WO2023051275A1 - Svm-based cold flow test detection method and system during diesel engine assembly - Google Patents

Svm-based cold flow test detection method and system during diesel engine assembly Download PDF

Info

Publication number
WO2023051275A1
WO2023051275A1 PCT/CN2022/119171 CN2022119171W WO2023051275A1 WO 2023051275 A1 WO2023051275 A1 WO 2023051275A1 CN 2022119171 W CN2022119171 W CN 2022119171W WO 2023051275 A1 WO2023051275 A1 WO 2023051275A1
Authority
WO
WIPO (PCT)
Prior art keywords
diesel engine
exhaust pressure
engine assembly
svm
support vector
Prior art date
Application number
PCT/CN2022/119171
Other languages
French (fr)
Chinese (zh)
Inventor
闫伟
王辉
李国祥
杨晓峰
孙俊伟
吴凡
李嘉颀
Original Assignee
山东大学
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 山东大学 filed Critical 山东大学
Priority to GB2318389.0A priority Critical patent/GB2622708A/en
Publication of WO2023051275A1 publication Critical patent/WO2023051275A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M15/00—Testing of engines
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M15/00—Testing of engines
    • G01M15/02—Details or accessories of testing apparatus
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M15/00—Testing of engines
    • G01M15/04—Testing internal-combustion engines
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M15/00—Testing of engines
    • G01M15/04—Testing internal-combustion engines
    • G01M15/05—Testing internal-combustion engines by combined monitoring of two or more different engine parameters
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines

Definitions

  • the invention relates to the field of diesel engines, in particular to an SVM-based diesel engine assembly cold test detection method and system.
  • Diesel engine is a very complex power machine, which can provide power source for various transportation equipment.
  • the hot test in the diesel engine assembly process is gradually replaced by the cold test. How to improve the assembly process?
  • the cold test detection technology has become the focus of research.
  • the exhaust parameter is one of the main parameters of diesel engine quality inspection, and the exhaust parameter in the cold test test is affected by the assembly parameters of the intake system, cylinder head body, piston and crankshaft connecting rod system.
  • the purpose of the present invention is to provide a diesel engine assembly cold test detection method based on SVM, which takes the exhaust pressure as the main parameter to determine its normal threshold range, and can be based on the characteristics of the input exhaust pressure
  • SVM diesel engine assembly cold test detection method
  • the vector classifies the quality of many parameters including air intake, crankshaft torque and exhaust, and can judge the quality of assembly according to the cold test parameters, which plays an extremely important role in improving the quality of diesel engine assembly.
  • the cold test detection method for diesel engine assembly based on SVM includes the following contents:
  • the threshold values of normal, small and large exhaust pressure are obtained
  • a sample set of diesel engine assembly quality inspection is constructed based on the cold test database, and the support vector machine algorithm (SVM) is used for training and testing to form a cold test test quality inspection support vector machine algorithm model.
  • SVM support vector machine algorithm
  • the assembly quality is identified, and an SVM-based cold test detection method for diesel engine assembly is formed.
  • the intake pressure, crankshaft torque and exhaust pressure of the diesel engine are measured by cold test equipment.
  • the cold test equipment obtains the intake pressure and exhaust pressure data of the diesel engine through the gas pressure sensor, and obtains the crankshaft torque data of the diesel engine through the torque sensor.
  • a diesel engine was obtained from the cold test database.
  • the above-mentioned SVM-based diesel engine assembly cold test detection method judges the distribution form of the exhaust pressure data by the ratio (Z-score) of the statistic of the parameter skewness value and the standard error;
  • the SVM-based diesel engine assembly cold test detection method adopts the 3 ⁇ principle of big data analysis to determine the normal, small and large thresholds of the exhaust pressure.
  • the numerical values of intake pressure and crankshaft torque are respectively normalized, combined with the classified exhaust pressure data, based on the cold test
  • the test database is used to build a sample set for diesel engine assembly quality inspection.
  • the SVM-based diesel engine assembly cold test detection method uses the radial basis kernel function to train and test the support vector machine algorithm model, and uses the cross-validation method to find the optimal penalty factor C and variance g, that is, in Within the set parameter range, determine the iteration length, combine the penalty factor and the variance in pairs, and select the combination with the highest accuracy in multiple groups of cross-validation.
  • the sample set of the diesel engine assembly quality inspection constructed is divided into a training set and a test set, and the training set is trained with a support vector machine algorithm to obtain a cold test test quality inspection
  • the support vector machine algorithm model uses the test set to determine the accuracy of the support vector machine algorithm model for assembly quality detection, thereby identifying the assembly quality of the diesel engine.
  • the present invention also provides an SVM-based diesel engine assembly cold test detection system, adopting the SVM-based diesel engine assembly cold test detection method, including:
  • Data processing unit use the intake pressure, crankshaft torque and exhaust pressure of the diesel engine to build a cold test test database, and use big data analysis to judge the distribution of exhaust pressure data in the assembly characteristic parameters; and according to the exhaust pressure data According to the distribution shape, the normal, small and large thresholds of assembly exhaust pressure are obtained, so as to construct a sample set for diesel engine assembly quality inspection;
  • Model building unit After determining the normal, small and large thresholds of the exhaust pressure, construct the support vector machine algorithm model, train and test the support vector machine algorithm model, and use the support vector machine algorithm model after the training test to assemble the diesel engine Quality is identified.
  • the present invention builds a cold test test database based on the intake pressure, crankshaft torque and exhaust pressure of the diesel engine, obtains the distribution form of the exhaust pressure data through a big data analysis method, and then determines multiple thresholds of the assembled exhaust pressure, It lays the foundation for the identification of assembly quality. After the threshold is determined, the feature vector is constructed and the support vector machine algorithm model is established, forming a diesel engine assembly cold test detection method.
  • the present invention trains and tests the support vector machine algorithm model by selecting a kernel function, looking for a penalty factor and a variance, thereby improving assembly quality and assembly precision, thereby improving assembly reliability and strong practicability.
  • the present invention can perform normalization processing on the feature vectors by assigning labels to the feature vectors for classification, which is beneficial to improving the accuracy of model establishment.
  • Fig. 1 is a flow chart of the SVM-based cold test detection method for diesel engine assembly according to one or more embodiments of the present invention.
  • the present invention proposes a diesel engine assembly cold test detection method based on SVM.
  • the diesel engine assembly cold test detection method based on SVM comprises the following contents:
  • the threshold values of normal, small and large exhaust pressure are obtained
  • a sample set of diesel engine assembly quality inspection is constructed based on the cold test database, and the support vector machine algorithm (SVM) is used for training and testing to form a cold test test quality inspection support vector machine algorithm model.
  • SVM support vector machine algorithm
  • the assembly quality is identified, and an SVM-based cold test detection method for diesel engine assembly is formed.
  • intake pressure, crankshaft torque and exhaust pressure are measured by cold test equipment
  • the cold test equipment obtains the intake pressure and exhaust pressure data of the diesel engine through the gas pressure sensor, obtains the crankshaft torque data of the diesel engine through the torque sensor, and obtains the cold test test database by collecting tens of thousands of diesel engines.
  • the distribution form of the exhaust pressure data is judged by the ratio (Z-score) of the statistic of the parameter skewness value to the standard error, and the calculation formula of the skewness value is The formula for calculating the standard error is where ⁇ is the sample mean and ⁇ is the sample standard deviation.
  • Z-score ⁇ [-2,2] the data is normally distributed; when Z-score>2, the data is positively skewed; when Z-score ⁇ -2, the data is negatively skewed .
  • the threshold range can be determined through the normal distribution, and the exhaust pressure is less than u-3 ⁇ or greater than u+3 ⁇ is an abnormal value; thus, according to different data distribution forms, the corresponding threshold determination method is given, which lays the foundation for the identification of assembly quality Base.
  • the 3 ⁇ principle of big data analysis is adopted to determine the thresholds of normal, low and high exhaust pressure.
  • the labels are 0, 1, and 2, where 0 represents the feature vector whose exhaust pressure is less than the minimum value of the threshold , 1 represents the eigenvector with the exhaust pressure within the threshold range, and 2 represents the eigenvector with the exhaust pressure greater than the maximum value of the threshold.
  • mapping operation normalize the values of the intake pressure and the crankshaft torque respectively, and map all their values into the range [-1,1].
  • the kernel function is selected as the radial basis function, and the formula is:
  • K(X i , X j ) exp(-g
  • Radial basis kernel function can effectively solve the problem of non-linear relationship between sample type and characteristic factors, and use cross-validation method to find the optimal penalty factor C and variance g, that is, within the set parameter range such as (-10, 10), Determine the iteration length, you can choose the iteration length to be 0.5, combine the penalty factor and the variance in pairs, and select the combination with the highest accuracy in the multi-group cross-validation.
  • the penalty factor C is selected as 0.33, and the variance g is 1.32.
  • the constructed sample set of diesel engine assembly quality inspection is divided into training set and test set, the training set is trained with support vector machine algorithm, and the support vector machine algorithm model of cold test test quality inspection is obtained, and the support vector machine algorithm model is determined by using the test set.
  • the algorithm model can detect the accuracy of the assembly quality, so as to identify the assembly quality of the diesel engine.
  • the exhaust pressure data of the diesel engine assembly is used as the basis to construct the support vector machine algorithm model, and the support vector machine algorithm model is trained and tested.
  • other assembly Characteristic parameters such as intake air and crankshaft torque can effectively ensure the accuracy of the support vector machine algorithm model for assembly quality detection.
  • the support vector machine algorithm model after training and testing can improve the accuracy of diesel engine assembly quality recognition, and correspondingly improve Reliability of diesel engine assembly.
  • This embodiment provides a diesel engine assembly cold test detection system based on SVM, using the SVM-based diesel engine assembly cold test detection method, including:
  • Data processing unit use the intake pressure, crankshaft torque and exhaust pressure of the diesel engine to build a cold test test database, and use big data analysis to judge the distribution of exhaust pressure data in the assembly characteristic parameters; and according to the exhaust pressure data According to the distribution shape, the normal, small and large thresholds of assembly exhaust pressure are obtained, so as to construct a sample set for diesel engine assembly quality inspection;
  • Model building unit After determining the normal, small and large thresholds of the exhaust pressure, construct the support vector machine algorithm model, train and test the support vector machine algorithm model, and use the support vector machine algorithm model after the training test to assemble the diesel engine The quality is identified, and the SVM-based diesel engine assembly cold test detection system is formed through the data processing unit and the model building unit.
  • the SVM-based diesel engine assembly cold test detection system is stored by a storage device such as a computer.
  • model building unit is based on the determined normal, small and large thresholds of the exhaust pressure.
  • 0 represents the eigenvector whose exhaust pressure is less than the minimum value of the threshold
  • 1 represents the eigenvector whose exhaust pressure is within the threshold range
  • 2 represents the eigenvector whose exhaust pressure is greater than the maximum value of the threshold.
  • mapping operation normalize the values of the intake pressure and the crankshaft torque respectively, and map all their values into the range [-1,1].
  • the distribution form of the exhaust pressure data is judged by the ratio (Z-score) of the statistic of the parameter skewness value to the standard error.
  • Z-score the ratio of the statistic of the parameter skewness value to the standard error.
  • the SVM model is trained and tested by selecting the radial basis function; the kernel function is a radial basis function, and the cross-validation method is used to find the optimal penalty factor C and variance g, that is, within the set parameter range such as (-10, 10) , to determine the iteration length, the iteration length can be selected as 0.5, the penalty factor and the variance are combined in pairs, and the combination with the highest accuracy is selected in the multi-group cross-validation.
  • the constructed sample set of diesel engine assembly quality inspection is divided into training set and test set, the training set is trained with support vector machine algorithm, and the support vector machine algorithm model of cold test test quality inspection is obtained, and the support vector machine algorithm model is determined by using the test set.
  • the algorithm model can detect the accuracy of the assembly quality, so as to identify the assembly quality of the diesel engine.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Combustion & Propulsion (AREA)
  • Chemical & Material Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Testing Of Engines (AREA)

Abstract

An SVM-based cold flow test detection method and system during diesel engine assembly. The method comprises: acquiring the intake pressure, crankshaft torque and exhaust pressure of a diesel engine, and constructing a cold flow test database; determining a distribution pattern of exhaust pressure data by means of bigdata analysis; according to the distribution pattern of the exhaust pressure data, obtaining threshold values for a normal exhaust pressure, a smaller exhaust pressure and a larger exhaust pressure; and after the threshold values are determined, constructing a sample set for diesel engine assembly quality detection on the basis of the cold flow test database, performing training and testing by using a support vector machine (SVM) algorithm, so as to form a cold flow test quality detection support vector machine algorithm model, and identifying the diesel engine assembly quality by means of the support vector machine algorithm model, such that an SVM-based cold flow test detection method during diesel engine assembly is formed. By means of the detection method and system, the accuracy of detection of diesel engine assembly quality is improved, thereby effectively identifying diesel engine assembly quality.

Description

基于SVM的柴油机装配冷试检测方法及***Method and system for cold test detection of diesel engine assembly based on SVM
本发明要求于2021年09月29日提交中国专利局、申请号为202111153792.2、发明名称为“基于SVM的柴油机装配冷试检测方法及***”的中国专利申请的优先权，其全部内容通过引用结合在本发明中。The present invention claims the priority of the Chinese patent application submitted to the China Patent Office on September 29, 2021, with the application number 202111153792.2, and the title of the invention is "SVM-based diesel engine assembly cold test detection method and system", the entire content of which is incorporated by reference In the present invention.
技术领域technical field
本发明涉及柴油机领域，尤其是基于SVM的柴油机装配冷试检测方法及***。The invention relates to the field of diesel engines, in particular to an SVM-based diesel engine assembly cold test detection method and system.
背景技术Background technique
本部分的陈述仅仅是提供了与本发明相关的背景技术信息，不必然构成在先技术。The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.
柴油机是一种十分复杂的动力机械，可为各种交通设备提供动力来源，随着社会对环境保护的需求越来越高，柴油机装配过程中热试逐渐被冷试所替代，如何提高装配过程中的冷试检测技术成为研究的重点。其中，排气参数是柴油机质量检测的主要参数之一，冷试测试中的排气参数受进气***、缸盖机体、活塞和曲轴连杆***装配参数的影响。发明人发现现有技术中没有针对柴油机排气分布特征来判断装配质量的方法或装置，对可能由装配引起的柴油机的故障无法准确控制。Diesel engine is a very complex power machine, which can provide power source for various transportation equipment. With the increasing demand for environmental protection in society, the hot test in the diesel engine assembly process is gradually replaced by the cold test. How to improve the assembly process? The cold test detection technology has become the focus of research. Among them, the exhaust parameter is one of the main parameters of diesel engine quality inspection, and the exhaust parameter in the cold test test is affected by the assembly parameters of the intake system, cylinder head body, piston and crankshaft connecting rod system. The inventors found that there is no method or device for judging the assembly quality based on the exhaust gas distribution characteristics of the diesel engine in the prior art, and it is impossible to accurately control the failure of the diesel engine that may be caused by the assembly.
发明内容Contents of the invention
针对现有技术存在的不足，本发明的目的是提供基于SVM的柴油机装配冷试检测方法，以排气压力为主要参数，确定其正常的阈值范围，并能根据输入的排气压力相关的特征向量将包括进气、曲轴转矩和排气等众多参数的好坏进行分类，可根据冷试参数判断装配质量的好坏，对于提高柴油机的装配质量具有极其重要的作用。In view of the deficiencies in the prior art, the purpose of the present invention is to provide a diesel engine assembly cold test detection method based on SVM, which takes the exhaust pressure as the main parameter to determine its normal threshold range, and can be based on the characteristics of the input exhaust pressure The vector classifies the quality of many parameters including air intake, crankshaft torque and exhaust, and can judge the quality of assembly according to the cold test parameters, which plays an extremely important role in improving the quality of diesel engine assembly.
为了实现上述目的，本发明是通过如下的技术方案来实现：In order to achieve the above object, the present invention is achieved through the following technical solutions:
基于SVM的柴油机装配冷试检测方法，包括如下内容：The cold test detection method for diesel engine assembly based on SVM includes the following contents:
获取柴油机的进气压力、曲轴转矩和排气压力，构建冷试测试数据库；Obtain the intake pressure, crankshaft torque and exhaust pressure of the diesel engine, and build a cold test database;
采用大数据分析的方式判断排气压力数据的分布形态；Use big data analysis to judge the distribution pattern of exhaust pressure data;
根据排气压力数据的分布形态，得到排气压力正常、偏小和偏大的阈值；According to the distribution form of the exhaust pressure data, the threshold values of normal, small and large exhaust pressure are obtained;
确定阈值后，基于冷试测试数据库构建柴油机装配质量检测的样本集，采用支持向量机算法(SVM)训练并测试，形成冷试测试质量检测支持向量机算法模型，通过支持向量机算法模型对柴油机装配质量进行识别，形成基于SVM的柴油机装配冷试检测方法。After the threshold is determined, a sample set of diesel engine assembly quality inspection is constructed based on the cold test database, and the support vector machine algorithm (SVM) is used for training and testing to form a cold test test quality inspection support vector machine algorithm model. The assembly quality is identified, and an SVM-based cold test detection method for diesel engine assembly is formed.
如上所述的基于SVM的柴油机装配冷试检测方法，所述柴油机的进气压力、曲轴转矩和排气压力由冷试测试设备测得。According to the SVM-based cold test detection method for diesel engine assembly, the intake pressure, crankshaft torque and exhaust pressure of the diesel engine are measured by cold test equipment.
如上所述的基于SVM的柴油机装配冷试检测方法，所冷试测试设备通过气体压力传感器获取柴油机的进气压力和排气压力数据，通过扭矩传感器获取柴油机的曲轴转矩数据，通过采集数万台柴油机获得冷试测试数据库。In the SVM-based cold test detection method for diesel engine assembly as described above, the cold test equipment obtains the intake pressure and exhaust pressure data of the diesel engine through the gas pressure sensor, and obtains the crankshaft torque data of the diesel engine through the torque sensor. A diesel engine was obtained from the cold test database.
如上所述的基于SVM的柴油机装配冷试检测方法，通过参数偏度值的统计量与标准误差的比值(Z-score)来判断所述排气压力数据的分布形态；The above-mentioned SVM-based diesel engine assembly cold test detection method judges the distribution form of the exhaust pressure data by the ratio (Z-score) of the statistic of the parameter skewness value and the standard error;
如上所述的基于SVM的柴油机装配冷试检测方法，当参数偏度值的统计量与标准误差的比值∈[-2,2]时，则数据为正态分布；当参数偏度值的统计量与标准误差的比值>2时，数据为正偏态分布；当参数偏度值的统计量与标准误差的比值<-2时，数据为负偏态分布。In the above-mentioned SVM-based diesel engine assembly cold test detection method, when the ratio of the parameter skewness value statistic to the standard error ∈ [-2,2], the data is normally distributed; when the parameter skewness value statistic When the ratio of the parameter skewness value to the standard error is >2, the data is positively skewed; when the ratio of the parameter skewness value to the standard error is <-2, the data is negatively skewed.
如上所述的基于SVM的柴油机装配冷试检测方法，采用大数据分析的3σ原则，确定所述的排气压力正常、偏小和偏大的阈值。As mentioned above, the SVM-based diesel engine assembly cold test detection method adopts the 3σ principle of big data analysis to determine the normal, small and large thresholds of the exhaust pressure.
如上所述的基于SVM的柴油机装配冷试检测方法，将进气压力的数值和曲轴转矩的数值分别作归一化处理，与分类后的排气压力数据相结合，基于所述的冷试测试数据库，构建柴油机装配质量检测的样本集。In the above-mentioned SVM-based diesel engine assembly cold test detection method, the numerical values of intake pressure and crankshaft torque are respectively normalized, combined with the classified exhaust pressure data, based on the cold test The test database is used to build a sample set for diesel engine assembly quality inspection.
如上所述的基于SVM的柴油机装配冷试检测方法，通过选取径向基核函数来训练测试所述的支持向量机算法模型，运用交叉验证法寻找最优的惩罚因子C和方差g，即在设定的参数范围内，确定迭代长度，将惩罚因子和方差两两组合，在多组交叉验证中选取精度最高的组合。As mentioned above, the SVM-based diesel engine assembly cold test detection method uses the radial basis kernel function to train and test the support vector machine algorithm model, and uses the cross-validation method to find the optimal penalty factor C and variance g, that is, in Within the set parameter range, determine the iteration length, combine the penalty factor and the variance in pairs, and select the combination with the highest accuracy in multiple groups of cross-validation.
如上所述的基于SVM的柴油机装配冷试检测方法，将构建的所述柴油机装配质量检测的样本集划分为训练集和测试集，用支持向量机算法对训练集训练，得到冷试测试质量检测支持向量机算法模型，利用测试集确定支持向量机算法模型对装配质量检测的准确性，从而对柴油机装配质量进行识别。In the above-mentioned SVM-based diesel engine assembly cold test detection method, the sample set of the diesel engine assembly quality inspection constructed is divided into a training set and a test set, and the training set is trained with a support vector machine algorithm to obtain a cold test test quality inspection The support vector machine algorithm model uses the test set to determine the accuracy of the support vector machine algorithm model for assembly quality detection, thereby identifying the assembly quality of the diesel engine.
第二方面，本发明还提供了基于SVM的柴油机装配冷试检测***，采用所述的基于SVM的柴油机装配冷试检测方法，包括：In the second aspect, the present invention also provides an SVM-based diesel engine assembly cold test detection system, adopting the SVM-based diesel engine assembly cold test detection method, including:
数据处理单元：利用柴油机的进气压力、曲轴转矩和排气压力构建冷试测试数据库，采用大数据分析的方式判断装配特征参数中排气压力数据的分布形态；并根据排气压力数据的分布形态，得到装配排气压力正常、偏小和偏大的阈值，从而构建柴油机装配质量检测的样本集；Data processing unit: use the intake pressure, crankshaft torque and exhaust pressure of the diesel engine to build a cold test test database, and use big data analysis to judge the distribution of exhaust pressure data in the assembly characteristic parameters; and according to the exhaust pressure data According to the distribution shape, the normal, small and large thresholds of assembly exhaust pressure are obtained, so as to construct a sample set for diesel engine assembly quality inspection;
模型建立单元：确定排气压力正常、偏小和偏大的阈值后，构建建立支持向量机算法模型，训练测试支持向量机算法模型，通过训练测试后的 支持向量机算法模型用于对柴油机装配质量进行识别。Model building unit: After determining the normal, small and large thresholds of the exhaust pressure, construct the support vector machine algorithm model, train and test the support vector machine algorithm model, and use the support vector machine algorithm model after the training test to assemble the diesel engine Quality is identified.
上述本发明的有益效果如下：The above-mentioned beneficial effects of the present invention are as follows:
1)本发明基于柴油机的进气压力、曲轴转矩和排气压力，构建冷试测试数据库，通过大数据分析方法得到排气压力数据的分布形态，进而确定装配排气压力的多个阈值，为装配质量的识别奠定了基础，在确定阈值后构建特征向量并建立支持向量机算法模型，形成了柴油机装配冷试检测方法。1) The present invention builds a cold test test database based on the intake pressure, crankshaft torque and exhaust pressure of the diesel engine, obtains the distribution form of the exhaust pressure data through a big data analysis method, and then determines multiple thresholds of the assembled exhaust pressure, It lays the foundation for the identification of assembly quality. After the threshold is determined, the feature vector is constructed and the support vector machine algorithm model is established, forming a diesel engine assembly cold test detection method.
2)本发明通过选取核函数，寻找惩罚因子和方差，来训练测试支持向量机算法模型，提高了装配质量和装配精度，从而提高了装配的可靠性，实用性强。2) The present invention trains and tests the support vector machine algorithm model by selecting a kernel function, looking for a penalty factor and a variance, thereby improving assembly quality and assembly precision, thereby improving assembly reliability and strong practicability.
3)本发明通过对特征向量给予标签进行分类，能够对特征向量进行归一化处理，有利于提高模型建立的精度。3) The present invention can perform normalization processing on the feature vectors by assigning labels to the feature vectors for classification, which is beneficial to improving the accuracy of model establishment.
附图说明Description of drawings
构成本发明的一部分的说明书附图用来提供对本发明的进一步理解，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention, and the schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations to the present invention.
图1是本发明根据一个或多个实施方式的基于SVM的柴油机装配冷试检测方法的流程图。Fig. 1 is a flow chart of the SVM-based cold test detection method for diesel engine assembly according to one or more embodiments of the present invention.
具体实施方式Detailed ways
应该指出，以下详细说明都是例示性的，旨在对本发明提供进一步的说明。除非另有指明，本发明使用的所有技术和科学术语具有与本发明所属技术领域的普通技术人员通常理解的相同含义。It should be noted that the following detailed description is exemplary and intended to provide further explanation of the present invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
需要注意的是，这里所使用的术语仅是为了描述具体实施方式，而非意图限制根据本发明的示例性实施方式。如在这里所使用的，除非本发明另外明确指出，否则单数形式也意图包括复数形式，此外，还应当理解的是，当在本说明书中使用术语“包含”和/或“包括”时，其指明存在特征、步骤、操作、器件、组件和/或它们的组合；It should be noted that the terminology used here is only for describing specific embodiments, and is not intended to limit exemplary embodiments according to the present invention. As used herein, unless the invention clearly states otherwise, the singular form is also intended to include the plural form. In addition, it should also be understood that when the terms "comprising" and/or "comprising" are used in this specification, their Indicate the presence of features, steps, operations, means, components and/or combinations thereof;
正如背景技术所介绍的，现有技术中存在柴油机装配质量检测无法进行准确识别的问题，为了解决如上的技术问题，本发明提出了基于SVM的柴油机装配冷试检测方法。As introduced in the background technology, there is a problem in the prior art that diesel engine assembly quality detection cannot be accurately identified. In order to solve the above technical problems, the present invention proposes a diesel engine assembly cold test detection method based on SVM.
本发明的一种典型的实施方式中，参考图1所示，基于SVM的柴油机装配冷试检测方法，包括如下内容：In a typical implementation of the present invention, with reference to shown in Fig. 1, the diesel engine assembly cold test detection method based on SVM comprises the following contents:
获取柴油机的进气压力、曲轴转矩和排气压力，构建冷试测试数据库；Obtain the intake pressure, crankshaft torque and exhaust pressure of the diesel engine, and build a cold test database;
采用大数据分析的方式判断排气压力数据的分布形态；Use big data analysis to judge the distribution pattern of exhaust pressure data;
根据排气压力数据的分布形态，得到排气压力正常、偏小和偏大的阈值；According to the distribution form of the exhaust pressure data, the threshold values of normal, small and large exhaust pressure are obtained;
确定阈值后，基于冷试测试数据库构建柴油机装配质量检测的样本集，采用支持向量机算法(SVM)训练并测试，形成冷试测试质量检测支持向量机算法模型，通过支持向量机算法模型对柴油机装配质量进行识别，形成基于SVM的柴油机装配冷试检测方法。After the threshold is determined, a sample set of diesel engine assembly quality inspection is constructed based on the cold test database, and the support vector machine algorithm (SVM) is used for training and testing to form a cold test test quality inspection support vector machine algorithm model. The assembly quality is identified, and an SVM-based cold test detection method for diesel engine assembly is formed.
进一步地，进气压力、曲轴转矩和排气压力由冷试测试设备测得；Further, intake pressure, crankshaft torque and exhaust pressure are measured by cold test equipment;
具体地，冷试测试设备通过气体压力传感器获取柴油机的进气压力和排气压力数据，通过扭矩传感器获取柴油机的曲轴转矩数据，通过采集数万台柴油机获得冷试测试数据库。Specifically, the cold test equipment obtains the intake pressure and exhaust pressure data of the diesel engine through the gas pressure sensor, obtains the crankshaft torque data of the diesel engine through the torque sensor, and obtains the cold test test database by collecting tens of thousands of diesel engines.
其中，对于排气压力数据的分布形态，通过参数偏度值的统计量与标准误差的比值(Z-score)的值来判断排气压力数据的分布形态，其中偏度值的计算公式为
Figure PCTCN2022119171-appb-000001
标准误差的计算公式为
Figure PCTCN2022119171-appb-000002
式中μ表示样本均值，σ表示样本标准差。当Z-score∈[-2,2]时，则数据为正态分布；当Z-score>2时，数据为正偏态分布；当Z-score<-2时，数据为负偏态分布。 Among them, for the distribution form of the exhaust pressure data, the distribution form of the exhaust pressure data is judged by the ratio (Z-score) of the statistic of the parameter skewness value to the standard error, and the calculation formula of the skewness value is
Figure PCTCN2022119171-appb-000001
The formula for calculating the standard error is
Figure PCTCN2022119171-appb-000002
where μ is the sample mean and σ is the sample standard deviation. When Z-score∈[-2,2], the data is normally distributed; when Z-score>2, the data is positively skewed; when Z-score<-2, the data is negatively skewed .
对于数据为正态分布时，确定平均值u、标准差σ，依据(u-3σ，u+3σ]，得到排气压力正常参数范围；对于数据为正偏态分布，当Z-score∈(2,3]时，对此类数据需整体进行开平方根的处理，即
Figure PCTCN2022119171-appb-000003
当Z-score>3时，对此类数据可以进行的操作是取自然对数(ln)，即X new＝lnX，转换为正态分布；对于数据为负偏态分布，当Z-score∈(-3,-2]时，需要用到的转换公式为
Figure PCTCN2022119171-appb-000004
当Z-score<-3时，需要用到的转换公式为X new＝ln(X max+1-X)，转换为正态分布。通过正态分布可以确定阈值范围，排气压力小于u-3σ或大于u+3σ即为异常值；这样针对不同的数据分布形态，给出了相应的阈值确定方法，为装配质量的识别奠定了基础。 When the data is normally distributed, determine the average value u and standard deviation σ, and get the normal parameter range of exhaust pressure according to (u-3σ, u+3σ]; for the data with positive skewed distribution, when Z-score∈( 2,3], this kind of data needs to be processed by the square root as a whole, that is,
Figure PCTCN2022119171-appb-000003
When Z-score>3, the operation that can be performed on this type of data is to take the natural logarithm (ln), that is, X new = lnX, and convert it to a normal distribution; for the data to be negatively skewed, when Z-score∈ (-3,-2], the conversion formula to be used is
Figure PCTCN2022119171-appb-000004
When Z-score<-3, the conversion formula to be used is X new =ln(X max +1-X), which is converted to a normal distribution. The threshold range can be determined through the normal distribution, and the exhaust pressure is less than u-3σ or greater than u+3σ is an abnormal value; thus, according to different data distribution forms, the corresponding threshold determination method is given, which lays the foundation for the identification of assembly quality Base.
本实施例中，采用大数据分析的3σ原则，确定排气压力正常、偏小和偏大的阈值。In this embodiment, the 3σ principle of big data analysis is adopted to determine the thresholds of normal, low and high exhaust pressure.
其中，构建支持向量机算法模型中，需要构建特征向量，并将特征向量给予标签进行分类，一些示例中，标签分别为0、1、2，其中0表示排气压力小于阈值最小值的特征向量，1表示排气压力在阈值范围内的特征向量，2表示排气压力大于阈值最大值的特征向量。Among them, in constructing the support vector machine algorithm model, it is necessary to construct feature vectors, and assign labels to the feature vectors for classification. In some examples, the labels are 0, 1, and 2, where 0 represents the feature vector whose exhaust pressure is less than the minimum value of the threshold , 1 represents the eigenvector with the exhaust pressure within the threshold range, and 2 represents the eigenvector with the exhaust pressure greater than the maximum value of the threshold.
进一步地，将进气压力的数值和曲轴转矩的数值分别作归一化处理，将其所有的值映射到范围[-1,1]中，映射操作的公式为 y＝2*(x-x min)/(x max-x min)-1，与分类后的排气压力数据相结合，基于所述的冷试测试数据库，构建柴油机装配质量检测的样本集。 Further, normalize the values of the intake pressure and the crankshaft torque respectively, and map all their values into the range [-1,1]. The formula for the mapping operation is y=2*(xx min )/(x max -x min )-1, combined with the classified exhaust pressure data, based on the cold test test database, a sample set for diesel engine assembly quality inspection is constructed.
进一步地，通过选取核函数来训练测试支持向量机算法模型；本实施例中，选取核函数为径向基函数，公式为：Further, train and test the support vector machine algorithm model by selecting the kernel function; in the present embodiment, the kernel function is selected as the radial basis function, and the formula is:
Figure PCTCN2022119171-appb-000005
Figure PCTCN2022119171-appb-000005
式中K(X i,X j)＝exp(-g||X i-X j|| 2),g＞0为径向基核函数，ε表示不敏感损失系数。径向基核函数能够有效解决样本类型与特征因素是非线性关系的问题，运用交叉验证法寻找最优的惩罚因子C和方差g，即在设定的参数范围如(-10，10)内，确定迭代长度，可选择迭代长度为0.5，将惩罚因子和方差两两组合，在多组交叉验证中选取精度最高的组合。本实施例中，选择惩罚因子C为0.33，方差g为1.32。 In the formula, K(X i , X j )=exp(-g||X i -X j || 2 ), g>0 is the radial basis kernel function, and ε represents the insensitive loss coefficient. Radial basis kernel function can effectively solve the problem of non-linear relationship between sample type and characteristic factors, and use cross-validation method to find the optimal penalty factor C and variance g, that is, within the set parameter range such as (-10, 10), Determine the iteration length, you can choose the iteration length to be 0.5, combine the penalty factor and the variance in pairs, and select the combination with the highest accuracy in the multi-group cross-validation. In this embodiment, the penalty factor C is selected as 0.33, and the variance g is 1.32.
进一步地，将构建的柴油机装配质量检测的样本集划分为训练集和测试集，用支持向量机算法对训练集训练，得到冷试测试质量检测支持向量机算法模型，利用测试集确定支持向量机算法模型对装配质量检测的准确性，从而对柴油机装配质量进行识别。Further, the constructed sample set of diesel engine assembly quality inspection is divided into training set and test set, the training set is trained with support vector machine algorithm, and the support vector machine algorithm model of cold test test quality inspection is obtained, and the support vector machine algorithm model is determined by using the test set. The algorithm model can detect the accuracy of the assembly quality, so as to identify the assembly quality of the diesel engine.
通过本发明提供的基于SVM的柴油机装配冷试检测方法，将柴油机装配的排气压力数据作为基础来构建支持向量机算法模型，并对支持向量机算法模型进行训练测试，此外，还考虑其他装配特征参数如进气、曲轴转矩，有效保证支持向量机算法模型对装配质量检测的准确性，通过训练测试后的支持向量机算法模型，可提高对柴油机装配质量识别的准确度，相 应提高了柴油机装配的可靠性。Through the SVM-based diesel engine assembly cold test detection method provided by the present invention, the exhaust pressure data of the diesel engine assembly is used as the basis to construct the support vector machine algorithm model, and the support vector machine algorithm model is trained and tested. In addition, other assembly Characteristic parameters such as intake air and crankshaft torque can effectively ensure the accuracy of the support vector machine algorithm model for assembly quality detection. The support vector machine algorithm model after training and testing can improve the accuracy of diesel engine assembly quality recognition, and correspondingly improve Reliability of diesel engine assembly.
实施例二Embodiment two
本实施例提供了基于SVM的柴油机装配冷试检测***，采用所述的基于SVM的柴油机装配冷试检测方法，包括：This embodiment provides a diesel engine assembly cold test detection system based on SVM, using the SVM-based diesel engine assembly cold test detection method, including:
数据处理单元：利用柴油机的进气压力、曲轴转矩和排气压力构建冷试测试数据库，采用大数据分析的方式判断装配特征参数中排气压力数据的分布形态；并根据排气压力数据的分布形态，得到装配排气压力正常、偏小和偏大的阈值，从而构建柴油机装配质量检测的样本集；Data processing unit: use the intake pressure, crankshaft torque and exhaust pressure of the diesel engine to build a cold test test database, and use big data analysis to judge the distribution of exhaust pressure data in the assembly characteristic parameters; and according to the exhaust pressure data According to the distribution shape, the normal, small and large thresholds of assembly exhaust pressure are obtained, so as to construct a sample set for diesel engine assembly quality inspection;
模型建立单元：确定排气压力正常、偏小和偏大的阈值后，构建建立支持向量机算法模型，训练测试支持向量机算法模型，通过训练测试后的支持向量机算法模型用于对柴油机装配质量进行识别，由此通过数据处理单元和模型建立单元形成基于SVM的柴油机装配冷试检测***。Model building unit: After determining the normal, small and large thresholds of the exhaust pressure, construct the support vector machine algorithm model, train and test the support vector machine algorithm model, and use the support vector machine algorithm model after the training test to assemble the diesel engine The quality is identified, and the SVM-based diesel engine assembly cold test detection system is formed through the data processing unit and the model building unit.
可以理解地是，通过存储设备如计算机对基于SVM的柴油机装配冷试检测***进行存储。It can be understood that the SVM-based diesel engine assembly cold test detection system is stored by a storage device such as a computer.
进一步地，模型建立单元根据确定的排气压力正常、偏小和偏大的阈值。Further, the model building unit is based on the determined normal, small and large thresholds of the exhaust pressure.
其中，构建支持向量机算法模型中，需要构建特征向量，并将特征向量给予0、1、2的标签进行分类。其中0表示排气压力小于阈值最小值的特征向量，1表示排气压力在阈值范围内的特征向量，2表示排气压力大于阈值最大值的特征向量。Among them, in constructing the support vector machine algorithm model, it is necessary to construct feature vectors, and assign labels of 0, 1, and 2 to the feature vectors for classification. Among them, 0 represents the eigenvector whose exhaust pressure is less than the minimum value of the threshold, 1 represents the eigenvector whose exhaust pressure is within the threshold range, and 2 represents the eigenvector whose exhaust pressure is greater than the maximum value of the threshold.
进一步地，将进气压力的数值和曲轴转矩的数值分别作归一化处理，将其所有的值映射到范围[-1,1]中，映射操作的公式为 y＝2*(x-x min)/(x max-x min)-1，与分类后的排气压力数据相结合，基于所述的冷试测试数据库，构建所述的柴油机装配质量检测的样本集。 Further, normalize the values of the intake pressure and the crankshaft torque respectively, and map all their values into the range [-1,1]. The formula for the mapping operation is y=2*(xx min )/(x max -x min )-1, combined with the classified exhaust pressure data, based on the cold test test database, construct the sample set for the diesel engine assembly quality inspection.
对于排气压力数据的分布形态，通过参数偏度值的统计量与标准误差的比值(Z-score)来判断排气压力数据的分布形态，当Z-score∈[-2,2]时，则数据为正态分布；当Z-score>2时，数据为正偏态分布；当Z-score<-2时，数据为负偏态分布。For the distribution form of the exhaust pressure data, the distribution form of the exhaust pressure data is judged by the ratio (Z-score) of the statistic of the parameter skewness value to the standard error. When Z-score∈[-2,2], Then the data is normally distributed; when Z-score>2, the data is positively skewed; when Z-score<-2, the data is negatively skewed.
通过选取径向基函数来训练测试SVM模型；核函数为径向基函数，运用交叉验证法寻找最优的惩罚因子C和方差g，即在设定的参数范围如(-10，10)内，确定迭代长度，可选择迭代长度为0.5，将惩罚因子和方差两两组合，在多组交叉验证中选取精度最高的组合。The SVM model is trained and tested by selecting the radial basis function; the kernel function is a radial basis function, and the cross-validation method is used to find the optimal penalty factor C and variance g, that is, within the set parameter range such as (-10, 10) , to determine the iteration length, the iteration length can be selected as 0.5, the penalty factor and the variance are combined in pairs, and the combination with the highest accuracy is selected in the multi-group cross-validation.
进一步地，将构建的柴油机装配质量检测的样本集划分为训练集和测试集，用支持向量机算法对训练集训练，得到冷试测试质量检测支持向量机算法模型，利用测试集确定支持向量机算法模型对装配质量检测的准确性，从而对柴油机装配质量进行识别。Further, the constructed sample set of diesel engine assembly quality inspection is divided into training set and test set, the training set is trained with support vector machine algorithm, and the support vector machine algorithm model of cold test test quality inspection is obtained, and the support vector machine algorithm model is determined by using the test set. The algorithm model can detect the accuracy of the assembly quality, so as to identify the assembly quality of the diesel engine.
以上所述仅为本发明的优选实施例而已，并不用于限制本发明，对于本领域的技术人员来说，本发明可以有各种更改和变化。凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (6)

  1. 基于SVM的柴油机装配冷试检测方法，其特征在于，包括如下内容：The cold test detection method for diesel engine assembly based on SVM is characterized in that it includes the following contents:
    获取柴油机的进气压力、曲轴转矩和排气压力，构建冷试测试数据库；Obtain the intake pressure, crankshaft torque and exhaust pressure of the diesel engine, and build a cold test database;
    采用大数据分析的方式判断排气压力数据的分布形态；Use big data analysis to judge the distribution pattern of exhaust pressure data;
    根据排气压力数据的分布形态，得到排气压力正常、偏小和偏大的阈值；According to the distribution form of the exhaust pressure data, the threshold values of normal, small and large exhaust pressure are obtained;
    确定阈值后，基于冷试测试数据库构建柴油机装配质量检测的样本集，采用支持向量机算法训练并测试，形成冷试测试质量检测支持向量机算法模型，通过支持向量机算法模型对柴油机装配质量进行识别，形成基于SVM的柴油机装配冷试检测方法；After the threshold is determined, a sample set of diesel engine assembly quality inspection is constructed based on the cold test database, and the support vector machine algorithm is used for training and testing to form a cold test test quality inspection support vector machine algorithm model, and the diesel engine assembly quality is evaluated through the support vector machine algorithm model. Identify and form an SVM-based cold test detection method for diesel engine assembly;
    通过参数偏度值的统计量与标准误差的比值来判断所述排气压力数据的分布形态；The distribution form of the exhaust pressure data is judged by the ratio of the statistic of the parameter skewness value to the standard error;
    当参数偏度值的统计量与标准误差的比值∈[-2,2]时，则数据为正态分布；当参数偏度值的统计量与标准误差的比值>2时，数据为正偏态分布；当参数偏度值的统计量与标准误差的比值<-2时，数据为负偏态分布；When the ratio of the statistic of the parameter skewness value to the standard error ∈ [-2,2], the data is normally distributed; when the ratio of the statistic of the parameter skewness value to the standard error > 2, the data is positively skewed When the ratio of the parameter skewness value to the standard error is <-2, the data is negatively skewed;
    采用大数据分析的3σ原则，确定所述的排气压力正常、偏小和偏大的阈值；Use the 3σ principle of big data analysis to determine the normal, small and large thresholds of the exhaust pressure;
    将进气压力的数值和曲轴转矩的数值分别作归一化处理，与分类后的排气压力数据相结合，基于所述的冷试测试数据库，构建柴油机装配质量检测的样本集。The values of intake pressure and crankshaft torque are normalized respectively, combined with the classified exhaust pressure data, and based on the cold test test database, a sample set for diesel engine assembly quality inspection is constructed.
  2. 根据权利要求1所述的基于SVM的柴油机装配冷试检测方法，其特征在于，所述柴油机的进气压力、曲轴转矩和排气压力由冷试测试设备 测得。The SVM-based diesel engine assembly cold test detection method according to claim 1, wherein the intake pressure, crankshaft torque and exhaust pressure of the diesel engine are measured by cold test equipment.
  3. 根据权利要求2所述的基于SVM的柴油机装配冷试检测方法，其特征在于，所述冷试测试设备通过气体压力传感器获取柴油机的进气压力和排气压力数据，通过扭矩传感器获取柴油机的曲轴转矩数据。The SVM-based diesel engine assembly cold test detection method according to claim 2, wherein the cold test equipment obtains the intake pressure and exhaust pressure data of the diesel engine through a gas pressure sensor, and obtains the crankshaft of the diesel engine through a torque sensor Torque data.
  4. 根据权利要求1所述的基于SVM的柴油机装配冷试检测方法，其特征在于，通过选取径向基核函数来训练测试所述的支持向量机算法模型，运用交叉验证法寻找最优的惩罚因子C和方差g，即在设定的参数范围内，确定迭代长度，将惩罚因子和方差两两组合，在多组交叉验证中选取精度最高的组合。according to the described SVM-based diesel engine assembly cold test detection method of claim 1, it is characterized in that, train and test described support vector machine algorithm model by selecting radial basis kernel function, use cross-validation method to find optimal penalty factor C and variance g, that is, within the set parameter range, determine the iteration length, combine the penalty factor and variance in pairs, and select the combination with the highest accuracy in the multi-group cross-validation.
  5. 根据权利要求4所述的基于SVM的柴油机装配冷试检测方法，其特征在于，将构建的所述柴油机装配质量检测的样本集划分为训练集和测试集，用支持向量机算法对训练集训练，得到冷试测试质量检测支持向量机算法模型，利用测试集确定支持向量机算法模型对装配质量检测的准确性，从而对柴油机装配质量进行识别。The diesel engine assembly cold test detection method based on SVM according to claim 4, characterized in that, the sample set of the diesel engine assembly quality inspection constructed is divided into a training set and a test set, and the training set is trained with a support vector machine algorithm , to obtain the support vector machine algorithm model of the cold test test quality inspection, and use the test set to determine the accuracy of the assembly quality inspection of the support vector machine algorithm model, so as to identify the assembly quality of the diesel engine.
  6. 基于SVM的柴油机装配冷试检测***，其特征在于，采用权利要求1所述的基于SVM的柴油机装配冷试检测方法，包括：The diesel engine assembly cold test detection system based on SVM is characterized in that, adopts the diesel engine assembly cold test detection method based on SVM according to claim 1, comprising:
    数据处理单元：利用柴油机的进气压力、曲轴转矩和排气压力构建冷试测试数据库，采用大数据分析的方式判断装配特征参数中排气压力数据的分布形态；并根据排气压力数据的分布形态，得到装配排气压力正常、偏小和偏大的阈值，从而构建柴油机装配质量检测的样本集；Data processing unit: use the intake pressure, crankshaft torque and exhaust pressure of the diesel engine to build a cold test test database, and use big data analysis to judge the distribution of exhaust pressure data in the assembly characteristic parameters; and according to the exhaust pressure data According to the distribution shape, the normal, small and large thresholds of assembly exhaust pressure are obtained, so as to construct a sample set for diesel engine assembly quality inspection;
    模型建立单元：确定排气压力正常、偏小和偏大的阈值后，构建建立支持向量机算法模型，训练测试支持向量机算法模型，通过训练测试后的 支持向量机算法模型用于对柴油机装配质量进行识别。Model building unit: After determining the normal, small and large thresholds of the exhaust pressure, construct the support vector machine algorithm model, train and test the support vector machine algorithm model, and use the support vector machine algorithm model after the training test to assemble the diesel engine Quality is identified.
PCT/CN2022/119171 2021-09-29 2022-09-16 Svm-based cold flow test detection method and system during diesel engine assembly WO2023051275A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
GB2318389.0A GB2622708A (en) 2021-09-29 2022-09-16 SVM-based cold flow test detection method and system during diesel engine assembly

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111153792.2A CN113884305B (en) 2021-09-29 2021-09-29 Diesel engine assembly cold test detection method and system based on SVM
CN202111153792.2 2021-09-29

Publications (1)

Publication Number Publication Date
WO2023051275A1 true WO2023051275A1 (en) 2023-04-06

Family

ID=79008371

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/119171 WO2023051275A1 (en) 2021-09-29 2022-09-16 Svm-based cold flow test detection method and system during diesel engine assembly

Country Status (3)

Country Link
CN (1) CN113884305B (en)
GB (1) GB2622708A (en)
WO (1) WO2023051275A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116433111A (en) * 2023-06-15 2023-07-14 潍柴动力股份有限公司 Construction method and quality evaluation method of crankshaft assembly torque quality detection system

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113884305B (en) * 2021-09-29 2022-06-28 山东大学 Diesel engine assembly cold test detection method and system based on SVM

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110093182A1 (en) * 2008-05-08 2011-04-21 Borgwarner Beru Systems Gmbh Estimating engine parameters based on dynamic pressure readings
CN108492399A (en) * 2018-02-11 2018-09-04 山东大学 Bull-dozer fault diagnosis expert system for diesel engine based on big data analysis and method
CN109726230A (en) * 2018-12-04 2019-05-07 重庆大学 A kind of method of big data analysis model prediction engine performance
CN110261122A (en) * 2019-06-20 2019-09-20 大连理工大学 A kind of boat diesel engine fault monitoring method based on piecemeal
CN111351668A (en) * 2020-01-14 2020-06-30 江苏科技大学 Diesel engine fault diagnosis method based on optimized particle swarm algorithm and neural network
CN111562111A (en) * 2020-06-05 2020-08-21 上海交通大学 Engine cold state test fault diagnosis method
CN111832617A (en) * 2020-06-05 2020-10-27 上海交通大学 Engine cold state test fault diagnosis method
CN113884305A (en) * 2021-09-29 2022-01-04 山东大学 Diesel engine assembly cold test detection method and system based on SVM

Family Cites Families (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1519183B1 (en) * 1996-07-19 2012-02-01 Toyota Jidosha Kabushiki Kaisha Method of testing assembled internal combustion engine
JP2000199428A (en) * 1998-10-29 2000-07-18 Hitachi Metals Ltd Evaluation model and evaluation method for exhaust manifold which connects catalytic carrier, and exhaust manifold obtained thereby
JP4776029B2 (en) * 2007-03-28 2011-09-21 Ｕｄトラックス株式会社 Power test oil circulation system for cold test bench
CN201034820Y (en) * 2007-04-10 2008-03-12 浙江大学鸣泉电子科技有限公司 Gasoline vehicle extraction flow analyzer
FR2923546A1 (en) * 2007-11-09 2009-05-15 Renault Sas Exhaust gas's mass flow measuring method for motor vehicle, involves performing gas analysis in pipes and in exhaust line for determining mass flow of exhaust gas so as to correctly adjust combustion parameters of engine
SE534475C2 (en) * 2010-01-18 2011-09-06 Scania Cv Ab Method and apparatus for preventing fuel accumulation in an exhaust system of a motor vehicle
WO2011118095A1 (en) * 2010-03-25 2011-09-29 Ｕｄトラックス株式会社 Engine exhaust purification device and engine exaust purification method
CN102680242B (en) * 2012-06-06 2014-09-17 哈尔滨工程大学 Fault diagnosing method for diesel engine based on swarm intelligence
JP6225934B2 (en) * 2015-02-27 2017-11-08 トヨタ自動車株式会社 Control device for internal combustion engine
CN105319071B (en) * 2015-09-21 2017-11-07 天津大学 Diesel Engine Fuel System Fault Diagnosis method based on least square method supporting vector machine
JP2018115997A (en) * 2017-01-19 2018-07-26 株式会社堀場製作所 Exhaust gas flow rate measurement unit and exhaust gas analysis device
NL2019853B1 (en) * 2017-11-03 2019-05-13 Daf Trucks Nv System and method for detecting malfunctioning turbo-diesel cylinders.
CN108387378B (en) * 2018-01-22 2019-11-15 西安航天动力试验技术研究所 A kind of engine test Propellant Supply low frequency pulsating suppressing method and system
CN110749450A (en) * 2018-07-24 2020-02-04 上海华依科技集团股份有限公司 Air inlet and exhaust plugging testing mechanism and method for engine cold test equipment
CN111175052A (en) * 2018-11-13 2020-05-19 上海华依科技集团股份有限公司 Engine gas distribution system fault testing system for engine cold test
CN109506942B (en) * 2018-12-04 2020-08-04 重庆大学 Method for analyzing correlation between engine cold test detection data and station by big data
CN110197222A (en) * 2019-05-29 2019-09-03 国网河北省电力有限公司石家庄供电分公司 A method of based on multi-category support vector machines transformer fault diagnosis
CN110308005A (en) * 2019-06-12 2019-10-08 上海市环境科学研究院 Fractions of Diesel Engine Exhaust Particulates object generation system and Fractions of Diesel Engine Exhaust Particulates object analogy method
CN111779573B (en) * 2020-06-28 2022-02-11 河南柴油机重工有限责任公司 Diesel engine online fault detection method and device

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110093182A1 (en) * 2008-05-08 2011-04-21 Borgwarner Beru Systems Gmbh Estimating engine parameters based on dynamic pressure readings
CN108492399A (en) * 2018-02-11 2018-09-04 山东大学 Bull-dozer fault diagnosis expert system for diesel engine based on big data analysis and method
CN109726230A (en) * 2018-12-04 2019-05-07 重庆大学 A kind of method of big data analysis model prediction engine performance
CN110261122A (en) * 2019-06-20 2019-09-20 大连理工大学 A kind of boat diesel engine fault monitoring method based on piecemeal
CN111351668A (en) * 2020-01-14 2020-06-30 江苏科技大学 Diesel engine fault diagnosis method based on optimized particle swarm algorithm and neural network
CN111562111A (en) * 2020-06-05 2020-08-21 上海交通大学 Engine cold state test fault diagnosis method
CN111832617A (en) * 2020-06-05 2020-10-27 上海交通大学 Engine cold state test fault diagnosis method
CN113884305A (en) * 2021-09-29 2022-01-04 山东大学 Diesel engine assembly cold test detection method and system based on SVM

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
CHEN, QINHE ET AL.: "Study on Relevance between Engine Bolt Tightening and Cylinder Cover Vibration", MECHANICAL SCIENCE AND TECHNOLOGY, vol. 37, no. 11, 30 November 2018 (2018-11-30), pages 1662 - 1669, XP009544990, ISSN: 1003-8728 *
YANG, JIA ET AL.: "Normal Domain Design of Multi-parameter in Engine Cold Test Based on SVDD", MACHINE DESIGN & RESEARCH, vol. 29, no. 03, 20 June 2013 (2013-06-20), XP009544989, ISSN: 1006-2343 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116433111A (en) * 2023-06-15 2023-07-14 潍柴动力股份有限公司 Construction method and quality evaluation method of crankshaft assembly torque quality detection system
CN116433111B (en) * 2023-06-15 2023-10-20 潍柴动力股份有限公司 Construction method and quality evaluation method of crankshaft assembly torque quality detection system

Also Published As

Publication number Publication date
CN113884305A (en) 2022-01-04
GB2622708A (en) 2024-03-27
CN113884305B (en) 2022-06-28
GB202318389D0 (en) 2024-01-17

Similar Documents

Publication Publication Date Title
WO2023051275A1 (en) Svm-based cold flow test detection method and system during diesel engine assembly
Tong et al. Convolutional neural network for asphalt pavement surface texture analysis
CN110163278B (en) Flame stability monitoring method based on image recognition
CN107328868B (en) A kind of Analysis of Acoustic Emission Signal method of quick identification ceramic coating failure type
TW201407154A (en) Integration of automatic and manual defect classification
CN111860106B (en) Unsupervised bridge crack identification method
Cai et al. A novel improved local binary pattern and its application to the fault diagnosis of diesel engine
WO2019015226A1 (en) Method for rapidly identifying wind speed distribution pattern
CN111797887A (en) Anti-electricity-stealing early warning method and system based on density screening and K-means clustering
CN110555235A (en) Structure local defect detection method based on vector autoregressive model
CN114487129B (en) Flexible material damage identification method based on acoustic emission technology
CN116226103A (en) Method for detecting government data quality based on FPGrow algorithm
CN109886314B (en) Kitchen waste oil detection method and device based on PNN neural network
CN114722641A (en) Lubricating oil state information integrated evaluation method and system for detection laboratory
CN115659323A (en) Intrusion detection method based on information entropy theory and convolution neural network
CN115129503A (en) Equipment fault data cleaning method and system
CN108595860B (en) Bridge construction vertical prestressed reinforcement detecting system based on computer
CN113673551A (en) Method and system for identifying bad data of electric power metering
Hu et al. A lightweight reconstruction network for surface defect inspection
Gu et al. Research on intelligent detection technology of surface defects of nuclear fuel rods based on machine vision
CN111581409A (en) Damage image feature database construction method and system and engine
CN108459948A (en) The determination method of fail data distribution pattern in Reliability evaluation
CN112884167B (en) Multi-index anomaly detection method based on machine learning and application system thereof
Ma et al. Network Intrusion Detection Method Based on FSSAE and GRU
TWI715428B (en) Identification method, identification device and identification system of cell image

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22874662

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 202318389

Country of ref document: GB

Kind code of ref document: A

Free format text: PCT FILING DATE = 20220916

NENP Non-entry into the national phase

Ref country code: DE