A method for calculating the sulfur content of fuel in a whole vehicle

Abstract

The application discloses a kind of whole vehicle fuel sulfur content calculation methods, belong to fuel sulfur content calculation technical field.The whole vehicle fuel sulfur content calculation method includes the following steps: S1: information collection;S2: training NARX neural network model;S3: report whether SCR catalyst is sulfur poisoning failure;S4: repair to the poisoning SCR catalyst;If driver uses low sulfur content oil, time is very long, possibly all does not cause catalyst poisoning, so adopt the product of sulfur content and time, to express the degree of poisoning comprehensively, using NARX neural network to calculate: the poisoning degree of engine next time, is influenced by its last time poisoning degree, the feature of NARX neural network is to use the output value of last time, as the input of next time participates in calculation, improves the accuracy of whether catalyst is poisoned detection.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of fuel sulfur content calculation, and more particularly to a method for calculating the sulfur content of fuel for a whole vehicle. BACKGROUND

[0002] Some drivers often use non-standard fuel with high sulfur content, which can easily cause poisoning of the SCR catalyst, reducing the conversion efficiency of the SCR. Currently, when the conversion efficiency is low, the sulfur poisoning fault of the SCR catalyst is reported. However, the efficiency of the SCR catalyst itself will change with the engine speed, load, exhaust temperature, etc., and it is difficult to distinguish these changes from the efficiency changes caused by sulfur poisoning through calibration. SUMMARY

[0003] To solve the defects in the above-mentioned technology, the application provides a method for calculating the sulfur content of fuel for a whole vehicle, which is essentially to improve the problem that the efficiency of the SCR catalyst itself will change with the engine speed, load, exhaust temperature, etc., and it is difficult to distinguish these changes from the efficiency changes caused by sulfur poisoning through calibration.

[0004] To achieve the above-mentioned purpose, the application provides a method for calculating the sulfur content of fuel for a whole vehicle, comprising the following steps:

[0005] S1: Collecting information, collecting the nitrogen and oxygen concentrations upstream and downstream of the catalyst, the engine speed, the engine fuel injection amount, the catalyst upstream exhaust temperature, the engine exhaust flow rate, and the urea injection amount through the ECU;

[0006] S2: Training the NARX neural network model, inputting the collected information into the nonlinear autoregressive NARX neural network model through the ECU, obtaining the model, and reading the corresponding inputs of the catalyst upstream nitrogen and oxygen concentrations, catalyst downstream nitrogen and oxygen concentrations, engine speed, engine fuel injection amount, catalyst upstream exhaust temperature, engine exhaust flow rate, and urea injection amount from the ECU, so as to calculate the product of sulfur content and time. The higher the sulfur content of the fuel used, the longer the time, and the greater the product;

[0007] S3: Reporting whether the SCR catalyst is sulfur poisoned, judging whether the SCR catalyst is sulfur poisoned according to whether the product of sulfur content and time reaches the set value of the poisoning of the SCR catalyst;

[0008] S4: Repairing the poisoned SCR catalyst, reminding the driver to repair the SCR catalyst when the sulfur poisoning of the SCR catalyst is reported.

[0009] In the implementation process, the ECU in step S1 is an electronic control unit, which is composed of a microcontroller, a memory, an input / output interface, an analog-to-digital converter, and a shaping and driving large-scale integrated circuit. The voltage operating range of the ECU is 6.5-16V, the operating current is 0.015-0.1A, and the operating temperature is -40℃~80℃. The ECU can withstand vibrations below 1000Hz. The CPU is the core part of the ECU. When the engine is running, it collects signals from various sensors, performs calculations, and converts the results into control signals to control the operation of the controlled object. It also controls the memory, input / output interface, and other external circuits. The program stored in the memory ROM is based on data obtained through precise calculations and extensive experiments. This inherent program continuously compares and calculates with the signals collected from various sensors when the engine is operating.

[0010] In the implementation process, the engine speed in step S1 is transmitted to the speed disc by a speed sensor. The engine speed is related to the number of work times per unit time or the size of the engine's effective power. That is, the engine's effective power changes with the speed. When indicating the size of the engine's effective power, the corresponding speed must also be indicated. The working condition of the engine at the rated power and the rated speed is called the rated working condition. The rated power is not the maximum power that the engine can produce. It is the maximum effective power limit for the engine's use.

[0011] In the implementation process, the engine fuel injection amount in step S1 is controlled by the ECU. The ECU calculates according to the engine intake air volume, speed, throttle opening, water temperature, intake air temperature, and other operating parameters measured by various sensors, and sends an electric pulse to the fuel injector according to the calculation results. By changing the width of each electric pulse, the duration of each fuel injection by the injector is controlled, thereby achieving the purpose of controlling the fuel injection amount. The wider the electric pulse width, the longer the fuel injection duration, and the larger the fuel injection amount. The engine operates under different conditions, and the requirements for the mixture concentration are also different. In particular, under some special conditions such as starting, rapid acceleration, and rapid deceleration, there are special requirements for the mixture concentration. The ECU controls the fuel injection amount in different ways according to the operating conditions measured by the relevant sensors. The control methods of fuel injection amount can be roughly divided into starting control, operating control, fuel cut control, and feedback control.

[0012] In the implementation process, the catalyst in step S1 is an SCR catalyst, which is a selective catalytic reduction method that improves the selectivity of N2 and reduces the consumption of NH3 under catalysis. The principle of SCR is that the reduction reaction can be carried out at a lower temperature under the action of a catalyst. The SCR system is composed of an ammonia supply system, an ammonia / air injection system, a catalytic reaction system, and a control system.

[0013] In the implementation process, the urea injection amount in step S1 is injected into the appropriate amount of urea at the appropriate time to participate in the conversion reaction, which is the key to meeting the NOx emission standard. The working principle of urea injection is mainly divided into pre-priming stage, injection metering stage and emptying stage according to the working stage.

[0014] In the implementation process, the upstream and downstream nitrogen oxide concentrations of the catalyst in step S1 are measured by upstream and downstream nitrogen oxide concentration sensors and transmitted to the ECU. The exhaust gas temperature upstream of the catalyst is measured by an upstream temperature sensor and transmitted to the ECU. The engine exhaust gas flow is measured by an air flow sensor and transmitted to the ECU.

[0015] In the implementation process, the NARX neural network model in step S2 is a nonlinear autoregressive neural network with external input, which is an effective time series prediction technology. First, a time series is divided into training set, validation set and test set according to time sequence, not random allocation. Then, the open-loop NARX is trained using the training set. After training, the open-loop NARX is converted to a closed loop. Then, the prediction performance of the closed-loop NARX on the validation set is checked, and whether to use the NARX network is determined according to the prediction result.

[0016] In the implementation process, the sulfur content in step S3 is the proportion of sulfur in the exhaust gas. The total sulfur content in fuel oil includes elemental sulfur, active sulfides and inactive sulfides. When the sulfur content in liquid fuel is high, SO2 and SO3 are formed after combustion.

[0017] In the implementation process, the sulfur poisoning of the SCR catalyst in step S3 is composed of an alarm, and the ECU controls the alarm.

[0018] Compared with the prior art, the present application has the following advantages:

[0019] When in use, information is collected, nitrogen and oxygen concentration upstream of the catalyst, nitrogen and oxygen concentration downstream of the catalyst, engine speed, engine fuel injection amount, exhaust temperature upstream of the catalyst, engine exhaust flow and urea injection amount are collected by the ECU, the NARX neural network model is trained, the collected information is input into the nonlinear autoregressive NARX neural network model by the ECU, the model is obtained, the corresponding input of nitrogen and oxygen concentration upstream of the catalyst, nitrogen and oxygen concentration downstream of the catalyst, engine speed, engine fuel injection amount, exhaust temperature upstream of the catalyst, engine exhaust flow and urea injection amount is read from the ECU, the product of sulfur content and time is calculated, it is reported whether the SCR catalyst is poisoned by sulfur, whether the product of sulfur content and time reaches the set value of SCR catalyst poisoning is judged, whether the SCR catalyst is poisoned is judged, the poisoned SCR catalyst is repaired, after the SCR catalyst is reported to be poisoned by sulfur, the driver is reminded to repair the SCR catalyst, the upstream nitrogen and oxygen concentration, the downstream nitrogen and oxygen concentration of the catalyst, the engine speed, the engine fuel injection amount, the exhaust temperature upstream of the catalyst, the engine exhaust flow and the urea injection amount are used as inputs, because these quantities can be directly obtained from the existing engine controller, more sensors do not need to be added, the cost will not be increased, and all the influences of sulfur poisoning can also be included, the product of sulfur content and time is used as the output, because if the driver uses high-sulfur-content oil but for a short time, or uses low-sulfur-content oil for a long time, the catalyst may not be poisoned, therefore, the product of sulfur content and time can comprehensively express the degree of poisoning, the NARX neural network is used for calculation: the poisoning degree of the engine at the next moment is affected by the poisoning degree at the last moment, the feature of the NARX neural network is that the output value at the last moment is used as the input for calculation at the next moment, which improves the accuracy of detection of whether the catalyst is poisoned. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 It is a system block diagram of a vehicle fuel sulfur content calculation method provided by the embodiment of the application. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical scheme and advantages of the application more clear, the application is further described in detail below with examples.

[0022] It should be understood that the specific embodiments described herein are only used to explain the application and not to limit the application.

[0023] In the examples, unless otherwise specified, the means used are conventional means in the art.

[0024] Please refer to Figure 1 The application provides a vehicle fuel sulfur content calculation method, which comprises the following steps:

[0025] S1: collecting information, collecting nitrogen and oxygen concentration upstream of the catalyst, nitrogen and oxygen concentration downstream of the catalyst, engine speed, engine fuel injection amount, catalyst upstream exhaust temperature, engine exhaust flow and urea injection amount through ECU, the ECU in step S1 is an electronic control unit, which is composed of a microcontroller, a memory, an input / output interface, an analog-to-digital converter and a shaping, driving large-scale integrated circuit, the voltage working range of the ECU is 6.5-16V, the working current is 0.015-0.1A, the working temperature is -40℃~80℃, and it can withstand vibration below 1000Hz, the CPU is the core part in the ECU, when the engine is running, it collects signals from various sensors, performs calculations, and converts the results of the calculations into control signals to control the work of the controlled object, it also implements control over the memory, input / output interface and other external circuits, the program stored in the memory ROM is written based on the data obtained through precise calculation and a large number of experiments, this inherent program constantly compares and calculates with the signals collected from various sensors when the engine is working, the engine speed in step S1 uses a speed sensor to transmit signals to the speed disc, the engine speed is related to the number of work times per unit time or the size of the engine effective power, that is, the effective power of the engine changes with the speed, when indicating the size of the engine effective power, the corresponding speed must be indicated, the working condition of the engine at the rated power and the rated speed is called the rated working condition, the rated power is not the maximum power that the engine can generate, it is the maximum effective power limit for use according to the purpose of the engine, the engine fuel injection amount in step S1 is controlled by ECU, ECU calculates according to various sensors measured engine intake, speed, throttle opening, water temperature, intake temperature and many other operating parameters, and sends electric pulses to the fuel injector according to the calculation results, the duration of each injection of each fuel injector is controlled by changing the width of each electric pulse, so as to achieve the purpose of controlling the fuel injection amount, the greater the width of the electric pulse, the longer the injection duration, the greater the fuel injection amount, the engine operates under different conditions, and the requirements for the mixture concentration are also different, especially in some special conditions such as starting, rapid acceleration and rapid deceleration, the mixture concentration has special requirements;

[0026] The ECU controls the fuel injection amount in different ways according to the operating conditions measured by the relevant sensors, and the control methods of the fuel injection amount can be roughly divided into start control, operation control, fuel cut control and feedback control. The catalyst in step S1 is an SCR catalyst, and the principle of the SCR catalyst is to improve the selectivity of N2 and reduce the consumption of NH3 under the catalytic action. The principle of the SCR system is to enable the reduction reaction to be carried out at a lower temperature under the action of the catalyst. The SCR system is composed of an ammonia supply system, an ammonia / air injection system, a catalytic reaction system and a control system. The urea injection amount in step S1 is injected into the appropriate amount of urea at the appropriate time to participate in the conversion reaction, which is the key to meeting the NOx emission standard. The working principle of urea injection is mainly divided into pre-priming stage, injection metering stage and emptying stage according to the working stage. The upstream and downstream nitrogen and oxygen concentrations of the catalyst in step S1 are measured by upstream and downstream nitrogen and oxygen concentration sensors and transmitted to the ECU. The upstream exhaust temperature of the catalyst is measured by an upstream temperature sensor and transmitted to the ECU. The engine exhaust flow is measured by an air flow sensor and transmitted to the ECU.

[0027] S2: Train the NARX neural network model. The ECU inputs the collected information into the nonlinear autoregressive NARX neural network model, obtains the model, and reads the upstream and downstream nitrogen and oxygen concentrations of the catalyst, the engine speed, the engine fuel injection amount, the upstream exhaust temperature of the catalyst, the engine exhaust flow and the urea injection amount corresponding to the input from the ECU. The product of sulfur content and time can be calculated. The higher the sulfur content of the fuel used and the longer the time, the greater the product. The NARX neural network model in step S2 is a nonlinear autoregressive neural network with external input, which is an effective time series prediction technology. First, a time series is divided into: training set, validation set and test set. The segmentation method is according to the time sequence, not random allocation. Then, the open-loop NARX is trained using the training set. After training, the open-loop NARX is converted into a closed loop. Then, the prediction performance of the closed-loop NARX on the validation set is checked, and whether to use the NARX network is decided according to the prediction result. If the prediction result meets the requirements, the accuracy on the test set is further checked to prevent overfitting. Otherwise, the NARX is retrained or its topology is optimized until it meets the requirements on the validation set and the test set.

[0028] S3: report whether the SCR catalyst is poisoned by sulfur, according to whether the product of sulfur content and time reaches the set value of poisoning of the SCR catalyst, judge whether the SCR catalyst is poisoned by sulfur, the sulfur content in step S3 is the proportion of sulfur in the exhaust gas, the total sulfur content in the fuel oil including elemental sulfur, active sulfide and non-active sulfide, when the sulfur content in the liquid fuel is high, SO2 and SO3 are formed after combustion, when the exhaust gas reaches the dew point, then form sulfurous acid and sulfuric acid and corrode the machinery, so the sulfur content is required to be less than a certain value, the SCR catalyst poisoning by sulfur in step S3 is reported by an alarm, the ECU controls the alarm, when the SCR catalyst is poisoned by sulfur, the ECU controls the alarm to alarm and remind the staff to maintain the SCR catalyst;

[0029] S4: maintain the poisoned SCR catalyst, when the SCR catalyst is reported to be poisoned by sulfur, remind the driver to maintain the SCR catalyst.

[0030] Specifically, the working principle of the whole vehicle fuel sulfur content calculation method is: when in use, information is collected, the nitrogen and oxygen concentrations upstream and downstream of the catalyst, the engine speed, the engine fuel injection amount, the catalyst upstream exhaust temperature, the engine exhaust flow and the urea injection amount are collected by the ECU, the NARX neural network model is trained, the collected information is input into the nonlinear autoregressive NARX neural network model, the model is obtained, the corresponding inputs of the catalyst upstream nitrogen and oxygen concentration, the catalyst downstream nitrogen and oxygen concentration, the engine speed, the engine fuel injection amount, the catalyst upstream exhaust temperature, the engine exhaust flow and the urea injection amount are read from the ECU, and the product of sulfur content and time can be calculated, whether the SCR catalyst is poisoned by sulfur is reported, according to whether the product of sulfur content and time reaches the set value of poisoning of the SCR catalyst, whether the SCR catalyst is poisoned by sulfur is judged, the poisoned SCR catalyst is maintained, when the SCR catalyst is reported to be poisoned by sulfur, the driver is reminded to maintain the SCR catalyst, the method and system use the upstream nitrogen and oxygen concentration, the catalyst downstream nitrogen and oxygen concentration, the engine speed, the engine fuel injection amount, the catalyst upstream exhaust temperature, the engine exhaust flow and the urea injection amount as inputs, because these quantities can be directly obtained from the existing engine controller, without the need to add more sensors, without increasing the cost, at the same time, all the influences of sulfur poisoning can be included, the product of sulfur content and time is used as the output, because if the driver uses high sulfur content oil, but the time is very short, or uses low sulfur content oil, the time is very long, it may not cause catalyst poisoning, therefore, the product of sulfur content and time is adopted to comprehensively express the degree of poisoning, the NARX neural network is used for calculation: the poisoning degree of the engine at the next moment is affected by the poisoning degree at the last moment, the feature of the NARX neural network is to use the output value at the last moment as the input for calculation at the next moment, which improves the accuracy of detection of whether the catalyst is poisoned.

[0031] The above embodiments only express several implementation manners of the present application, which are more specific and detailed, but cannot be understood as limiting the patent application scope.

[0032] It should be noted that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present application, which all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method of calculating the fuel sulfur content of a vehicle, characterized by, The method and system include the following steps: S1: Information collection, including nitrogen and oxygen concentration upstream of the catalyst, nitrogen and oxygen concentration downstream of the catalyst, engine speed, engine fuel injection quantity, exhaust temperature upstream of the catalyst, engine exhaust flow rate, and urea injection quantity, collected by the ECU. S2: Train the NARX neural network model. The ECU inputs the collected information into the nonlinear autoregressive NARX neural network model. After obtaining the model, the ECU reads the corresponding inputs from the upstream nitrogen and oxygen concentration of the catalyst, the downstream nitrogen and oxygen concentration of the catalyst, the engine speed, the engine fuel injection quantity, the upstream exhaust temperature of the catalyst, the engine exhaust flow rate, and the urea injection quantity. Then, the product of sulfur content and time can be calculated. S3: Report whether the SCR catalyst is poisoned by sulfur. Determine whether the SCR catalyst is poisoned by sulfur based on whether the product of sulfur content and time reaches the set value for SCR catalyst poisoning. S4: Repair the poisoned SCR catalyst. When sulfur poisoning of the SCR catalyst is reported, remind the driver to repair the SCR catalyst.

2. The method of calculating the fuel sulfur content of a complete vehicle according to claim 1, wherein, The ECU in step S1 is an electronic control unit, which consists of a microcontroller, memory, input / output interface, analog-to-digital converter, and large-scale integrated circuits for shaping and driving. The ECU operates in a voltage range of 6.5-16V, a current range of 0.015-0.1A, and a temperature range of -40℃ to 80℃. It can withstand vibrations below 1000Hz. The CPU is the core component of the ECU. When the engine is running, it collects signals from various sensors, performs calculations, and converts the results into control signals to control the operation of the controlled object. It also controls the memory, input / output interface, and other external circuits. The program stored in the ROM is written based on data obtained through precise calculations and extensive experiments. This inherent program continuously compares and calculates with the signals collected from various sensors when the engine is running.

3. The method of calculating the fuel sulfur content of a complete vehicle according to claim 1, wherein, In step S1, the engine speed is transmitted to the tachometer by a speed sensor. The engine speed is related to the number of work cycles per unit time or the effective power of the engine. That is, the effective power of the engine changes with the speed. When describing the effective power of the engine, the corresponding speed must be specified at the same time. The working condition of the engine at the rated power and rated speed is called the rated operating condition. The rated power is not the maximum power that the engine can produce. It is the maximum usable limit of effective power determined according to the purpose of the engine.

4. The method of claim 1, wherein, In step S1, the engine fuel injection quantity is controlled by the ECU. The ECU calculates various operating parameters such as engine intake air volume, speed, throttle opening, water temperature, and intake air temperature measured by various sensors according to a set program, and sends electrical pulses to the injectors according to the calculation results. By changing the width of each electrical pulse, the duration of each injection by each injector is controlled, thereby achieving the purpose of controlling the fuel injection quantity. The wider the electrical pulse, the longer the injection duration and the greater the fuel injection quantity. The engine operates under different conditions, and the requirements for the air-fuel mixture concentration are also different. Under some special conditions, including starting, rapid acceleration, and rapid deceleration, there are special requirements for the air-fuel mixture concentration. The ECU must control the fuel injection quantity in different ways according to the operating conditions measured by relevant sensors. The fuel injection quantity control methods are divided into several types: start control, operation control, fuel cut-off control, and feedback control.

5. The method of calculating the fuel sulfur content of a complete vehicle as claimed in claim 1, wherein, The catalyst in step S1 is an SCR catalyst, which stands for Selective Catalytic Reduction. The principle of SCR is to improve the selectivity of N2 and reduce the consumption of NH3 under the action of catalysis. The principle of SCR is that the reduction reaction can be carried out at a lower temperature under the action of the catalyst. The SCR system consists of an ammonia supply system, an ammonia / air injection system, a catalytic reaction system and a control system.

6. The method for calculating the sulfur content of fuel in a vehicle according to claim 1, characterized in that, The urea injection in step S1, which involves injecting an appropriate amount of urea at the right time to participate in the conversion reaction, is the key to achieving NOx emission standards. The working principle of urea injection is divided into three stages: pre-injection stage, injection metering stage, and venting stage.

7. The method for calculating the sulfur content of fuel in a vehicle according to claim 1, characterized in that, In step S1, the nitrogen and oxygen concentrations upstream and downstream of the catalyst are measured by the upstream nitrogen and oxygen concentration sensor and transmitted to the ECU. The exhaust temperature upstream of the catalyst is measured by the upstream temperature sensor and transmitted to the ECU. The engine exhaust flow rate is measured by the air flow sensor and transmitted to the ECU.

8. The method for calculating the sulfur content of fuel in a vehicle according to claim 1, characterized in that, The NARX neural network model in step S2 is called a nonlinear autoregressive neural network with external input. It is an effective time series prediction technique. First, a time series is divided into a training set, a validation set, and a test set. The division method is based on the chronological order of time, rather than random allocation. Then, the training set is used to train the open-loop NARX. After training, the open-loop NARX is transformed into a closed-loop NARX. Next, the prediction performance of the closed-loop NARX on the validation set is examined, and the decision on whether to adopt the NARX network is made based on the prediction results.

9. The method for calculating the sulfur content of fuel in a vehicle according to claim 1, characterized in that, The sulfur content in step S3 is the proportion of sulfur to exhaust gas. The total sulfur content in fuel oil includes elemental sulfur, active sulfides and inactive sulfides. When the sulfur content in liquid fuel is high, SO2 and SO3 are formed after combustion.

10. The method for calculating the sulfur content of fuel in a vehicle according to claim 1, characterized in that, In step S3, the alarm for sulfur poisoning of the SCR catalyst is generated by an alarm device controlled by the ECU.

Images