CN113239732A - Engine knock intensity calculation method, system and readable storage medium - Google Patents
Engine knock intensity calculation method, system and readable storage medium Download PDFInfo
- Publication number
- CN113239732A CN113239732A CN202110392081.4A CN202110392081A CN113239732A CN 113239732 A CN113239732 A CN 113239732A CN 202110392081 A CN202110392081 A CN 202110392081A CN 113239732 A CN113239732 A CN 113239732A
- Authority
- CN
- China
- Prior art keywords
- knock intensity
- data set
- subset
- knock
- training
- 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.)
- Granted
Links
- 238000004364 calculation method Methods 0.000 title claims abstract description 53
- 238000012360 testing method Methods 0.000 claims abstract description 36
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000004590 computer program Methods 0.000 claims description 4
- 230000010076 replication Effects 0.000 claims description 2
- 238000005070 sampling Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 16
- 230000005484 gravity Effects 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 238000002485 combustion reaction Methods 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
Abstract
The application relates to the technical field of engine detection, in particular to a method and a system for calculating engine knock intensity and a readable storage medium. The method comprises the following steps: providing a training data set and a testing data set; training a knock intensity calculation primary model based on a training data set; calculating and determining a large error subset in the test data set; the large error subset is a set of knock intensity signals, wherein in the test data set, the intensity error between the calculated value of the knock intensity calculation model and the actual value of the knock intensity exceeds a specified threshold value; determining the distribution rule of the large error subset; determining a data subset in a training data set, wherein the data subset conforms to a distribution rule; expanding the data of the data subset based on an oversampling method so that the newly expanded data form an expanded data set; adding the extended data set into a training data set, and completing the training data to form a new training data set; and re-training an engine knock intensity calculation model with smaller error based on the new training data set.
Description
Technical Field
The application relates to the technical field of engine knock intensity detection, in particular to an engine knock intensity calculation method and system and a readable storage medium.
Background
Generally, an automobile engine realizes the repeated operation of the engine through the actions of four strokes of mixed gas suction, compression, combustion work and exhaust. When the engine sucks in the mixture of fuel vapor and air and the mixture does not reach the designed ignition position in the compression stroke, factors beyond control cause the fuel-air mixture to be ignited and combusted automatically. The large impact force generated by the combustion at this time is opposite to the movement direction of the piston, so that the engine is vibrated, which becomes knocking.
Engine knock can adversely affect both engine power and fuel consumption. If the knock intensity is too high, the phenomena of engine shaking and knocking can also occur, so that a spark plug, a valve and a piston are broken or ablated, the knock intensity reaches a certain degree, super knock is achieved, and the engine can be damaged instantly.
In the related art, a knock sensor is generally used to identify the engine knock phenomenon, but the knock sensor is difficult to identify the knock with small intensity, and the engine is damaged to different degrees when the knock problem is identified.
Disclosure of Invention
The application provides an engine knock intensity calculation method, an engine knock intensity calculation system and a readable storage medium, which can solve the problem of low-intensity knock which is difficult to identify in the related art.
In order to identify an engine knock intensity in real time so as to take necessary control measures in advance to reduce the knock intensity and prevent occurrence of super knock, a first aspect of the present application provides an engine knock intensity calculation method including:
providing a training data set and a testing data set;
training a knock intensity calculation primary model based on the training data set;
calculating a primary model based on the knock intensity, computationally determining a large subset of errors in the test data set; the large error subset is a set of knock intensity signals, wherein in the test data set, the intensity error between the calculated value of the knock intensity calculation model and the actual value of the knock intensity exceeds a specified threshold value;
determining the distribution rule of the large error subsets;
determining a data subset in the training data set, wherein the data subset conforms to the distribution rule;
expanding the data of the subset of data by at least 2 times based on an oversampling method such that the newly expanded data forms an expanded data set;
adding the extended data set into the training data set, and performing data completion on the training data to form a new training data set;
and re-training a knock intensity calculation model based on the new training data set.
Optionally, the test data set includes a plurality of knock intensity signals for testing, and knock intensity actual values corresponding to the knock intensity signals.
Optionally, the step of calculating a primary model based on the knock intensity, and calculating and determining a large error subset in the test data set, where the large error subset is a knock intensity signal set in which an error between a calculated value of the knock intensity calculation model and an actual value of the knock intensity exceeds a specified threshold value, includes:
calculating a calculation value of a knock intensity calculation model for each of the knock intensity signals in the test data set based on the knock intensity calculation primary model;
calculating an intensity error between a calculated value of the knock intensity calculation model corresponding to each knock intensity signal and the actual knock intensity;
determining the set of knock intensity signals for which the intensity error exceeds a specified threshold as a large error subset.
Optionally, the step of counting a distribution rule of the large error subset includes:
and determining a mode value interval of the knock intensity actual values based on the knock intensity actual values of the knock intensity signals in the large error subset.
Optionally, the number of knock intensity signals in the mode value range accounts for 80% to 90% of the total number of knock intensity signals in the large error subset.
Optionally, the training data set includes a plurality of knock intensity signals for training, and knock intensity actual values corresponding to the knock intensity signals.
Optionally, the step of determining the data subset in the training data set that conforms to the distribution rule includes:
based on the training data set, determining the set of knock intensity signals for which the corresponding knock intensity actual values lie in the mode value interval as a data subset of the training data set.
Optionally, the step of expanding the data of the data subset by at least 2 times based on the oversampling method, so that the newly expanded data forms an expanded data set, includes:
and performing sampling replication on the data in the data subsets based on an oversampling method, so that the data in the data subsets are repeated by at least 2 times to form an expanded data set.
In order to identify the engine knock intensity in real time so as to take necessary control measures in advance to reduce the knock intensity and prevent the onset of super knock, a second aspect of the present application provides an engine knock intensity calculation system for executing the engine knock intensity calculation method according to the first aspect of the present application.
In order to identify the engine knock intensity in real time so as to take necessary control measures in advance to reduce the knock intensity and prevent the onset of super knock, a third aspect of the present application provides a readable storage medium having stored therein a computer program, characterized in that the computer program, when executed by a processor, is capable of implementing the engine knock intensity calculation method as described in the first aspect of the present application.
The technical scheme at least comprises the following advantages: by adding the proportion of the subset with large error characteristics in the training data set, namely the ratio of the data subset in the new training data set, the knock intensity calculation model is trained again according to the new training data set E, and the knock intensity calculation model has smaller error and higher accuracy when calculating the knock intensity of the knock intensity signal.
Drawings
In order to more clearly illustrate the detailed description of the present application or the technical solutions in the prior art, the drawings needed to be used in the detailed description of the present application or the prior art description will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart illustrating a method for calculating engine knock intensity according to an embodiment of the present disclosure;
fig. 2 shows a schematic diagram of a data structure of a training data set a ═ { ax [1: n ], ay [1: n ] };
fig. 3 shows a data structure diagram of a test data set B ═ { bx [1: m ], by [1: m ] };
fig. 4 shows a schematic diagram of a process of training a knock intensity calculation primary model based on a training data set a ═ { ax [1: n ], ay [1: n ] };
fig. 5 is a schematic diagram illustrating a process of determining a large error subset in a test data set, for example, the test data set B ═ { bx [1: m ], by [1: m ] };
fig. 6 shows a schematic process diagram for determining the distribution rule of the large error subset B' ═ { bx [ p: q ], by [ p: q ] };
fig. 7 shows a schematic diagram of a process for determining a data subset in a training data set a ═ ax [1: n ], ay [1: n ] } that conforms to a distribution rule;
fig. 8 shows a schematic process diagram of a new extended data set D formed after three times the data subset a' ═ { ax [ v: w ], ay [ v: w ] };
fig. 9 shows a schematic process diagram of a new training data set E formed after appending the new extended data set D to the training data set a ═ ax [1: n ], ay [1: n ].
Detailed Description
The technical solutions in the present application will be described clearly and completely with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description and simplicity of description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present application. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present application, it is to be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; the connection can be mechanical connection or electrical connection; the two elements may be directly connected or indirectly connected through an intermediate medium, or may be communicated with each other inside the two elements, or may be wirelessly connected or wired connected. The specific meaning of the above terms in the present application can be understood in a specific case by those of ordinary skill in the art.
In addition, the technical features mentioned in the different embodiments of the present application described below may be combined with each other as long as they do not conflict with each other.
Fig. 1 shows a flowchart of a method for calculating engine knock according to an embodiment of the present application, and referring to fig. 1, the method for calculating engine knock includes the following steps:
step S1: a training data set and a test data set are provided.
The training data set comprises a large number of knock intensity signals used for model training and knock intensity actual values corresponding to the knock intensity signals; the test data set includes a plurality of knock intensity signals for the model test, and knock intensity actual values corresponding to the knock intensity signals. The training data set and the testing data set are acquired by carrying out experimental tests on 4 cylinders of the same rack engine in advance.
For clarity, let training dataset a be { ax [1: n ], ay [1: n ] }, test dataset B be { bx [1: m ], by [1: m ] }, n and m all belonging to positive integers. Please refer to fig. 2 and fig. 3, wherein fig. 2 shows a data structure diagram of a training data set a ═ { ax [1: n ], ay [1: n ] }, and fig. 3 shows a data structure diagram of a testing data set B ═ { bx [1: m ], by [1: m ] }.
Wherein ax represents a knock intensity signal array in the training data set A, and ax [1: n ] represents n knock intensity signals including ax [1] to ax [ n ] in the knock intensity signal array; ay represents the knock intensity actual value array in the training data set A, ay [1: n ] represents that the knock intensity actual value array comprises ay [1] to ay [ n ] and n knock intensity actual values, and the knock intensity signal array is in one-to-one correspondence with the knock intensity actual value array, namely, a knock intensity signal ax [1] corresponds to the knock intensity actual value ay [1], and a knock intensity signal ax [2] corresponds to the knock intensity actual value ay [2] … ….
Wherein, measuring bx represents a knock intensity signal array in the test data set B, bx [1: m ] represents that m knock intensity signals are included in the knock intensity signal array from bx [1] to bx [ m ]; by represents the knock intensity actual value array in the test data set B, by [1: m ] represents that the knock intensity actual value array comprises by [1] to by [ m ] and m knock intensity actual values, and the knock intensity signal array is in one-to-one correspondence with the knock intensity actual value array, namely, the knock intensity signal bx [1] corresponds to the knock intensity actual value by [1], the knock intensity signal bx [2] corresponds to the knock intensity actual value by [2] … …, and the knock intensity signal bx [ n ] corresponds to the knock intensity actual value by [ m ].
Step S2: based on the training data set, a knock intensity calculation primary model is trained.
Referring to fig. 4, a schematic diagram of a primary model for calculating knock intensity based on training data set a ═ { ax [1: n ], ay [1: n ] }isshown. In this embodiment, a knock intensity calculation primary model may be preliminarily trained through a machine learning algorithm based on a training data set.
Step S3: a primary model is calculated based on the knock intensity, and a large subset of errors in the test data set is computationally determined.
The large error subset is a subset of the test data set, and the large error subset is a set of knock intensity signals in the test data set for which an intensity error between a calculated value of the knock intensity calculation model and an actual value of the knock intensity exceeds a specified threshold.
In this embodiment, the step S3 of determining the large error subset in the test data set includes:
step S31: and calculating a calculation value of a knock intensity calculation model of each knock intensity signal in the test data set based on the knock intensity calculation primary model.
Step S32: and calculating an intensity error between the calculated value of the knock intensity calculation model corresponding to each knock intensity signal and the actual intensity of the knock.
Step S33: determining the set of knock intensity signals for which the intensity error exceeds a specified threshold as a large error subset.
Referring to fig. 5, a data flow diagram for determining a large error subset in a test data set is illustrated, taking test data set B ═ { bx [1: m ], by [1: m ] }.
FIG. 5 shows that the knock intensity signal array bx [1: m ] is calculated to obtain a calculated value array bz [1: m ] of a knock intensity calculation model through the knock intensity calculation primary model, the calculated value array bz [1: m ] of the knock intensity calculation model comprises bz [1] to bz [ m ], and m knock intensity calculated values are obtained. Then, the calculated knock intensity value of the calculated knock intensity array bz [1: m ] and the intensity error b | z-y | [1: m ] between the corresponding actual knock intensity values in the actual knock intensity array by [1: m ] are calculated. In this embodiment, the calculation formula of the intensity error may be:
b|z-y|[1:m]=|(bz[1:m]-by[1:m])/by[1:m]| (1)
the equation (1) represents an intensity error b | z-y | 1: m equal to the absolute value of the ratio of the difference bz 1: m-by 1: m between each calculated value of knock intensity and the corresponding actual value of knock intensity to the corresponding actual value of knock intensity by 1: m. For example, the intensity error b | z-y | 1 between the calculated knock intensity bz [1] and the actual knock intensity by [1] may be: b | z-y | 1 | (bz [1] -by [1])/by [1] |.
If the intensity error b | z-y | p: q is determined to exceed a specified threshold, where p and q are positive integers in the range of 1 to m, and q is greater than p. Thereby determining the knock intensity signal array bx [ p: q ] as a large error subset B ', that is, the large error subset B' ═ { bx [ p: q ], by [ p: q }. Wherein bx [ p: q ] represents a number of continuous or discontinuous knock intensity signals within the large error subset B' including the range of knock intensity signals bx [ p ] to bx [ q ]. by p q represents that the large error subset B' comprises a knock intensity actual value by p to a knock intensity actual value by q, and the knock intensity actual values are in one-to-one correspondence with knock intensity signals, namely the knock intensity signal bx p corresponds to the knock intensity actual value by p … …, and the knock intensity signal bx q corresponds to the knock intensity actual value by q.
In other embodiments, the calculation method of the intensity error can be set according to the needs.
Step S4: and determining the distribution rule in the large error subset.
Referring to fig. 6, which shows a schematic diagram of a process of determining a distribution rule of the large error subset B '═ { bx [ p: q ], by [ p: q ] }, it can be seen from fig. 6 that a mode value interval C of the knock intensity actual value by [ p: q ] can be determined as [ C1, C2] based on the distribution rule of the knock intensity actual value by [ p: q ] in the large error subset B'. In this embodiment, the mode value interval of the actual knock intensity by [ p: q ] may cover 80% to 90% of the knock intensity signals bx [ p: q ] in the large error subset B'.
Step S5: and determining a data subset in the training data set, wherein the data subset conforms to the distribution rule.
Fig. 7 is a schematic diagram illustrating a process of determining a data subset satisfying a distribution rule in the training data set a ═ { ax [1: n ], ay [1: n }. Continuing to take the training data set a ═ { ax [1: n ], ay [1: n ] } as an example, based on the distribution rule of the large error subset B 'determined in step S4, that is, the mode value interval C of the large error subset B' is [ C1, C2], determining that the training data set a ═ { ax [1: n ], ay [1: n ] }, and the knock intensity actual value ay [ y: w ] is located in the mode value interval C. In the training data set A, a knock intensity signal corresponding to the knock intensity actual value ay [ y: w ] is ax [ v: w ], where v and w are positive integers in the range of 1 to n, and w is greater than v. Namely, the data subset a' in the training data set a, which conforms to the distribution rule, is { ax [ v: w ], ay [ v: w }.
Step S6: the data of the subset of data is expanded at least 2 times based on an oversampling method such that the expanded data forms an expanded data set.
Please refer to fig. 8, which shows a schematic process diagram of a new extended data set D formed after expanding the data subset a 'by three times { ax [ v: w ], ay [ v: w ] }, wherein the extended data set D simply copies all data in the data subset a' three times.
In other embodiments, the expansion of the data subset may also be set according to the requirement of iteratively calculating the error magnitude, and a multiple that allows the trained model to obtain the minimum intensity error is found.
Step S7: and supplementing the extended data set into the training data set to form a new training data set.
Please refer to fig. 9, which shows a schematic process diagram of a new training data set E formed after appending the new extended data set D into the training data set a ═ ax [1: n ], ay [1: n ]. As can be seen in fig. 9, the new training data set E has more data parts than the original training data set a of the extended data set D.
As can be seen from the above description, the extended data set D is a simple copy of the data subset a '═ { ax [ v: w ], ay [ v: w ] }, and the mode value interval C of the data subset a' ═ ax [ v: w ], ay [ v: w ] } conforming to the large error subset B 'is [ C1, C2], that is, the data subset a' ═ { ax [ v: w ], ay [ v: w ] } conforming to the large error characteristic.
Step S8: and re-training a knock intensity calculation model based on the new training data set.
The new training data set E of the present embodiment is configured such that the knock intensity calculation model is newly trained from the original training data set E by adding the specific gravity of the subset having a large error characteristic in the original training data set a, that is, the specific gravity of the data subset a '═ ax [ v: w ], ay [ v: w ] } in the new training data set E is greater than the specific gravity of the data subset a' in the original training data set a, so that the knock intensity calculation model can have a smaller error when calculating the knock intensity of the knock intensity signal.
It should be understood that the above examples are only for clarity of illustration and are not intended to limit the embodiments. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. And obvious variations or modifications of this invention are intended to be covered by the scope of the invention as expressed herein.
Claims (10)
1. An engine knock intensity calculation method, characterized by comprising:
providing a training data set and a testing data set;
training a knock intensity calculation primary model based on the training data set;
calculating a primary model based on the knock intensity, computationally determining a large subset of errors in the test data set; the large error subset is a set of intensity errors between a calculated value of an engine knock intensity model and an actual value of the knock intensity in the test data set and knock intensity signals exceeding a certain threshold;
determining the distribution rule of the large error subsets;
determining a data subset in the training data set, wherein the data subset conforms to the distribution rule;
expanding the data of the subset of data by at least 2 times based on an oversampling method such that the newly expanded data forms an expanded data set;
adding the extended data set into the training data set, and completing the training data to form a new training data set;
and re-training a knock intensity calculation model based on the new training data set.
2. The engine knock intensity calculation method according to claim 1, wherein the test data set includes a plurality of knock intensity signals for testing, and knock intensity actual values corresponding to the knock intensity signals.
3. The engine knock intensity calculation method according to claim 2, wherein said calculating a primary model based on said knock intensity, and determining a large subset of errors in said test data set, said large subset of errors being a set of knock intensity signals having an error between a calculated value of a knock intensity model and an actual value of a knock intensity exceeding a specified threshold, comprises:
calculating a knock intensity calculation value for each of the knock intensity signals in the test data set based on the knock intensity calculation primary model;
calculating an error between the calculated knock value and the actual knock intensity corresponding to each knock intensity signal;
determining the set of knock intensity signals for which the engine intensity error exceeds a specified threshold as a large error subset.
4. The engine knock calculation method according to claim 1, wherein said step of counting the distribution law of said large error subset comprises:
and determining a mode value interval of the knock intensity actual values based on the knock intensity actual values of the knock intensity signals in the large error subset.
5. The engine knock intensity calculation method according to claim 4, wherein the number of knock intensity signals in the mode interval is 80% to 90% of the total number of knock intensity signals in the large error subset.
6. The engine knock intensity calculation method according to claim 4, wherein the training data set includes a plurality of knock intensity signals for training, and knock intensity actual values corresponding to the knock intensity signals.
7. The engine knock intensity calculation method of claim 6, wherein said step of determining a subset of data in said training data set that conforms to said distribution law comprises:
based on the training data set, determining the set of knock intensity signals for which the corresponding knock intensity actual values lie in the mode value interval as a data subset of the training data set.
8. The engine knock calculation method of claim 1, wherein said step of expanding the data of said subset of data by a factor of at least 2 based on oversampling, such that the newly expanded data forms an expanded data set, comprises:
and performing sampling replication on the data in the data subsets based on an oversampling method, so that the data in the data subsets are repeated by at least 2 times to form an expanded data set.
9. An engine knock intensity calculation system for executing the engine knock intensity calculation method according to any one of claims 1 to 8.
10. A readable storage medium in which a computer program is stored, the computer program being capable of implementing the engine knock intensity calculation method according to any one of claims 1 to 8 when executed by a processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110392081.4A CN113239732B (en) | 2021-04-13 | 2021-04-13 | Engine knock intensity calculation method, system and readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110392081.4A CN113239732B (en) | 2021-04-13 | 2021-04-13 | Engine knock intensity calculation method, system and readable storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113239732A true CN113239732A (en) | 2021-08-10 |
CN113239732B CN113239732B (en) | 2024-04-19 |
Family
ID=77128155
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110392081.4A Active CN113239732B (en) | 2021-04-13 | 2021-04-13 | Engine knock intensity calculation method, system and readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113239732B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007077968A (en) * | 2005-09-16 | 2007-03-29 | Denso Corp | Knock determining device for internal combustion engine |
US20160298553A1 (en) * | 2013-11-25 | 2016-10-13 | Sem Ab | Engine Management Using Knock Data |
CN111652327A (en) * | 2020-07-16 | 2020-09-11 | 北京思图场景数据科技服务有限公司 | Model iteration method, system and computer equipment |
-
2021
- 2021-04-13 CN CN202110392081.4A patent/CN113239732B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007077968A (en) * | 2005-09-16 | 2007-03-29 | Denso Corp | Knock determining device for internal combustion engine |
US20160298553A1 (en) * | 2013-11-25 | 2016-10-13 | Sem Ab | Engine Management Using Knock Data |
CN111652327A (en) * | 2020-07-16 | 2020-09-11 | 北京思图场景数据科技服务有限公司 | Model iteration method, system and computer equipment |
Non-Patent Citations (1)
Title |
---|
R.F.TURKSON;颜伏伍;侯献军;M.K.A.ALI;刘博;: "基于人工神经网络的发动机点火提前角预测模型", 汽车科技, no. 04 * |
Also Published As
Publication number | Publication date |
---|---|
CN113239732B (en) | 2024-04-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112629648B (en) | Ignition timing control device for internal combustion engine | |
US9316565B2 (en) | Exhaust manifold pressure based misfire detection for internal combustion engines | |
Gish et al. | Determination of true engine friction | |
JP5674903B1 (en) | In-cylinder pressure estimation device for internal combustion engine | |
CN112629649B (en) | Ignition timing control device for internal combustion engine | |
CN112629650B (en) | Ignition timing control device for internal combustion engine | |
CN111120094B (en) | Engine fire detection method and device, storage medium and terminal | |
CN111551368A (en) | Knock detection device and knock detection method for internal combustion engine | |
JP4453581B2 (en) | Engine control parameter experiment plan setting method, program for causing computer to execute the experiment plan setting method, computer-readable recording medium storing the program, and engine control parameter experiment plan setting apparatus | |
CN113239732A (en) | Engine knock intensity calculation method, system and readable storage medium | |
US9869257B2 (en) | System and method for predicting peak pressure values using knock sensor | |
US9303615B2 (en) | Ignition timing control device for an internal combustion engine | |
US10584653B2 (en) | Systems and methods for spark timing retardation | |
CN117150884A (en) | Engine fault investigation method and system based on digital twin | |
US6595043B2 (en) | Pressure measurement system | |
CN104153931B (en) | Method for detecting deflagration phenomenon in explosive motor | |
JP2007177723A (en) | Ignition timing control device for internal combustion engine | |
JP2000170589A (en) | Control device for internal combustion engine | |
CN111999067A (en) | Engine physical top dead center calibration measuring and calculating method, equipment and vehicle | |
JP2011111906A (en) | Diagnostic device for cylinder pressure sensor | |
KR102264304B1 (en) | Misfire diagnosis method and device of Multi cylinder four-stroke engine | |
KR102300965B1 (en) | Misfire diagnosis method and device of Multi cylinder four-stroke engine | |
CN117153282B (en) | Method, device, medium and equipment for calculating reference temperature of combustion chamber of gas turbine | |
KR102667018B1 (en) | Misfire diagnosis method and device of engine | |
CN115292807A (en) | Method and system for determining damping of battery pack, electronic device and storage medium |
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 |