.

2. The method according to claim 1, characterized in that Methods for generating labels of jam probability magnitude include: Modeling the filter of the differential pressure sensor to obtain a filtration simulation model; Calling the filtration simulation model to simulate and measure the material blockage of the upstream pressure interface of the differential pressure sensor to obtain a blockage simulation data set; Extract blockage features according to the blockage simulation data set, and output blockage thickness, blockage uniformity, and blockage area index; A label of the size of the blockage probability is generated according to the blockage thickness, blockage uniformity and blockage area index.

3. The method according to claim 1, characterized in that The output end of the variable gain amplifier is connected to the input end of the differential pressure sensor, and the output end of the differential pressure sensor is feedback-connected to the input end of the variable gain amplifier; When the variable gain amplifier receives the control signal of the differential pressure sensor, it performs signal gain based on the signal gain multiple to determine a gain control signal; The differential pressure sensor performs differential pressure sensing on the exhaust gas after-treatment device according to the gain control signal.

4. The method according to claim 1, characterized in that The jam probability index is input into a gain amplification control module for matching to obtain a matching sensing range, and the method further includes: Setting a step control granularity, wherein the step control granularity corresponds to a plurality of sensing ranges, and the radius of each sensing range is different; The gain amplification control module determines the range to which the jam probability index belongs based on the step control granularity, and acquires a matching sensing range according to the range corresponding to the jam probability index.

5. The method according to claim 1, characterized in that Obtaining the initialization gain multiple of the gain amplifier, the method comprising: Collecting historical signal gain multiples of the gain amplification control module; Perform frequency identification according to the historical signal gain multiple, and output a characteristic signal gain multiple, wherein the characteristic signal gain multiple has the highest occurrence frequency; The characteristic signal gain multiple is set as the initialization gain multiple of the gain amplifier.

6. The method according to claim 1, characterized in that The signal gain multiple is determined by calculating the current sensing range of the differential pressure sensor and the matching sensing range. The calculation formula is as follows: ； in, It is the signal gain multiple, which reflects the degree of signal amplification of the sensor and ensures that the signal output by the sensor is within the appropriate measurement range. is the matching sensing range of the differential pressure sensor, is the current sensing range of the differential pressure sensor, , is the differential pressure signal based on the sensor time t output Signal correction factor.

7. The method according to claim 3, characterized in that The differential pressure sensor monitors the differential pressure signal of the exhaust aftertreatment device through a sensor connection pipeline, and the differential pressure signal is transmitted to the variable gain amplifier through a sensor interface, and the variable gain amplifier adjusts the signal gain through a communication interface.

Images