CN113326992B - Optimization method for smoke limit value of electric control diesel engine under working condition of rapidly increasing accelerator - Google Patents

Optimization method for smoke limit value of electric control diesel engine under working condition of rapidly increasing accelerator Download PDF

Info

Publication number
CN113326992B
CN113326992B CN202110731750.6A CN202110731750A CN113326992B CN 113326992 B CN113326992 B CN 113326992B CN 202110731750 A CN202110731750 A CN 202110731750A CN 113326992 B CN113326992 B CN 113326992B
Authority
CN
China
Prior art keywords
torque
working condition
rotating speed
distribution method
diesel engine
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202110731750.6A
Other languages
Chinese (zh)
Other versions
CN113326992A (en
Inventor
杨蓉
卓能聪
覃频频
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangxi University
Original Assignee
Guangxi University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangxi University filed Critical Guangxi University
Priority to CN202110731750.6A priority Critical patent/CN113326992B/en
Publication of CN113326992A publication Critical patent/CN113326992A/en
Application granted granted Critical
Publication of CN113326992B publication Critical patent/CN113326992B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks

Abstract

The invention discloses an optimization method of an electric control diesel engine smoke limit value under the working condition that an accelerator is rapidly increased, belongs to the technical field of diesel engines, and particularly relates to the optimization method of the electric control diesel engine smoke limit value under the working condition that the accelerator is rapidly increased. The method is characterized in that a prediction model of circulation air inflow, circulation combustion efficiency and emission in the transient process is established based on an LSTM neural network prediction algorithm in combination with steady-state test data and transient test data of the diesel engine, the concentration of circulation mixed gas is determined according to the power demand of each circulation, a transition path is designed in combination with a theory, the transition path is analyzed one by using the prediction model with the aim of stable circulation combustion efficiency and low emission to obtain an optimal transition path, a smoke limit value MAP graph in the transient process is formed in combination with state parameters of the diesel engine, and the optimization of smoke limit values is realized.

Description

Optimization method for smoke limit value of electric control diesel engine under working condition of rapidly increasing accelerator
Technical Field
The invention relates to the technical field of diesel engines, in particular to an optimization method of an electric control diesel engine smoke limit value under the working condition that an accelerator is rapidly increased.
Background
In actual life, a diesel engine is often in a transient working condition that an accelerator is greatly suddenly increased, such as the initial stage of accelerating the diesel vehicle to climb a steep slope, the time of suddenly increasing the load of a bulldozer, the initial stage of contacting a deep ploughing machine with soil and the like. Under such conditions, the cyclic fuel injection quantity is suddenly and greatly increased, and the intake response lags behind the fuel injection change, so that the combustion and emission performance of the diesel engine greatly deviates from the steady-state condition, and the deterioration tendency tends to occur. So many countries now propose transient condition test items in particular in the economy and emission performance test standards of diesel engines and diesel vehicles.
In order to solve the problems of high oil consumption and poor emission under the condition that an accelerator is rapidly increased, the current widely-used electronic control high-pressure common rail diesel engine adopts a Smoke limiting control strategy, namely, a Smoke limiting value (Smok _ limit) is additionally arranged in the control strategy, when the variation amplitude of the accelerator of the diesel engine is large, the maximum value of the circulating fuel injection quantity is limited, and the situation that the combustion is deteriorated and black Smoke is generated due to overhigh concentration of mixed gas in the circulating combustion process is prevented. And the smoke limit value of the electric control diesel engine is reasonably set and optimized, so that the combustion and emission performance of the diesel engine under the transient working condition can be optimized.
Therefore, the application provides a method for optimizing the smoke limit value of the electric control diesel engine under the condition that the throttle is rapidly increased.
Disclosure of Invention
The invention aims to provide an optimization method of the smoke limit of the electric control diesel engine under the working condition that the accelerator is rapidly increased, a transition path is designed in combination with theory, the transition path is analyzed one by utilizing a prediction model with the aim of stable cyclic combustion efficiency and low emission so as to obtain an optimal transition path, and the optimization of the smoke limit is realized by combining experiments and theory.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
the optimization method of the smoke limit value of the electric control diesel engine under the working condition that the accelerator is rapidly increased comprises the following steps,
s10, according to test parameters of the rotating speed, the torque, the accelerator opening and the accelerator opening change rate of each accelerator working condition, calculating to obtain the change parameters of the rotating speed and the torque of the accelerator working condition which are not tested by using transient models of various accelerator working conditions which are respectively established based on an LSTM neural network; the working conditions of the oil filling door comprise a constant-rotating-speed oil filling door working condition, a constant-torque oil filling door working condition and an acceleration oil filling door working condition;
step S20, calculating to obtain the number of cycles that each cylinder of the diesel engine passes during the transition process according to the accelerator change time obtained by the test and the change parameter calculation result of the rotating speed;
step S30, analyzing and calculating the inertia moment change of the moving mass caused by the rotation speed change of each cycle in the transition process according to the Newton law, integrating the energy required by the rotation speed and the torque change, and determining the power demand of each cycle;
step S40, obtaining a combustion efficiency predicted value and a cycle air inflow predicted value according to the power demand of each cycle by using a cycle air inflow, a cycle combustion efficiency and emission prediction model of the transient process, reversely deducing the total heat release and the cycle oil injection of each cycle, and combining the cycle air inflow predicted value to obtain the concentration of the cycle mixed gas; the method comprises the following steps that a prediction model of cycle air inflow, cycle combustion efficiency and emission in the transient process is established based on an LSTM neural network prediction algorithm in combination with steady-state test and transient test data of a diesel engine;
s50, analyzing each designed transition path of each oil filling door working condition one by adopting a prediction model of the circulation air inflow, the circulation combustion efficiency and the emission of the transient process and taking the stability of the circulation combustion efficiency and the low emission as targets to determine the optimal transition path of each oil filling door working condition;
and step S60, analyzing and obtaining the air-fuel ratio or the excess air coefficient of the optimal path of each fuel filler door working condition, and combining with the state parameters of the diesel engine to form a smoke limit value MAP graph in the transient process.
In step S10, the test parameters are obtained as follows: aiming at 3 working conditions of the oil filling door, an orthogonal test is designed, tests of different initial rotating speeds, different initial torques, different initial accelerator opening degrees and different change rates of the oil filling door opening degrees are carried out, and test parameters of the rotating speed, the torque, the accelerator opening degree and the change rate of the oil filling door opening degree of each working condition of the oil filling door are obtained through the tests.
As an option, in step S40, the cyclic intake air amount, the cyclic combustion efficiency, and the initial value of the neuron weight of the LSTM neural network of the prediction model of the transient are optimized through a genetic algorithm, so as to improve the defect of random selection of the initial value of the neuron weight of the neural network, and improve the prediction accuracy of the prediction model.
As an option, in step S50, the transition path of each filler door operating condition is designed by an average distribution method and an uneven distribution method, which are as follows:
aiming at the working condition of the constant-speed refueling door, a torque average distribution method and a torque non-uniform distribution method are applied to design a plurality of groups of transition paths; the torque average distribution method is to uniformly distribute the torque change of the whole process according to the initial and final states of the torque and the predicted number of cycles; the torque uneven distribution method means that the torque change rate changes in a segmented way in the whole process;
aiming at the working condition of a constant-torque refueling door, a plurality of groups of transition paths are designed by using a rotating speed average distribution method and a rotating speed uneven distribution method; the rotating speed average distribution method is to uniformly distribute the rotating speed change of the whole process according to the initial and final states of the rotating speed and the predicted number of cycles; the uneven distribution method of the rotating speed means that the changing rate of the rotating speed changes in sections in the whole process;
aiming at the working condition of an accelerating refueling door, a plurality of groups of transition paths are designed by using a rotating speed and torque average distribution method and a rotating speed and torque non-uniform distribution method; the rotating speed and torque average distribution method is to uniformly distribute the rotating speed and torque change in the whole process according to the initial and final states of the rotating speed and torque and the predicted number of cycles; the rotation speed and torque non-uniform distribution method means that the change rate of the rotation speed and the torque changes in stages in the whole process.
Due to the adoption of the technical scheme, the invention has the following beneficial effects:
1. the method is characterized in that a prediction model of circulation air inflow, circulation combustion efficiency and emission in the transient process is established based on an LSTM neural network prediction algorithm in combination with steady-state test data and transient test data of the diesel engine, the concentration of circulation mixed gas is determined according to the power demand of each circulation, a transition path is designed in combination with a theory, the transition path is analyzed one by using the prediction model with the aim of stable circulation combustion efficiency and low emission to obtain an optimal transition path, a smoke limit value MAP graph in the transient process is formed in combination with state parameters of the diesel engine, and the optimization of smoke limit values is realized.
2. The optimized smoke limit value is used for the electric control diesel engine, and is beneficial to reducing the smoke intensity and PM emission of the diesel engine under the working condition that the accelerator is rapidly increased.
3. The invention can provide a new method for determining the smoke limit of the electric control diesel engine, reduce the test workload and reduce the test time and cost of the engine pedestal.
Drawings
FIG. 1 is a flowchart of example 1 of the present invention.
Fig. 2 is a flowchart of embodiment 2 of the present invention.
Fig. 3 is a schematic view of the torque-average distribution method of embodiment 2 of the present invention.
FIG. 4 is a schematic view of the unidirectional torque maldistribution method of embodiment 2 of the present invention.
Fig. 5 is a schematic view of the rotational speed and torque average split method of embodiment 2 of the invention.
Detailed Description
The following further describes the embodiments of the present invention with reference to the drawings.
Example 1
As shown in fig. 1, the method for optimizing the smoke limit of the electronically controlled diesel engine under the rapid acceleration condition of the present embodiment 1 includes the following steps,
s01, respectively establishing transient models of various fuel filler door working conditions based on an LSTM neural network, wherein the fuel filler door working conditions comprise a constant-rotation-speed fuel filler door working condition, a constant-torque fuel filler door working condition and an acceleration fuel filler door working condition; wherein, this step can also adjust the position as required, such as between step S05 and step S10;
step S03, establishing a prediction model of the circulation air inflow, the circulation combustion efficiency and the emission in the transient process based on the LSTM neural network prediction algorithm and combined with the steady-state test and the transient test data of the diesel engine; this step may also adjust the position as desired, such as between steps S05 and S10, or between steps S30 and S40, etc.;
s05, designing an orthogonal test aiming at 3 working conditions of the refueling door, developing tests of different initial rotating speeds, different initial torques, different initial accelerator opening degrees and different accelerator opening degree change rates, and testing to obtain test parameters of the rotating speed, the torque, the accelerator opening degree and the accelerator opening degree change rate of each working condition of the refueling door;
step S10, according to the test parameters of the rotating speed, the torque, the accelerator opening and the accelerator opening change rate of each accelerator working condition, calculating the change parameters of the rotating speed and the torque of the accelerator working condition which are not tested by using the transient models of various accelerator working conditions which are respectively established based on the LSTM neural network, and calculating the change parameters of the rotating speed and the torque;
step S20, calculating to obtain the number of cycles that each cylinder of the diesel engine passes during the transition process according to the accelerator change time obtained by the test and the change parameter calculation result of the rotating speed;
step S30, analyzing and calculating the inertia moment change of the moving mass caused by the rotation speed change of each cycle in the transition process according to the Newton law, integrating the energy required by the rotation speed and the torque change, and determining the power demand of each cycle;
step S40, obtaining a combustion efficiency predicted value and a cycle air inflow predicted value according to the power demand of each cycle by using a cycle air inflow, a cycle combustion efficiency and emission prediction model of the transient process, reversely deducing the total heat release and the cycle oil injection of each cycle, and combining the cycle air inflow predicted value to obtain the concentration of the cycle mixed gas;
s45, designing transition paths of the working conditions of the refueling doors by adopting an average distribution method and an uneven distribution method; this step may also adjust the position as needed, such as between step S20 and step S30, or between step S30 and step S40; the specific content of the step is as follows:
aiming at the working condition of the constant-speed refueling door, a torque average distribution method and a torque non-uniform distribution method are applied to design a plurality of groups of transition paths; the torque average distribution method is to uniformly distribute the torque change of the whole process according to the initial and final states of the torque and the predicted number of cycles; the torque uneven distribution method means that the torque change rate changes in a segmented way in the whole process;
aiming at the working condition of a constant-torque refueling door, a plurality of groups of transition paths are designed by using a rotating speed average distribution method and a rotating speed uneven distribution method; the rotating speed average distribution method is to uniformly distribute the rotating speed change of the whole process according to the initial and final states of the rotating speed and the predicted number of cycles; the uneven distribution method of the rotating speed means that the changing rate of the rotating speed changes in sections in the whole process;
aiming at the working condition of an accelerating refueling door, a plurality of groups of transition paths are designed by using a rotating speed and torque average distribution method and a rotating speed and torque non-uniform distribution method; the rotating speed and torque average distribution method is to uniformly distribute the rotating speed and torque change in the whole process according to the initial and final states of the rotating speed and torque and the predicted number of cycles; the uneven distribution method of the rotating speed and the torque refers to that the changing rate of the rotating speed and the torque changes in a segmented way in the whole process;
s50, analyzing each designed transition path of each oil filling door working condition one by adopting a prediction model of the circulation air inflow, the circulation combustion efficiency and the emission of the transient process and taking the stability of the circulation combustion efficiency and the low emission as targets to determine the optimal transition path of each oil filling door working condition;
and step S60, analyzing and obtaining the air-fuel ratio or the excess air coefficient of the optimal path of each fuel filler door working condition, and combining with the state parameters of the diesel engine to form a smoke limit value MAP graph in the transient process.
The MAP of the smoke limit value is a three-dimensional or multi-dimensional data table commonly existing in the electronic control unit of the current electronic control diesel engine. In the existing electric control diesel engine products, some electric control diesel engine products take air inlet flow and engine rotating speed as main input parameters and take a smoke limiting lambda value as an output result; the air intake flow and the engine rotating speed of some products are used as main input parameters, the cooling water temperature and the atmospheric pressure of the engine are used as correction parameters, and the smoke limit lambda value is finally used as an output result. This is also the final optimization goal of the present invention.
As described above, a prediction model of the circulation air inflow, the circulation combustion efficiency and the emission of the transient process is established based on an LSTM neural network prediction algorithm in combination with the steady-state test data and the transient test data of the diesel engine, the concentration of circulation mixed gas is determined according to the power demand of each circulation, a transition path is designed in combination with a theory, the transition path is analyzed one by using the prediction model with the aim of the stability and the low discharge of the circulation combustion efficiency, so as to obtain an optimal transition path, a smoke limit value MAP graph of the transient process is formed in combination with the state parameters of the diesel engine, and the optimization of the smoke limit value is realized.
Example 2
This example 2 will be further illustrated.
As shown in fig. 2, the method for optimizing the smoke limit of the electronically controlled diesel engine under the rapid acceleration condition of embodiment 2 includes the following steps,
(1) and establishing a prediction model of the circulation air inflow, the circulation combustion efficiency and the emission in the transient process based on an LSTM neural network prediction algorithm by combining the test data of a steady-state test (universal characteristic) and a transient test (only torque change, only rotating speed change and simultaneous torque and rotating speed change). This step is an existing technique and can be used to predict NOx emissions using the LSTM neural network according to the existing technique and will not be described herein.
(2) And optimizing the initial value of the weight of the neuron of the LSTM neural network by combining a genetic algorithm, and overcoming the defect of random selection of the initial value of the weight of the neuron of the neural network so as to improve the circulating air inflow, the circulating combustion efficiency and the prediction accuracy of the prediction model of the emission in the transient process. This step is an existing technique, and can be implemented according to the existing technique, and is not described herein again.
(3) Aiming at the typical sudden acceleration working condition of the whole vehicle accelerator, namely the working condition of a constant-rotating-speed accelerator (vehicle climbing), the working condition of a constant-torque accelerator (vehicle straight-way acceleration) and the working condition of an acceleration accelerator (vehicle acceleration climbing), an orthogonal test is designed, and test researches of different initial rotating speeds, different initial torques, different initial accelerator opening degrees and different accelerator opening degree change rates are carried out.
(4) According to the test result, a transient model of 3 working conditions of the fuel filler door is respectively established by combining an LSTM neural network, the input parameters of the transient model are rotating speed, torque, accelerator and accelerator change rate, the output parameters are rotating speed and torque, and the change characteristics of the rotating speed and the torque (the change parameters of the rotating speed and the torque) are obtained.
(5) And (4) according to the accelerator change time obtained by the test, combining the rotating speed change parameter calculation result (the rotating speed change prediction result obtained by calculation prediction) to obtain the cycle number possibly experienced by each cylinder of the diesel engine in the transition process.
(6) And (3) designing a transition path for the working condition of the constant-speed refueling door by using a torque average distribution method and a torque non-uniform distribution method.
The average distribution method is to distribute the torque variation of the whole process evenly according to the initial and final states of the torque and the predicted number of cycles. As shown in FIG. 3, which is a schematic diagram of the torque averaging distribution method, the large dots are the initial and final states, and the uniform grid indicates that the torque is uniformly changed.
The non-uniform distribution method includes 2 types: unidirectional and multidirectional. The unidirectional means that the torque change rate in the whole process is monotonically increased or monotonically decreased from small to large or from large to small. As shown in fig. 4 below, the diagram is a schematic diagram of the one-way non-uniform distribution method, where the diagram (4a) is monotonically increasing, the diagram (4b) is monotonically decreasing, and the width of the grid in the diagram represents the change rate.
The multidirectional torque change rate is changed in various conditions in the whole process, the initial exploration is carried out, in order to simplify the research process, the change rate can be set to be 2-3 types, and the change rate can be divided into the following 5 types: large-small, large-small-large, small-large-small, and small-large; the whole process is divided into 3 segments according to the number of cycles calculated, and each segment corresponds to 1 change rate. Then, if the result of the later test is not ideal, the number of segments and the number of change rates can be increased. Finally, with the accumulation of research data and the summarization of rules, the number of change rates and the change rules can be fixed at the later stage. The later stage test result refers to that after the relevant research is finished, the engine bench test is carried out according to the research result to obtain a bench test result; the optimization result of the smoking limit value obtained by earlier research is obtained based on theoretical analysis; after theoretical analysis is finished, the engine bench test is required to be verified.
(7) And designing a transition path for the working condition of the constant-torque refueling door by using a rotating speed average distribution method and a rotating speed uneven distribution method. The specific design method for the rotation speed is similar to the step (6), and is not described herein again.
(8) And (3) designing a plurality of groups of transition paths by applying a rotating speed and torque average distribution method and a rotating speed and torque non-uniform distribution method to the working condition of the accelerator.
The rotation speed and torque average distribution method is to uniformly distribute the torque variation in the whole process according to the initial and final states of the rotation speed and the torque and the predicted number of cycles, and as shown in fig. 5, the rotation speed and torque average distribution method is a schematic diagram.
The non-uniform distribution method includes 2 types: unidirectional and multidirectional. The one-way mode comprises 4 conditions, namely, the torque change rate and the torque change rate in the whole process are simultaneously and monotonically increased; the torque change rate and the torque change rate in the whole process are monotonically decreased; the torque change rate is monotonically increased, and the rotating speed change rate is monotonically decreased; the torque change rate is monotonously decreased, and the rotating speed change rate is monotonously increased.
The multidirectional non-uniform distribution method is designed in the same principle as the step (6), but is more complex at the moment, and the change rules of the rotating speed and the torque are considered at the moment because the rotating speed and the torque change simultaneously, and then the combination analysis is carried out.
(9) And analyzing and calculating the change of the inertia moment of the moving mass caused by the change of the rotating speed of each cycle in the transition process according to the Newton law, and determining the power demand of each cycle by integrating the energy required by the change of the rotating speed and the torque.
(10) Reversely deducing the total heat release quantity and the circular fuel injection quantity of each circle according to the circular power demand and a combustion efficiency predicted value obtained by a prediction model of the circular air inflow, the circular combustion efficiency and the discharge in the transient process; and determining the concentration of the circulating mixed gas by combining the predicted value of the circulating air inflow.
(11) And analyzing each transition path one by utilizing a prediction model of the circulation air inflow, the circulation combustion efficiency and the emission in the transient process and combining main combustion and emission characteristic parameters obtained by research and representing the transient process of the diesel engine, and determining the optimal transition path by taking low emission and stable combustion efficiency as targets. Wherein the combustion characteristic parameters include: combustion efficiency and total heat release from combustion; the emission characteristic parameters mainly include: NOXAnd PM.
(12) The air-fuel ratio or excess air ratio of the optimal path is summarized and combined with engine state parameters to form a transient smoke limit MAP.
As described above, the transient process in the present invention mainly refers to a process in which each operating parameter and performance parameter of the engine change due to a change in the accelerator, and refers to a process in which the operating state of the engine changes from one stable state to another stable state.
The transition path mentioned in the invention mainly refers to the change rule of the rotating speed and the torque of the engine along with time in the accelerator change process, and can embody a relevant path curve by drawing through a rectangular coordinate system taking the rotating speed as an x axis and the torque as a y axis.
The optimized smoke limit value is used for the electric control diesel engine, and is beneficial to reducing the smoke intensity and PM emission of the diesel engine under the working condition that the accelerator is rapidly increased. The invention can provide a new method for determining the smoke limit of the electric control diesel engine, reduce the test workload and reduce the test time and cost of the engine pedestal.
The foregoing description is directed to the details of preferred and exemplary embodiments of the invention, and not to the limitations defined thereby, which are intended to cover all modifications and equivalents of the invention as may come within the spirit and scope of the invention.

Claims (4)

1. An optimization method for the smoke limit value of an electric control diesel engine under the working condition that an accelerator is rapidly increased is characterized in that: including the following in-eluding matters,
s10, according to test parameters of the rotating speed, the torque, the accelerator opening and the accelerator opening change rate of each accelerator working condition, calculating to obtain the change parameters of the rotating speed and the torque of the accelerator working condition which are not tested by using transient models of various accelerator working conditions which are respectively established based on an LSTM neural network; the working conditions of the oil filling door comprise a constant-rotating-speed oil filling door working condition, a constant-torque oil filling door working condition and an acceleration oil filling door working condition;
step S20, calculating to obtain the number of cycles that each cylinder of the diesel engine passes during the transition process according to the accelerator change time obtained by the test and the change parameter calculation result of the rotating speed;
step S30, analyzing and calculating the inertia moment change of the moving mass caused by the rotation speed change of each cycle in the transition process according to the Newton law, integrating the energy required by the rotation speed and the torque change, and determining the power demand of each cycle;
step S40, obtaining a combustion efficiency predicted value and a cycle air inflow predicted value according to the power demand of each cycle by using a cycle air inflow, a cycle combustion efficiency and emission prediction model of the transient process, reversely deducing the total heat release and the cycle oil injection of each cycle, and combining the cycle air inflow predicted value to obtain the concentration of the cycle mixed gas; the method comprises the following steps that a prediction model of cycle air inflow, cycle combustion efficiency and emission in the transient process is established based on an LSTM neural network prediction algorithm in combination with steady-state test and transient test data of a diesel engine;
s50, analyzing each designed transition path of each oil filling door working condition one by adopting a prediction model of the circulation air inflow, the circulation combustion efficiency and the emission of the transient process and taking the stability of the circulation combustion efficiency and the low emission as targets to determine the optimal transition path of each oil filling door working condition;
and step S60, analyzing and obtaining the air-fuel ratio or the excess air coefficient of the optimal path of each fuel filler door working condition, and combining with the state parameters of the diesel engine to form a smoke limit value MAP graph in the transient process.
2. The optimization method of the smoke limit value of the electric control diesel engine under the working condition of rapidly increasing the accelerator according to claim 1, which is characterized in that: in step S10, the test parameters are obtained as follows: aiming at 3 working conditions of the oil filling door, an orthogonal test is designed, tests of different initial rotating speeds, different initial torques, different initial accelerator opening degrees and different change rates of the oil filling door opening degrees are carried out, and test parameters of the rotating speed, the torque, the accelerator opening degree and the change rate of the oil filling door opening degree of each working condition of the oil filling door are obtained through the tests.
3. The optimization method of the smoke limit value of the electric control diesel engine under the working condition of rapidly increasing the accelerator according to claim 1, which is characterized in that: in step S40, the initial values of the neuron weights of the LSTM neural network of the prediction model of the cycle intake air amount, the cycle combustion efficiency, and the emissions of the transient are optimized by the genetic algorithm.
4. The optimization method of the smoke limit value of the electric control diesel engine under the working condition of rapidly increasing the accelerator according to claim 1, which is characterized in that: in step S50, the transition path of each filler door operating condition is designed by an average distribution method and an uneven distribution method, which are specifically as follows:
aiming at the working condition of the constant-speed refueling door, a torque average distribution method and a torque non-uniform distribution method are applied to design a plurality of groups of transition paths; the torque average distribution method is to uniformly distribute the torque change of the whole process according to the initial and final states of the torque and the predicted number of cycles; the torque uneven distribution method means that the torque change rate changes in a segmented way in the whole process;
aiming at the working condition of a constant-torque refueling door, a plurality of groups of transition paths are designed by using a rotating speed average distribution method and a rotating speed uneven distribution method; the rotating speed average distribution method is to uniformly distribute the rotating speed change of the whole process according to the initial and final states of the rotating speed and the predicted number of cycles; the uneven distribution method of the rotating speed means that the changing rate of the rotating speed changes in sections in the whole process;
aiming at the working condition of an accelerating refueling door, a plurality of groups of transition paths are designed by using a rotating speed and torque average distribution method and a rotating speed and torque non-uniform distribution method; the rotating speed and torque average distribution method is to uniformly distribute the rotating speed and torque change in the whole process according to the initial and final states of the rotating speed and torque and the predicted number of cycles; the rotation speed and torque non-uniform distribution method means that the change rate of the rotation speed and the torque changes in stages in the whole process.
CN202110731750.6A 2021-06-30 2021-06-30 Optimization method for smoke limit value of electric control diesel engine under working condition of rapidly increasing accelerator Active CN113326992B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110731750.6A CN113326992B (en) 2021-06-30 2021-06-30 Optimization method for smoke limit value of electric control diesel engine under working condition of rapidly increasing accelerator

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110731750.6A CN113326992B (en) 2021-06-30 2021-06-30 Optimization method for smoke limit value of electric control diesel engine under working condition of rapidly increasing accelerator

Publications (2)

Publication Number Publication Date
CN113326992A CN113326992A (en) 2021-08-31
CN113326992B true CN113326992B (en) 2022-03-29

Family

ID=77425206

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110731750.6A Active CN113326992B (en) 2021-06-30 2021-06-30 Optimization method for smoke limit value of electric control diesel engine under working condition of rapidly increasing accelerator

Country Status (1)

Country Link
CN (1) CN113326992B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114635806B (en) * 2022-03-21 2023-08-18 重庆隆鑫发动机有限公司 Motorcycle and engine fueling transient fuel injection control method, system and device thereof

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
AT508817A2 (en) * 2010-12-06 2011-04-15 Avl List Gmbh METHOD AND ARRANGEMENT FOR DETERMINING A SMOKE-LIMITING IDENTIFICATION OF A COMBUSTION ENGINE
CN106762181A (en) * 2016-11-28 2017-05-31 哈尔滨工程大学 A kind of transient state EGR control methods based on neutral net
WO2017222454A1 (en) * 2016-06-21 2017-12-28 Scania Cv Ab Method and system for controlling the amount of fuel in connection to operating an internal combustion engine
CN111894748A (en) * 2020-07-24 2020-11-06 东风商用车有限公司 Control method and control system for smoke limit under idle working condition of engine
CN112096532A (en) * 2019-06-18 2020-12-18 北京福田康明斯发动机有限公司 Method and system for limiting fuel injection quantity
CN112555039A (en) * 2020-11-13 2021-03-26 潍柴动力股份有限公司 Engine control method and system based on smoke intensity working condition recognition

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012047191A1 (en) * 2010-10-04 2012-04-12 International Engine Intellectual Property Company, Llc Engine controlling emissions during transient operations

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
AT508817A2 (en) * 2010-12-06 2011-04-15 Avl List Gmbh METHOD AND ARRANGEMENT FOR DETERMINING A SMOKE-LIMITING IDENTIFICATION OF A COMBUSTION ENGINE
WO2017222454A1 (en) * 2016-06-21 2017-12-28 Scania Cv Ab Method and system for controlling the amount of fuel in connection to operating an internal combustion engine
CN106762181A (en) * 2016-11-28 2017-05-31 哈尔滨工程大学 A kind of transient state EGR control methods based on neutral net
CN112096532A (en) * 2019-06-18 2020-12-18 北京福田康明斯发动机有限公司 Method and system for limiting fuel injection quantity
CN111894748A (en) * 2020-07-24 2020-11-06 东风商用车有限公司 Control method and control system for smoke limit under idle working condition of engine
CN112555039A (en) * 2020-11-13 2021-03-26 潍柴动力股份有限公司 Engine control method and system based on smoke intensity working condition recognition

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Experimental Assessment of Turbocharged Diesel Engine Transient Emissions during acceleration, Load Change and Starting;C.D.Rakopoulos等;《SEA》;20100430;第18节 *
冒烟限制对重型电控柴油机加速瞬态性能影响试验研究;曹卢等;《现代制造技术与装备》;20170131(第1期);第69-72+74页 *
基于烟度限值的柴油机怠速瞬态过程性能评价方法与试验;楼狄明等;《农业工程学报》;20170228;第33卷(第4期);第111-116页 *
重型增压柴油机瞬变工况的燃烧阶段特性研究;杨蓉等;《内燃机工程》;20150430;第36卷(第4期);第41-45+52页 *

Also Published As

Publication number Publication date
CN113326992A (en) 2021-08-31

Similar Documents

Publication Publication Date Title
KR102644366B1 (en) Engine Virtual Test Environment System and Engine Management System Mapping Method thereof
WO2000009876A1 (en) Engine combustion control device
US9322357B2 (en) Soot accumulation model for setpoint modification
US20210189990A1 (en) Vehicle controller, vehicle control system, and learning device for vehicle
CN110966063B (en) Method and device for controlling variable valve timing system of Miller cycle engine
CN113326992B (en) Optimization method for smoke limit value of electric control diesel engine under working condition of rapidly increasing accelerator
CN114970404B (en) Engine oil consumption calculation and optimization method based on in-cylinder combustion CFD analysis
CN102213150B (en) Method and system for enabling cylinder balancing at low idle speed using crankshaft speed sensor
US20130197785A1 (en) Nox control
US8463530B2 (en) Method for operating auto ignition combustion engine
CN103225551A (en) Oxygen concentration setpoint modification
CN112393908A (en) Bench test method for pollutant emission of non-road diesel engine
CN102678361A (en) Engine control system with algorithm for actuator control
JP2012013637A (en) Adaptation method and adaptation device for engine control parameter
US10132318B2 (en) Turbocharger control
US9328691B2 (en) Setpoint bank control architecture
Kamaruddin et al. PID controller for idle speed control
Mu et al. Study with fuzzy self-turning PID controller on idle speed controlling of automobile
CN117418944B (en) Control method and device of EGR (exhaust gas Recirculation) system, electronic equipment and storage medium
Kang et al. Modeling and optimal control for torque tracking of spark-ignition engines with low pumping loss
Xu et al. Modeling Analysis on Improving Cylinder Balance in A Gasoline Engine Using Electromagnetic Valve Train
Wagner et al. Characterizing lean spark ignition combustion instability in terms of a low-order map
Kako SICE benchmark problem: starting speed control of SI engines
JP2009209779A (en) Adaptation method and adaptation device of engine control parameter
Beham et al. Development of a new automatic calibration method for control of variable valve timing

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant