Plateau light automobile emission data analysis and research method
Through multi-source data acquisition and machine learning algorithm combined with plateau environmental factors, an emission prediction model suitable for plateau is constructed, which solves the problem of emission prediction deviation and lack of integration of data in plateau environments in the existing technology, and achieves high-precision emission prediction and policy support .
Patent Information
- Application Number
- CN202411198484.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-29
- Publication Date
- 2024-12-27
AI Technical Summary
The existing light vehicle emission data analysis method fails to fully consider factors such as thin air and low air pressure in the plateau environment, resulting in deviations in the prediction results of the emission prediction model in the plateau environment, and the plateau light vehicle emission data lacks data fusion and details. analyze.
Multi-source data acquisition, data cleaning and preprocessing, combined with meteorological data, statistical analysis and machine learning algorithms are used to study the impact of plateau environmental factors on emission characteristics, and typical emission patterns are identified through cluster analysis and correlation rules mining to build suitable for plateaus. The environmental emission prediction model introduces plateau environmental factors as input variables to improve prediction accuracy.
It has achieved the accuracy of emission prediction models in plateau environments, supported the formulation and evaluation of environmental protection policies, improved data utilization efficiency, helped manufacturers identify and improve models with poor emission performance, and improved product environmental protection performance and market competitiveness.
Smart Images
Figure CN119202579A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of automobile emission analysis, and in particular to an emission data analysis and research method for plateau light automobiles. Background Art
[0002] 1) The particularity of the plateau environment
[0003] The plateau area has unique geographical environmental characteristics, such as high altitude, thin air, low oxygen content, and large temperature changes. These environmental factors will have a significant impact on the engine performance, fuel combustion efficiency and emission characteristics of light vehicles.
[0004] The transportation infrastructure construction in the plateau area is constantly improving, the number of light vehicles is gradually increasing, and their emission problems are becoming increasingly prominent.
[0005] (2) Urgency of the emission problem
[0006] Light-duty vehicle emissions are one of the important sources of urban air pollution. Due to the particularity of the geographical environment, the atmospheric environment carrying capacity of plateau areas is relatively weak. Once polluted, it is difficult to recover. Therefore, it is of great significance to strengthen the analysis and control of plateau light-duty vehicle emissions data.
[0007] (3) Necessity of the study
[0008] Most of the existing emission data analysis methods and standards are based on the environmental conditions in plain areas, and their applicability to plateau areas needs further verification. By studying the characteristics of plateau light vehicle emission data, we can reveal the influence of plateau environment on vehicle emissions and provide technical support for the formulation of more scientific and reasonable emission control strategies.
[0009] Defects and shortcomings of existing technology
[0010] (1) Lack of specific methods for plateau environments
[0011] Existing light vehicle emission data analysis and research methods are mostly based on plain environments, and fail to fully consider the impact of factors such as thin air and low air pressure in plateau environments on engine emission characteristics, resulting in deviations in the prediction results of emission prediction models in plateau environments.
[0012] (2) Lack of data fusion for plateau light vehicle emission data
[0013] The emission data of light vehicles in the plateau are multi-source and heterogeneous, including structured data, semi-structured data, unstructured data, etc. They come from laboratory analysis, RDE testing, on-board diagnostic system (OBD) data, etc. These data are distributed in different regions and different test equipment, and are stored on different media. There is no unified aggregation and lack of data fusion.
[0014] (3) Insufficient data analysis
[0015] Existing data analysis methods often focus on the total amount of emissions, but lack in-depth analysis of detailed information such as emission composition, particle size distribution, post-processing devices, oil products, engines, manufacturers, etc. In plateau environments, due to factors such as incomplete combustion and reduced efficiency of post-processing systems, emission composition and particle size distribution may change significantly, and existing data analysis methods are difficult to accurately capture these changes. Summary of the invention
[0016] The present invention aims to solve the above problems of the prior art. A method for analyzing and studying the emission data of light vehicles in plateaus is proposed. The technical solution of the present invention is as follows:
[0017] A plateau light vehicle emission data analysis and research method, comprising the following steps:
[0018] Data collection and preprocessing steps: Collect emission data of light vehicles in plateau environment, clean the collected data, combine meteorological data of plateau area with emission data to form a comprehensive data set; Emission characteristic analysis steps: Use statistical analysis, machine learning algorithms and other methods to study the impact mechanism of plateau environmental factors on emission characteristics and identify key influencing factors; Based on the comprehensive data set, use cluster analysis and association rule mining technology to identify the typical emission patterns of light vehicles in plateau environment;
[0019] Establish a plateau emission model: Based on the results of emission characteristic analysis, improve the traditional emission prediction model, introduce plateau environmental factors as input variables, and build an emission prediction model suitable for plateau environment; use measured data in plateau environment to train and verify the model, adjust model parameters, evaluate the adaptability of the model in different plateau areas and different vehicle models, and ensure the generalization ability of the model.
[0020] Furthermore, the data collection and preprocessing steps specifically include:
[0021] Multi-source data collection steps: Use multiple channels including laboratory analysis, RDE testing, and OBD systems to collect emission data of light vehicles in plateau environments, including structured data, semi-structured data, and unstructured data;
[0022] Data cleaning and standardization steps: Clean the collected data, remove noise and outliers, and standardize the format;
[0023] Environmental parameter integration step: Combine the meteorological data including air pressure, temperature, humidity, and oxygen content of the plateau area with the emission data to form a comprehensive data set.
[0024] Furthermore, a machine learning algorithm is used to study the impact mechanism of plateau environmental factors on emission characteristics. The steps of the machine learning algorithm include clarifying business indicators and model prediction targets, data collection, data preprocessing, feature engineering, model selection, model training, model evaluation, model fusion and online deployment.
[0025] Specifically include:
[0026] Clarify business indicators and model prediction targets: clarify the impact mechanism of plateau environmental factors on emission characteristics through the model and the expected output results;
[0027] Data collection: Collect relevant data sets based on the problem definition;
[0028] Data preprocessing: clean, remove duplicates, and process missing values for the collected data;
[0029] Feature engineering: extracting, transforming, and selecting relevant features to make the original data more suitable for model training and prediction;
[0030] Model selection: Select appropriate machine learning models, including linear regression, decision tree, and neural network, based on the nature of the problem and the characteristics of the data;
[0031] Model training: Use the training dataset to train the selected model and adjust the model parameters to achieve the best performance;
[0032] Model evaluation: Use the test data set to evaluate the performance of the model. The evaluation indicators include accuracy, precision, recall, F1-score, and root mean square error (RMSE).
[0033] Model fusion: Fusion of the results of multiple models to obtain more accurate predictions.
[0034] Online deployment: Deploy the trained model to the production environment to process new data in real time or batches and give prediction results.
[0035] Furthermore, the association rule mining adopts the improved Apriori algorithm to identify the typical emission pattern of light vehicles in plateau environment. The Apriori algorithm is one of the most influential algorithms for mining frequent item sets of Boolean association rules. Its core is a recursive algorithm based on the two-stage frequent set idea; all item sets with support greater than the minimum support are called frequent item sets, or frequent sets for short;
[0036] The basic idea of the algorithm is: first find all frequent sets, the frequency of occurrence of these item sets is at least the same as the predefined minimum support. Then generate strong association rules from the frequent sets, which must meet the minimum support and minimum credibility. Then use the frequent sets found in step 1 to generate the desired rules, and generate all rules that only contain items in the set, where each rule has only one item on the right. The definition of the middle rule is adopted here. Once these rules are generated, only those rules that are greater than the minimum credibility given by the user are retained. In order to generate all frequent sets, a recursive method is used; the possibility of generating a large number of candidate sets and the possibility of repeatedly scanning the database are the two major disadvantages of the Apriori algorithm; the Apriori algorithm is improved by a partition-based method. The partition-based algorithm first divides the database into several non-intersecting blocks logically, considers a block separately each time and generates all frequent sets for it, then merges the generated frequent sets to generate all possible frequent sets, and finally calculates the support of these item sets.
[0037] Furthermore, the emission prediction model suitable for plateau environment is an XGBoost-SVR emission prediction model that adds plateau environmental factors as input variables, specifically including: adding singular spectrum analysis to process the original time series and eliminating outliers in the time series; using the XGBoost model to make a preliminary prediction of the instantaneous emissions of CO and PN of plateau light vehicles, and using the SVR model to perform residual correction to obtain the final prediction results; comparing the prediction results with the actual values measured using PEMS equipment in actual road emission tests.
[0038] The advantages and beneficial effects of the present invention are as follows:
[0039] 1. Promote environmental policy formulation and evaluation
[0040] Based on the optimized emission prediction model, an emission inventory of light vehicles in the plateau area can be compiled to provide data support for environmental protection and emission control. Using the emission inventory data, the effects of different environmental protection policies can be evaluated and a scientific basis can be provided for policy making.
[0041] 2. Data Fusion
[0042] Through a unified data management platform and data fusion technology, the integration and utilization of multi-source heterogeneous data are achieved, improving the efficiency and value of data utilization.
[0043] 3. Improve emission prediction accuracy
[0044] The present invention fully considers the influence of special factors such as thin air, low air pressure and low oxygen content in plateau environment on engine emission characteristics. By introducing these environmental factors as input variables, the prediction accuracy of the emission prediction model in plateau environment is significantly improved.
[0045] 4. Support manufacturers’ product improvement and R&D
[0046] Vehicle manufacturers can use the emission data analysis results provided by the present invention to identify vehicle models or configurations with poor emission performance, thereby making targeted product improvements, which helps to improve the environmental performance and market competitiveness of products. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 It is a flow chart of the processing flow and analysis method of emission data of the present invention. DETAILED DESCRIPTION
[0048] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.
[0049] The technical solution of the present invention to solve the above technical problems is:
[0050] Preferably, Figure 1 As shown, a plateau light vehicle emission data analysis and research method comprises the following steps:
[0051] Data collection and preprocessing steps: Collect emission data of light vehicles in plateau environment, clean the collected data, combine meteorological data of plateau area with emission data to form a comprehensive data set; Emission characteristic analysis steps: Use statistical analysis, machine learning algorithms and other methods to study the impact mechanism of plateau environmental factors on emission characteristics and identify key influencing factors; Based on the comprehensive data set, use cluster analysis and association rule mining technology to identify the typical emission patterns of light vehicles in plateau environment;
[0052] Establish a plateau emission model: Based on the results of emission characteristic analysis, improve the traditional emission prediction model, introduce plateau environmental factors as input variables, and build an emission prediction model suitable for plateau environment; use measured data in plateau environment to train and verify the model, adjust model parameters, evaluate the adaptability of the model in different plateau areas and different vehicle models, and ensure the generalization ability of the model.
[0053] Preferably, the data collection and preprocessing steps specifically include:
[0054] Multi-source data collection steps: Use multiple channels including laboratory analysis, RDE testing, and OBD systems to collect emission data of light vehicles in plateau environments, including structured data, semi-structured data, and unstructured data;
[0055] Data cleaning and standardization steps: Clean the collected data, remove noise and outliers, and standardize the format;
[0056] Environmental parameter integration step: Combine the meteorological data including air pressure, temperature, humidity, and oxygen content of the plateau area with the emission data to form a comprehensive data set.
[0057] Preferably, a machine learning algorithm is used to study the impact mechanism of plateau environmental factors on emission characteristics. The steps of the machine learning algorithm include clarifying business indicators and model prediction targets, data collection, data preprocessing, feature engineering, model selection, model training, model evaluation, model fusion and online deployment; specifically including:
[0058] Clarify business indicators and model prediction targets: clarify the impact mechanism of plateau environmental factors on emission characteristics through the model and the expected output results;
[0059] Data collection: Collect relevant data sets based on the problem definition;
[0060] Data preprocessing: clean, remove duplicates, and process missing values for the collected data;
[0061] Feature engineering: extracting, transforming, and selecting relevant features to make the original data more suitable for model training and prediction;
[0062] Model selection: Select appropriate machine learning models, including linear regression, decision tree, and neural network, based on the nature of the problem and the characteristics of the data;
[0063] Model training: Use the training dataset to train the selected model and adjust the model parameters to achieve the best performance;
[0064] Model evaluation: Use the test data set to evaluate the performance of the model. The evaluation indicators include accuracy, precision, recall, F1-score, and root mean square error (RMSE).
[0065] Model fusion: Fusion of the results of multiple models to obtain more accurate predictions.
[0066] Online deployment: Deploy the trained model to the production environment to process new data in real time or batches and give prediction results.
[0067] Preferably, the association rule mining adopts the improved Apriori algorithm to identify the typical emission pattern of light vehicles in plateau environment. The Apriori algorithm is one of the most influential algorithms for mining frequent item sets of Boolean association rules. Its core is a recursive algorithm based on the two-stage frequent set idea; all item sets with support greater than the minimum support are called frequent item sets, referred to as frequent sets;
[0068] The basic idea of the algorithm is: first find all frequent sets, the frequency of occurrence of these item sets is at least the same as the predefined minimum support. Then generate strong association rules from the frequent sets, which must meet the minimum support and minimum credibility. Then use the frequent sets found in step 1 to generate the desired rules, and generate all rules that only contain items in the set, where each rule has only one item on the right. The definition of the middle rule is adopted here. Once these rules are generated, only those rules that are greater than the minimum credibility given by the user are retained. In order to generate all frequent sets, a recursive method is used; the possibility of generating a large number of candidate sets and the possibility of repeatedly scanning the database are the two major disadvantages of the Apriori algorithm; the Apriori algorithm is improved by a partition-based method. The partition-based algorithm first divides the database into several non-intersecting blocks logically, considers a block separately each time and generates all frequent sets for it, then merges the generated frequent sets to generate all possible frequent sets, and finally calculates the support of these item sets.
[0069] Preferably, the emission prediction model suitable for plateau environment is an XGBoost-SVR emission prediction model that adds plateau environmental factors as input variables, specifically including: adding singular spectrum analysis to process the original time series and eliminating outliers in the time series; using the XGBoost model to make a preliminary prediction of the instantaneous emissions of CO and PN of plateau light vehicles, and using the SVR model to perform residual correction to obtain the final prediction results; comparing the prediction results with the actual values measured using PEMS equipment in actual road emission tests.
[0070] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the plateau light vehicle emission data analysis and research method as described in any one of the items is implemented.
[0071] A non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any of the above-mentioned plateau light vehicle emission data analysis and research methods.
[0072] A plateau light vehicle emission data analysis and research system, comprising:
[0073] Data collection and preprocessing module: used to collect emission data of light vehicles in plateau environments, clean the collected data, combine the meteorological data of the plateau area with the emission data, and form a comprehensive data set;
[0074] Emission characteristic analysis module: used to study the impact mechanism of plateau environmental factors on emission characteristics and identify key influencing factors using methods including statistical analysis and machine learning algorithms; based on a comprehensive data set, cluster analysis and association rule mining technology are used to identify the typical emission patterns of light vehicles in plateau environments; Plateau emission model: used to improve the traditional emission prediction model based on the results of emission characteristic analysis, introduce plateau environmental factors as input variables, and construct an emission prediction model suitable for plateau environments; use measured data in plateau environments to train and verify the model, adjust model parameters, and evaluate the adaptability of the model to different plateau regions and different models to ensure the generalization ability of the model.
[0075] The specific embodiments are:
[0076] 1.1 Data collection and preprocessing
[0077] Multi-source data collection: Utilize multiple channels such as laboratory analysis, RDE testing, and OBD systems to collect emission data of light vehicles in plateau environments, including structured data (such as emission concentration), semi-structured data (such as vehicle operating status), and unstructured data (such as test videos and images).
[0078] Data cleaning and standardization: Clean the collected data, remove noise and outliers, and standardize the format to ensure data consistency and comparability.
[0079] Integration of environmental parameters: Meteorological data (such as air pressure, temperature, humidity, oxygen content, etc.) in plateau areas are combined with emission data to form a comprehensive data set.
[0080] 1.2 Emission characteristics analysis
[0081] Statistical analysis, machine learning and other methods are used to study the impact mechanism of plateau environmental factors on emission characteristics and identify key influencing factors. Based on the comprehensive data set, cluster analysis, association rule mining and other technologies are used to identify the typical emission patterns of light vehicles in plateau environments.
[0082] 1.3 Establishing a plateau emission model
[0083] According to the results of emission characteristics analysis, the traditional emission prediction model is improved, and plateau environmental factors are introduced as input variables to build an emission prediction model suitable for plateau environment. The model is trained and verified using measured data in plateau environment, and the model parameters are adjusted to improve the prediction accuracy. The adaptability of the model in different plateau areas and different vehicle models is evaluated to ensure the generalization ability of the model.
[0084] Application of results
[0085] (1) Policy formulation and evaluation
[0086] Emission inventory data is used to evaluate the effectiveness of different environmental protection policies, such as environmental protection standards compliance assessment, to provide a scientific basis for policy making.
[0087] (2) Emissions Inventory Preparation
[0088] Based on the optimized emission prediction model, an emission inventory of light vehicles in plateau areas is compiled to provide data support for ecological environmental protection and emission control.
[0089] (3) Application by vehicle manufacturers
[0090] Vehicle manufacturers can use the emission data analysis results and emission prediction models provided by the present invention to understand the emission performance of their products in plateau environments. By comparing the emission data of different models and engine configurations, manufacturers can identify models or configurations with poor emission performance, so as to make targeted product improvements, including optimizing the engine combustion system, adjusting the fuel injection strategy, improving the exhaust after-treatment system, etc., to improve the emission performance of products in plateau environments, so as to enhance the brand image and consumer trust.
[0091] Key technical points
[0092] (1) Multi-source heterogeneous data fusion technology
[0093] Integrate and analyze plateau emission data from different data sources, formats and fields, clean and pre-process the data by removing redundant data, filling missing values, and resolving data inconsistencies, to form visible, manageable and controllable data, providing a data foundation for data mining and model building.
[0094] (2) Construction of emission prediction model adapted to plateau environment
[0095] Based on the traditional emission prediction model, plateau environmental factors are introduced as input variables, and an emission prediction model suitable for plateau environment is constructed through algorithm optimization and model training. The model is verified using measured data in plateau environment, and model parameters are adjusted to ensure the prediction accuracy of the model in plateau environment.
[0096] (3) Emission characteristics analysis and emission pattern recognition
[0097] Using statistical analysis, machine learning and other methods, we conduct in-depth research on the impact mechanism of plateau environmental factors on emission characteristics, helping manufacturers understand emission performance in plateau environments. Based on comprehensive data sets, we use cluster analysis, association rule mining and other technologies to identify typical emission patterns of light vehicles in plateau environments, providing a scientific basis for model optimization.
[0098] 2.2 Points to be protected
[0099] (1) Protection of technical methods
[0100] Protect the specific algorithms and steps used for multi-source heterogeneous data fusion to prevent unauthorized use or copying by others. Protect the construction method, optimization strategy and key parameter settings of the emission prediction model to ensure technical barriers and competitive advantages.
[0101] (2) Data resource protection
[0102] Protect the original data sources and comprehensive data sets used for model building and verification to prevent data leakage and illegal use. Protect the specific steps and parameter settings of data cleaning, standardization, fusion and other processing processes to ensure the accuracy and reliability of data processing results.
[0103] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
[0104] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0105] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0106] The above embodiments should be understood to be only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the contents of the present invention, technicians can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.
Claims
1. A plateau light vehicle emission data analysis and research method, characterized in that: The following steps are involved: Data collection and preprocessing steps: Collect emission data of light vehicles in plateau environment, clean the collected data, combine the meteorological data of plateau area with emission data to form a comprehensive data set; Emission characteristics analysis steps: Use statistical analysis, machine learning algorithms and other methods to study the impact mechanism of plateau environmental factors on emission characteristics and identify key influencing factors; Based on the comprehensive data set, use cluster analysis and association rule mining technology to identify the typical emission patterns of light vehicles in plateau environments; Establish a plateau emission model: Based on the results of emission characteristic analysis, improve the traditional emission prediction model, introduce plateau environmental factors as input variables, and build an emission prediction model suitable for plateau environment; use measured data in plateau environment to train and verify the model, adjust model parameters, evaluate the adaptability of the model in different plateau areas and different vehicle models, and ensure the generalization ability of the model.
2. The plateau light vehicle emission data analysis and research method according to claim 1 is characterized in that: The data collection and preprocessing steps specifically include: Multi-source data collection steps: Use multiple channels including laboratory analysis, RDE testing, and OBD systems to collect emission data of light vehicles in plateau environments, including structured data, semi-structured data, and unstructured data; Data cleaning and standardization steps: Clean the collected data, remove noise and outliers, and standardize the format; Environmental parameter integration step: Combine the meteorological data including air pressure, temperature, humidity, and oxygen content of the plateau area with the emission data to form a comprehensive data set.
3. The plateau light vehicle emission data analysis and research method according to claim 1 is characterized in that: A machine learning algorithm is used to study the impact mechanism of plateau environmental factors on emission characteristics. The steps of the machine learning algorithm include clarifying business indicators and model prediction targets, data collection, data preprocessing, feature engineering, model selection, model training, model evaluation, model fusion and online deployment; specifically, they include: Clarify business indicators and model prediction targets: clarify the impact mechanism of plateau environmental factors on emission characteristics through the model and the expected output results; Data collection: Collect relevant data sets based on the problem definition; Data preprocessing: clean, remove duplicates, and process missing values for the collected data; Feature engineering: extracting, transforming, and selecting relevant features to make the original data more suitable for model training and prediction; Model selection: Select appropriate machine learning models, including linear regression, decision tree, and neural network, based on the nature of the problem and the characteristics of the data; Model training: Use the training dataset to train the selected model and adjust the model parameters to achieve the best performance; Model evaluation: Use the test data set to evaluate the performance of the model. The evaluation indicators include accuracy, precision, recall, F1-score, and root mean square error (RMSE). Model fusion: Fusion of the results of multiple models to obtain more accurate predictions. Online deployment: Deploy the trained model to the production environment to process new data in real time or batches and give prediction results.
4. The plateau light vehicle emission data analysis and research method according to claim 1 is characterized in that: The association rule mining adopts the improved Apriori algorithm to identify the typical emission pattern of light vehicles in plateau environment. The Apriori algorithm is one of the most influential algorithms for mining frequent item sets of Boolean association rules. Its core is a recursive algorithm based on the two-stage frequent set idea; all item sets with support greater than the minimum support are called frequent item sets, referred to as frequent sets; The basic idea of the algorithm is: first find all frequent sets, the frequency of occurrence of these item sets is at least the same as the predefined minimum support. Then generate strong association rules from the frequent sets, which must meet the minimum support and minimum credibility. Then use the frequent sets found in step 1 to generate the desired rules, and generate all rules that only contain items in the set, where each rule has only one item on the right. The definition of the middle rule is adopted here. Once these rules are generated, only those rules that are greater than the minimum credibility given by the user are retained. In order to generate all frequent sets, a recursive method is used; the possibility of generating a large number of candidate sets and the possibility of repeatedly scanning the database are the two major disadvantages of the Apriori algorithm; the Apriori algorithm is improved by a partition-based method. The partition-based algorithm first divides the database into several non-intersecting blocks logically, considers a block separately each time and generates all frequent sets for it, then merges the generated frequent sets to generate all possible frequent sets, and finally calculates the support of these item sets.
5. The plateau light vehicle emission data analysis and research method according to claim 1 is characterized in that: The emission prediction model suitable for plateau environment is an XGBoost-SVR emission prediction model that adds plateau environmental factors as input variables, specifically including: adding singular spectrum analysis to process the original time series and eliminating outliers in the time series; using the XGBoost model to make a preliminary prediction of the instantaneous emissions of CO and PN of plateau light vehicles, and using the SVR model to perform residual correction to obtain the final prediction result; comparing the prediction result with the actual value measured by the PEMS device in the actual road emission test.