Optimization Design Method of Car Fuel Consumption Correction Model Based on PEMS
By using a PEMS-based car fuel consumption correction model, which comprehensively considers factors such as vehicle speed, acceleration, altitude, and specific power of motor vehicles, the problem of existing technologies being unable to comprehensively assess the influence of multiple factors is solved, and more accurate fuel consumption assessment is achieved.
Patent Information
- Application Number
- CN202310931713.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-07-27
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-07-27
AI Technical Summary
Existing methods for assessing vehicle fuel consumption fail to comprehensively consider multiple influencing factors, leading to inaccurate assessments.
A PEMS-based fuel consumption correction model for automobiles was adopted. The model was designed by comprehensively considering factors such as vehicle speed, acceleration, altitude and specific power of motor vehicles through multiple linear regression analysis, and then corrected by carbon balance fuel consumption calculation method.
It improves the accuracy and persuasiveness of fuel consumption assessment, reflects the actual operating conditions of light-duty gasoline vehicles, and the determination of correction coefficients makes the model more rigorous.
Smart Images

Figure CN117150725B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of automotive fuel consumption assessment technology, and in particular to an optimization design method for a car fuel consumption correction model based on PEMS. Background Technology
[0002] The rapid development of the automotive industry has led to a continuous increase in car ownership. This phenomenon has also brought about a serious energy problem, namely the rapid increase in vehicle exhaust emissions, which has a significant impact on air quality and health. Generally speaking, the factors affecting vehicle fuel consumption models can be divided into three categories: vehicle technology factors, environmental conditions factors, and operation management factors. Vehicle technology factors include vehicle weight, engine displacement, fuel type, transmission type, and vehicle power-to-weight ratio; environmental conditions factors include road gradient, wind speed, temperature, altitude, and road surface conditions; operation management factors include road grade, traffic volume, intersection type and control method, as well as vehicle speed, acceleration, and queue length. However, current estimation methods only address single influencing factors and cannot assess fuel consumption based on multiple influencing factors. Summary of the Invention
[0003] In view of this, the purpose of this invention is to provide an optimization design method for a car fuel consumption correction model based on PEMS, which comprehensively considers the influence of factors such as vehicle speed, acceleration, altitude and specific power of motor vehicles on car fuel consumption. Most other studies only consider the influence of one factor on fuel consumption, but in practice, comprehensive consideration is more convincing.
[0004] To achieve the above objectives, the present invention adopts the following technical solution: a PEMS-based optimization design method for a car fuel consumption correction model, comprising the following steps:
[0005] Step 1: Propose a fuel consumption correction model;
[0006] Step 2: Validate the fuel consumption model;
[0007] Step 3: Determine the range of coefficients for the modified model.
[0008] In a preferred embodiment, step 1 specifically involves: calculating the basic fuel consumption of the vehicle during driving and the average fuel consumption at various average speeds; based on this, analyzing the relative influence of speed, acceleration, altitude, and VSP on fuel consumption; and designing a fuel consumption correction model by comprehensively considering the above influencing factors.
[0009] Suggested multiple linear regression equation:
[0010]
[0011] Among them: Q iLet V be the dependent variable, where i = 1, 2, ..., 20, representing vehicle fuel consumption; i Let A be the independent variable, where i = 1, 2, ..., 20, representing the average speed of the vehicle; i Let i be the independent variable, where i = 1, 2, ..., 20, representing the average vehicle acceleration; VSP i Let H be the independent variable, where i = 1, 2, ..., 20, representing the specific power of the vehicle; i , where i = 1, 2, ..., 20, represents the altitude of the vehicle; k1 is the intercept, and k2, k3, k4, k5 are the regression coefficients; after calculating the regression model using the multiple linear regression analysis prediction method, the significance of the regression equation should be tested.
[0012] Linear regression analysis revealed a correlation coefficient R = 0.747 between vehicle average speed and fuel consumption, indicating a linear correlation between vehicle average speed and fuel consumption, and a good fit of the linear fuel consumption model. Based on this, assuming the data conforms to the conditions of a quaternary linear regression equation, the fuel consumption model equation is derived as follows:
[0013] Q i =15.025-0.11V i -0.294A i -0.051H i +0.026VSP i
[0014] In the formula: the correlation coefficient R of the linear equation is 0.775, indicating that the linear model has a good fit.
[0015] In a preferred embodiment, step 2 specifically involves: the null hypothesis is k = 0, meaning the regression coefficient is equal to 0; when the regression coefficient is zero, there is no relationship between the two variables; the F-statistic is used as the test statistic, and its formula is as follows:
[0016]
[0017] In the formula: SSR is the regression sum of squares; SSE is the residual sum of squares; p is the number of independent variables; n is the sample size;
[0018] The F-statistic follows an F-distribution with (4, n-5) degrees of freedom. SPSS software was selected for the significance test of the model. SPSS automatically calculates the observed value F of the test statistic. If F > F(4, 15), the relationship between the independent variable and vehicle fuel consumption is significant. Otherwise, the relationship between the independent variable and vehicle fuel consumption is considered to be insignificant.
[0019] From the PEMS data, we know that p = 4 and n = 20. SPSS analysis shows that SSR is the regression sum of squares and SSE is the residual sum of squares, with an calculated F-value of 5.63. Taking a significance level of α = 0.01, and consulting the F-distribution critical value table, we find that F(4, 15) = 4.89. Since F = 5.63 > F(4, 15) = 4.89, the regression line obtained from the experimental data is highly significant. The significance test results of this fuel consumption correction model using SPSS statistical analysis software show that the model has good significance and meets the requirements for significance testing.
[0020] In a preferred embodiment, step 3 specifically involves: calculating the speed correction coefficient H for fuel consumption in each speed range based on the actual fuel consumption of the experimental vehicle and the model result fuel consumption, according to the following formula. vi ;
[0021]
[0022] As speed increases, the speed correction factor also increases slightly.
[0023] The acceleration correction coefficient H for vehicle fuel consumption in each acceleration range is calculated using the following formula. Ai ;
[0024]
[0025] As acceleration increases, the velocity correction coefficient also increases slightly.
[0026] The fuel consumption correction factor H for each altitude range is calculated using the following formula. Hi ;
[0027]
[0028] As altitude increases, the altitude correction factor also increases slightly.
[0029] The fuel consumption correction factor for each altitude range is calculated using the following formula.
[0030]
[0031] As VSP decreases, the VSP correction factor also increases slightly.
[0032] Compared with the prior art, the present invention has the following beneficial effects:
[0033] (1) The basic fuel consumption model is designed by carbon balance fuel consumption calculation method. The basic fuel consumption model is corrected by comprehensively considering speed, acceleration, altitude and VSP through multiple linear regression model, so as to obtain a more accurate fuel consumption model and better reflect the actual operating conditions of light gasoline vehicles.
[0034] (2) The effect of vehicle instantaneous speed on fuel consumption is quite significant. In the range of 0-60 km / h, the instantaneous fuel consumption gradually decreases with increasing speed, reaching a minimum value around 60 km / h. Based on common knowledge of motor vehicle fuel consumption, after 60 km / h, the instantaneous fuel consumption gradually increases with increasing speed. In the speed range of 25-45 km / h, the instantaneous fuel consumption is stable in the range of 5-12.5 L / 100km.
[0035] (3) By dividing the speed, acceleration, altitude and VSP into intervals, the correction coefficients of each influencing factor in different intervals are obtained. The determination of the correction coefficients makes the design of the vehicle fuel consumption model more rigorous and convincing. Attached Figure Description
[0036] Figure 1 This is a structural diagram of the PEMS system according to a preferred embodiment of the present invention;
[0037] Figure 2 This is a technical roadmap of a preferred embodiment of the present invention;
[0038] Figure 3 This is a preferred embodiment of the present invention showing the instantaneous mass graphs of CO2 and HC.
[0039] Figure 4 This is a graph showing the relationship between instantaneous vehicle speed and travel time according to a preferred embodiment of the present invention;
[0040] Figure 5 This is a graph showing the relationship between instantaneous vehicle speed and instantaneous vehicle fuel consumption according to a preferred embodiment of the present invention.
[0041] Figure 6 This is a graph showing the relationship between instantaneous vehicle acceleration and travel time according to a preferred embodiment of the present invention.
[0042] Figure 7 This is a graph showing the relationship between instantaneous vehicle acceleration and instantaneous vehicle fuel consumption according to a preferred embodiment of the present invention.
[0043] Figure 8 This is a second-by-second VSP distribution diagram of a preferred embodiment of the present invention;
[0044] Figure 9 This is a graph showing the relationship between VSP (Variable Savings Per Second) and instantaneous fuel consumption of a vehicle in a preferred embodiment of the present invention.
[0045] Figure 10 This is a diagram showing the elevation change during the driving process of a preferred embodiment of the present invention.
[0046] Figure 11 This is a graph showing the relationship between altitude and instantaneous fuel consumption of a vehicle, representing a preferred embodiment of the present invention. Detailed Implementation
[0047] The present invention will be further described below with reference to the accompanying drawings and embodiments.
[0048] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of this application. Unless otherwise specified, 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 application pertains.
[0049] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations according to this application; as used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise; furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.
[0050] 1 Overall Scheme Design
[0051] 1.1 Requirements Analysis
[0052] 1.1.1 Experimental Requirements
[0053] This experiment requires the calculation of vehicle speed, acceleration, altitude, and VSP data, and based on this, a vehicle fuel consumption correction model will be designed. PEMS can calculate vehicle speed data per second under various traffic conditions on the road, including speed and altitude, which has significant advantages compared to other testing methods. The experimental design is as follows: Figure 1 As shown:
[0054] The main modules and functions of the PEMS system are shown in Table 1-1, where the fuel economy meter and THC analyzer are both gas analyzers.
[0055] Table 1-1 Main Modules of PEMS System
[0056]
[0057] 1.1.2 Model Requirements
[0058] The fuel consumption correction model is designed based on factors affecting vehicle fuel consumption. These factors include vehicle structure, distance traveled, environmental conditions, vehicle condition, as well as driving speed and acceleration. These factors can be categorized as random factors, natural factors, and human factors. To comprehensively establish the correction model, this fuel consumption model only considers quantifiable factors such as vehicle speed, altitude, and vehicle power-to-weight ratio. Factors such as vehicle condition and driving skill level, although they have a significant impact on fuel consumption, are considered to be at a normal level and are not considered.
[0059] Using carbon balance fuel consumption calculation to measure the fuel consumption of cars has become a major method for analyzing fuel economy both domestically and internationally. This design uses carbon balance fuel consumption calculation as the basic fuel consumption model. The following three points need to be noted during analysis and research:
[0060] (1) The carbon-containing compounds that the exhaust gas analyzer of the PEMS equipment can detect are CO2, CO and HC;
[0061] (2) The total carbon content remained unchanged before and after the test;
[0062] (3) The CO2, CO and HC in the exhaust gas come only from the fuel.
[0063] Based on the basic fuel consumption model, the relationship between speed, acceleration, altitude, and VSP and instantaneous fuel consumption of the vehicle is analyzed. A fuel consumption correction model is established using a multiple linear regression equation, and correction coefficients for relevant parameters are determined. A PEMS-based vehicle fuel consumption correction model is designed.
[0064] 1.2 Scheme Design
[0065] 1.2.1 PEMS Testing System
[0066] The PEMS device needs to be tested under normal driving conditions, and the test time is usually 15 to 30 minutes. During the test, the PEMS device records and saves the vehicle's exhaust emission data. The six steps for installing and testing the PEMS device are shown in Table 1-2 below:
[0067] Table 1-2 PEMS Equipment Installation and Testing
[0068]
[0069]
[0070] 1.2.2 Selection of Experimental Conditions
[0071] To accurately reflect the vehicle's operating conditions, eliminate the influence of driver behavior on fuel consumption, and make the test results more rigorous, this experiment selected a driver with 15 years of driving experience. Through research and analysis, this design chose a light vehicle model with a relatively high ownership rate to make the experiment more representative. The parameter information of the experimental vehicle is shown in Tables 1-3 below.
[0072] Table 1-3 Basic parameters of the test vehicle
[0073]
[0074] Based on the meteorological data, the environmental conditions of the actual road in the experiment are shown in Table 1-4. As can be seen from the table, the experimental environment is in a relatively stable state, and the influence of environmental factors can be ignored in the subsequent analysis.
[0075] Table 1-4 Actual Road Test Environment
[0076]
[0077] The technical roadmap for this design is as follows: Figure 2 As shown.
[0078] 2. Analysis of Factors Affecting Passenger Fuel Consumption Based on PEMS
[0079] 2.1 Characteristics of Real-World Fuel Consumption of Private Cars in Urban Areas
[0080] 2.1.1 Experimental Data Acquisition
[0081] The data selected in this design are the data measured by relevant experiments based on PEMS under actual working conditions of light gasoline vehicles that meet the China V emission standard. The PEMS equipment used in this experiment is the SEMTECH-ECOSTAR vehicle exhaust gas testing equipment from SENSORS, USA.
[0082] This design primarily uses raw experimental data such as speed, time, altitude, and instantaneous CO2 mass as the basis for calculations to study the relationship between speed, acceleration, VSP, and altitude and vehicle fuel consumption. Partial data on time, speed, altitude, and instantaneous CO2 mass collected by the PEMS device during the experiment are shown in Table 2-1.
[0083] Table 2-1 Examples of PEMS Raw Data Acquisition
[0084]
[0085] 2.1.2 Data Analysis and Processing
[0086] The magnitude of a car's acceleration during movement is equal to the change in its velocity per unit time. The unit of acceleration 'a' is meters per second squared, and its symbol is m / s². 2 Its formula (2-1) is as follows:
[0087]
[0088] In the formula:
[0089] v t+1 The instantaneous speed of the vehicle at second t+1 is expressed in m / s.
[0090] v t Let be the instantaneous speed of the vehicle at second t, in m / s;
[0091] Δt represents the unit of time during the vehicle's movement, measured in seconds.
[0092] The concept of VSP was first proposed by Jimenez-Palacios of MIT in his doctoral dissertation. It is defined as the power output of a motor vehicle engine per ton of displacement, with the unit being kW / t.
[0093] Currently, there are relatively mature VSP calculation formulas for motor vehicles. These formulas are based on the instantaneous speed and instantaneous acceleration of the vehicle, and the VSP value can be obtained by substituting these values into the formula. For example, formula (2-2) is the instantaneous VSP (unit: kW / t) calculation formula proposed by Jimenez-Palacios:
[0094] VSP=v·(1.1a+0.132)+0.000302·v 3 (2-2)
[0095] In the formula:
[0096] v is the instantaneous speed of the vehicle, measured in m / s;
[0097] 'a' is the instantaneous acceleration of the vehicle, measured in m / s². 2 .
[0098] Based on the velocity and time data obtained from the experiment, partial data of acceleration and VSP were calculated using formulas (2-1) and (2-2), as shown in Table 2-2.
[0099] Table 2-2 Examples of Acceleration and Power Specific of Motor Vehicles
[0100]
[0101]
[0102] 2.2 Basic Model of Instantaneous Fuel Consumption of Automobile
[0103] The combustion of fuel in a car is a chemical reaction that obeys the law of conservation of mass, meaning that the total mass of carbon atoms in the fuel before combustion is equal to the total mass of carbon atoms in the exhaust gas produced after combustion in the car engine. The method for calculating instantaneous fuel consumption based on this law is called the carbon balance fuel consumption calculation method. This model is established by acquiring the instantaneous mass of CO2, CO, and HC in the exhaust gas emitted by the car during driving using a PEMS device, and then calculating the instantaneous fuel consumption of the car using the carbon balance fuel consumption calculation method.
[0104] Figure 3 The instantaneous mass of CO2 and hydrocarbons (HC) emitted by the vehicle during operation:
[0105] Instantaneous data, such as carbon dioxide (CO2), obtained by PEMS can be used to calculate the instantaneous fuel consumption of a motor vehicle. After measuring the mass of each carbon-containing gas in the vehicle exhaust using PEMS equipment, it is necessary to determine the proportion of carbon atoms in CO2, CO, and HC, as shown in the table below. The hydrogen-to-carbon ratio of HC in exhaust emissions is the same as that in gasoline, which is 1.85.
[0106] Table 2-3 Mass fraction of carbon atoms in CO2, CO, and HC in exhaust gas
[0107]
[0108] The main components of automobile fuel are a mixture of carbon (C) and hydrogen (H). During fuel combustion, due to the space of the engine's combustion chamber...
[0109] Limited combustion means fuel cannot burn completely, resulting in exhaust emissions containing not only CO2 and H2O, but also trace amounts of CO and hydrocarbons (HC). Therefore, the fuel combustion reaction during vehicle operation is as follows:
[0110] C m H n →CO2+CO+HC
[0111] (2-3)
[0112] To ensure the accuracy of fuel consumption testing, my country's current standard, "Method for Testing and Evaluating Fuel Consumption of Road Transport Vehicles" (GB / T 18566-2011), provides a model for calculating fuel consumption, with a hydrogen-to-carbon ratio of 1.85 for gasoline. Assuming fuel consumption per unit mileage is q (g / km), according to the carbon balance fuel consumption calculation method, the total mass of carbon atoms before and after fuel combustion is the same. Therefore:
[0113] 0.8664 × q = 0.2727 × M CO2 +0.4286×M CO +0.8664×M HC (2-4)
[0114] In the formula:
[0115] M CO2 The instantaneous mass of CO2 is expressed in g / km.
[0116] M CO The instantaneous mass of CO is expressed in g / km.
[0117] M HC The instantaneous mass of HC is expressed in g / km.
[0118] From the transformation of equation (2-4) above, we can obtain:
[0119]
[0120] The instantaneous fuel consumption Q (L / 100km) of a car, calculated from experimental data using the carbon balance fuel consumption calculation method, is as follows:
[0121]
[0122] In the formula: ρ——the density of gasoline at 25℃. The comparison method in the experiment is the mass method. The selection of fuel density has no impact on error analysis, so a fixed value of 0.725kg / L is selected according to the standard.
[0123] Based on the PEMS calculation system, the M can be calculated using the carbon balance fuel consumption calculation method. CO2 M HC Based on this, the instantaneous fuel consumption Q of the car can be calculated using formula (2-7), and some of the data is shown in Table 2-4.
[0124] Table 2-4 Examples of fuel consumption calculated using the carbon balance fuel consumption method
[0125]
[0126] 2.3 Analysis of Factors Affecting Car Fuel Consumption
[0127] 2.3.1 The impact of speed on vehicle fuel consumption
[0128] The test vehicle's speed change during driving is as follows Figure 4 As shown, the instantaneous speed of the vehicle exhibited relatively stable fluctuations during the experiment. However, as the test progressed, real-time road traffic conditions became increasingly complex and variable, leading to acceleration and deceleration, resulting in significant speed changes. The vehicle maintained a stable speed of 10-40 km / h from 1 to 900 seconds and 50-75 km / h from 901 to 1200 seconds. This speed variation during the experiment is representative for analyzing the instantaneous fuel consumption of a passenger car.
[0129] The relationship between instantaneous speed and instantaneous fuel consumption during the entire test is as follows: Figure 5 As shown in the graph, during vehicle operation, the speed range is 0-80 km / h, and the instantaneous fuel consumption exhibits a trend of first decreasing and then increasing. In the 0-50 km / h range, the instantaneous fuel consumption gradually decreases with increasing speed, reaching a minimum around 50 km / h. Subsequently, in the 50-80 km / h range, the instantaneous fuel consumption gradually increases with increasing speed. In the 25-45 km / h range, the instantaneous fuel consumption stabilizes within the range of 5-12.5 L / 100km.
[0130] 2.3.2 The impact of acceleration on vehicle fuel consumption
[0131] The change in acceleration of the test vehicle during driving is as follows Figure 6 As shown, during the experiment, the instantaneous acceleration of the motor vehicle exhibited stable fluctuations. The changes in acceleration were closely related to the changes in velocity per unit time. As the test progressed, real-time road traffic conditions became increasingly complex and variable, leading to sudden accelerations and decelerations, resulting in significant changes in acceleration values. Throughout the entire test, the total time for acceleration and deceleration was roughly equal.
[0132] Throughout the test, the relationship between acceleration and instantaneous fuel consumption was as follows: Figure 7 As shown in the figure, the acceleration variation range of the vehicle during driving is concentrated in the range of -1m / s². 2 -1m / s 2 Within this range, the vehicle's instantaneous fuel consumption is concentrated between 5-15L / 100km. During the test, due to the complex real-time road conditions, the vehicle may suddenly accelerate or decelerate at any time, so the instantaneous fuel consumption varies greatly under the same acceleration.
[0133] 2.3.3 The impact of VSP on vehicle fuel consumption
[0134] VSP is the parameter closest to reality among the parameters for calculating vehicle fuel consumption. Substituting the speed and acceleration measured based on PEMS into formula (2.2) yields the VSP per second. Statistical analysis of the VSP values per second yields the following statistical data: Figure 8 As shown.
[0135] Throughout the entire test, the relationship between the changes in VSP (Variable Savings Per Second) and the instantaneous fuel consumption of the vehicle is as follows: Figure 9 As shown in the figure, the vehicle's VSP variation range per second during driving is concentrated between -5KW / t and 15KW / t, and the instantaneous fuel consumption within this range is concentrated between 5-15L / 100km. During the test, due to the complex real-time road conditions, the vehicle may experience sudden acceleration or deceleration at any time, leading to significant changes in the vehicle's instantaneous speed and acceleration. Therefore, under the same VSP conditions, the vehicle's instantaneous fuel consumption will exhibit diverse variations.
[0136] 2.3.4 The impact of altitude on vehicle fuel consumption
[0137] The test vehicle's altitude change during driving is as follows: Figure 10As shown, the altitude of the vehicle fluctuated steadily during the experiment. However, as the test progressed and the vehicle's route changed, the altitude variation became quite significant.
[0138] Throughout the test, the relationship between altitude and instantaneous fuel consumption was as follows: Figure 11 As shown in the figure, at an altitude of 10-15 meters, the vehicle's instantaneous fuel consumption is stable in the range of 5-10L / 100km. As the altitude increases during the test, the vehicle's instantaneous fuel consumption also tends to increase relatively.
[0139] 3. Fuel Consumption Correction Model Design
[0140] 3.1 Fuel Consumption Correction Model
[0141] 3.1.1 Basic Principles of the Fuel Consumption Correction Model
[0142] Because a vehicle's fuel consumption varies even when its speed, acceleration, altitude, and VSP (vehicle speed and fuel consumption per unit area) are the same, it is necessary to calculate the vehicle's basic fuel consumption during driving and its average fuel consumption at various average speeds. Based on this, the relative impacts of speed, acceleration, altitude, and VSP on fuel consumption should be analyzed, and a fuel consumption correction model should be designed by comprehensively considering these influencing factors.
[0143] 3.1.2 Multiple Linear Regression Equation
[0144] Multiple linear regression is a classic statistical model that predicts a continuous dependent variable from independent variables. As a more scientific method, multiple linear regression analysis is widely used in various fields for both quantitative and qualitative problems. The multiple linear regression equation is a method of establishing a linear regression model by analyzing the relationship between two or more independent variables and one dependent variable. Its specific structure is as follows:
[0145]
[0146] Where α0 is a constant term, and α i (i = 1, 2, ..., i) is Y's pair with X i The regression coefficient parameter value, for a typical multiple linear regression with multiple independent variables, is α for each regression coefficient. i This indicates the average increase in dependent variable Y when independent variable X increases by one unit, assuming other independent variables remain constant in the regression equation.
[0147] Based on this, the structure of the multiple linear regression equation in this design is as follows:
[0148]
[0149] in:
[0150] Q i Let i be the dependent variable, where i = 1, 2, ..., 20, representing vehicle fuel consumption;
[0151] V i Let i be the independent variable, where i = 1, 2, ..., 20, representing the average speed of the vehicle;
[0152] A i Let be the independent variable, where i = 1, 2, ..., 20, representing the average acceleration of the vehicle;
[0153] VSP i Let i be the independent variable, where i = 1, 2, ..., 20, representing the specific power of the vehicle;
[0154] H i Let i be the independent variable, where i = 1, 2, ..., 20, representing the altitude of the vehicle.
[0155] k1 is the intercept, and k2, k3, k4, and k5 are the regression coefficients.
[0156] After calculating the regression model using multiple linear regression analysis, the significance of the regression equation needs to be tested. To ensure the experiment is more representative and to reduce fuel consumption deviations caused by sudden acceleration, deceleration, or other unexpected situations during vehicle operation, this design divides the collected data into 20 groups. The average value of the data per minute represents the vehicle's driving conditions within that interval. The calculated data is shown in Table 3-1 below:
[0157] Table 3-1 Experimental Results Grouping Data
[0158]
[0159] Linear regression analysis of the table above using SPSS software yielded a correlation coefficient R = 0.747 between vehicle average speed and fuel consumption, indicating a good fit between the average speed and fuel consumption. Based on this, assuming the data conforms to the conditions of a quaternary linear regression equation, the fuel consumption model equation is obtained through analysis:
[0160] Q i =15.025-0.11V i -0.294A i -0.051H i +0.026VSP i (3-1)
[0161] In the formula: the correlation coefficient R of the linear equation is 0.775, indicating a good fit of the linear model. Substituting the above data, the baseline...
[0162] The results calculated using Formula 3-1 and the comparison are shown in Table 3-2 below:
[0163] Table 3-2 Regression Model Calculation Results and Comparison
[0164]
[0165] As can be seen from Table 3-2, the average deviation rate of this model is 8.03%, and the correlation coefficient is relatively high at 0.775. Therefore, the model used in this design is simple and practical, and has a good fitting effect, and can be used as a vehicle fuel consumption model.
[0166] 3.2 Fuel Consumption Model Validation
[0167] To verify the rationality of the designed model, a significance test is needed. A significance test examines whether the relationship between the independent and dependent variables is significant. Specifically, it examines whether the linear relationship between speed, acceleration, altitude, VSP, and vehicle fuel consumption is significant within the model. Based on the established fuel consumption model, the model is tested using SPSS software through an F-test.
[0168] During model testing, the null hypothesis is: k = 0, meaning the regression coefficient is equal to 0. When the regression coefficient is zero, there is no relationship between the two variables. The F-statistic is used as the test statistic, and its formula is (3-2):
[0169]
[0170] In the formula:
[0171] SSR stands for Sum of Squares of Regression;
[0172] SSE is the sum of squared residuals;
[0173] p is the number of independent variables;
[0174] n is the number of samples.
[0175] The F-statistic follows an F-distribution with (4, n-5) degrees of freedom. SPSS software was used to test the significance of the model. SPSS automatically calculates the observed value F of the test statistic. If F > F(4, 15), the relationship between the independent variable and vehicle fuel consumption is significant; otherwise, the relationship is considered insignificant.
[0176] From the PEMS data, p = 4 and n = 20. SPSS analysis shows that SSR is the regression sum of squares and SSE is the residual sum of squares, yielding an F-value of 5.63. Taking a significance level of α = 0.01, and consulting the F-distribution critical value table, we find F(4, 15) = 4.89. Since F = 5.63 > F(4, 15) = 4.89, the regression line obtained from the experimental data is highly significant. The significance test results of the fuel consumption correction model in this design using SPSS statistical analysis software show that the model has good significance and meets the requirements for significance testing.
[0177] 3.3 Establishment of the Correction Coefficient Range
[0178] A correction factor is a calculation method used to adjust the model when there is a deviation between the model's predicted values and the actual measured values, thereby improving the model's goodness of fit to the actual measurements. Introducing a correction factor can improve the accuracy and reliability of model predictions, making them more suitable for real-world applications. The application of correction factors can, to some extent, eliminate the discrepancy between measured data and model predictions, improving the reliability and practicality of the prediction results.
[0179] Among the factors affecting vehicle fuel consumption, the correlation coefficient between vehicle average speed and fuel consumption is the largest, indicating that the relationship between vehicle average speed and vehicle fuel consumption is the strongest. Based on the actual fuel consumption of the experimental vehicle and the fuel consumption results from the model, the speed correction coefficient H for fuel consumption in each speed range is calculated according to formula (3-3). vi .
[0180]
[0181] The calculation results of the speed correction factor are shown in Table 3-3 below:
[0182] Table 3-3 Speed Range and Correction Factor H vi
[0183]
[0184] As can be seen from Table 3-3, when the speed is between 0-20 km / h, the speed correction factor is 1.00. As the speed increases, the speed correction factor also increases slightly.
[0185] The acceleration correction coefficient H for vehicle fuel consumption in each acceleration range is calculated using formula (3-4). Ai .
[0186]
[0187] The calculation results of the acceleration correction coefficient are shown in Table 3-4 below:
[0188] Table 3-4 Acceleration Range and Correction Factor H Ai
[0189]
[0190] As can be seen from Table 3-4, when the acceleration is in the range of (-2) to (2) m / s 2 When the acceleration is greater than 2 m / s², the acceleration correction factor is 1.00. 2 At this point, the velocity correction coefficient also increases slightly with increasing acceleration. When the acceleration is less than -2 m / s², the correction factor also increases slightly. 2 At that time, as the acceleration decreases, the velocity correction coefficient also increases slightly.
[0191] The fuel consumption correction factor H for each altitude range is calculated using formula (3-5). Hi .
[0192]
[0193] Altitude correction factor H Hi The calculation results are shown in Table 3-5 below:
[0194] Table 3-5 Altitude Ranges and Correction Factors H Hi
[0195]
[0196] As can be seen from Table 3-5, when the altitude is between 0 and 5 m, the correction factor is 1.00. As the altitude increases, the altitude correction factor also increases slightly.
[0197] The fuel consumption correction coefficient for each altitude range is calculated using formula (3-6).
[0198]
[0199] Vehicle power-to-weight ratio correction factor The calculation results are shown in Table 3-6 below:
[0200] Table 3-6 VSP Range and Correction Factor
[0201]
[0202] As shown in Table 3-6, when VSP is between -20 and 20 km / t, the correction factor is 1.00. When VSP is greater than 20 kW / t, the VSP correction factor also increases slightly with increasing VSP. When VSP is less than -20 kW / t, the VSP correction factor also increases slightly with decreasing VSP.
[0203] This design studies motor vehicles, selecting those meeting the China V emission standard, comparing the advantages and disadvantages of different testing methods, and determining the appropriate on-board emission testing method. Instantaneous data on motor vehicle emissions are collected, and a fuel consumption correction model is established based on this data. The relationship between instantaneous speed, instantaneous acceleration, vehicle altitude, and specific power of the vehicle and its fuel consumption is analyzed. The carbon balance fuel consumption calculation method is used to calculate the instantaneous fuel consumption of the motor vehicle. Based on the measured effects of speed, acceleration, altitude, and VSP on vehicle fuel consumption, correction coefficients for each influencing factor are determined, allowing for a more accurate prediction of the characteristics of fuel consumption changes on the road. The main research results and conclusions are as follows:
[0204] (1) The basic fuel consumption model is designed by carbon balance fuel consumption calculation method. The basic fuel consumption model is corrected by comprehensively considering speed, acceleration, altitude and VSP through multiple linear regression model, so as to obtain a more accurate fuel consumption model and better reflect the actual operating conditions of light gasoline vehicles.
[0205] (2) The effect of vehicle instantaneous speed on fuel consumption is quite significant. In the range of 0-60 km / h, the instantaneous fuel consumption gradually decreases with increasing speed, reaching a minimum value around 60 km / h. Based on common knowledge of motor vehicle fuel consumption, after 60 km / h, the instantaneous fuel consumption gradually increases with increasing speed. In the speed range of 25-45 km / h, the instantaneous fuel consumption is stable in the range of 5-12.5 L / 100km.
[0206] (3) By dividing the speed, acceleration, altitude and VSP into intervals, the correction coefficients of each influencing factor in different intervals are obtained. The determination of the correction coefficients makes the design of the vehicle fuel consumption model more rigorous and convincing.
[0207] The innovative aspects of this design are as follows:
[0208] (1) In terms of research methods and techniques, this design adopted advanced PEMS equipment and used vehicle emission testing method, which made the designed model more representative.
[0209] (2) The PEMS device was used to measure various parameters of the motor vehicle in real time, and instantaneous data of the motor vehicle during driving were obtained. The carbon balance fuel consumption calculation method was used to establish a basic fuel consumption model. The basic fuel consumption was corrected by the multiple linear regression model to establish a more accurate fuel consumption correction model.
[0210] (3) This design takes into account the effects of vehicle speed, acceleration, altitude and specific power of motor vehicles on fuel consumption. Most other studies only consider the effects of one factor on fuel consumption. However, in practice, comprehensive consideration is more convincing.
Claims
1. A method for optimizing the design of a car fuel consumption correction model based on PEMS, characterized in that, Includes the following steps: Step 1: Propose a fuel consumption correction model; Step 2: Validate the fuel consumption correction model; Step 3: Determine the range of coefficients for the modified model; Step 1 specifically involves: calculating the basic fuel consumption of the vehicle during driving and the average fuel consumption at each average speed; based on this, analyzing the relative influence of speed, acceleration, altitude, and VSP on fuel consumption; and designing a fuel consumption correction model by comprehensively considering the influencing factors. Establish a multiple linear regression equation: wherein: is the dependent variable, i = 1, 2, …, 20, representing the vehicle fuel consumption; is the independent variable, representing the vehicle average speed; is the independent variable, representing the vehicle average acceleration; is the independent variable, representing the vehicle specific power; is the independent variable, representing the vehicle altitude; is the intercept, , , , is the regression coefficient; after calculating the regression model by using the multiple linear regression analysis prediction method, the regression equation significance test is performed on the model; Linear regression analysis revealed a correlation coefficient of 0.747 between the vehicle's average speed and the vehicle's fuel consumption correction model. The equation for the fuel consumption correction model is as follows: In the formula: the correlation coefficient of the linear equation is 0.775, indicating a good fit of the linear model; The step 3 is specifically: according to the actual fuel consumption of the experimental vehicle and the model result fuel consumption, the speed correction coefficient H of the fuel consumption of each speed interval is calculated according to the following formula vi ; The acceleration correction factor for the motor vehicle fuel consumption for each acceleration interval is calculated according to the following formula ; The fuel consumption correction coefficient of each altitude interval is calculated according to the following formula ; The fuel consumption correction coefficient of each altitude interval is calculated according to the following formula ; 。 2. The PEMS-based passenger car fuel consumption correction model optimization design method of claim 1, wherein, Step 2 specifically involves: assuming k=0, meaning the regression coefficient is equal to 0, when the regression coefficient is zero, there is no relationship between the two variables; using the F-statistic as the test statistic, its formula is as follows: In the formula: SSR is the regression sum of squares; SSE is the residual sum of squares; p is the number of independent variables; n is the sample size; The F-statistic follows an F-distribution with (4, n-5) degrees of freedom. SPSS software is used to test the significance of the model. SPSS automatically calculates the observed value F of the test statistic. If F > F(4, 15), the relationship between the independent variable and vehicle fuel consumption is significant; otherwise, the relationship between the independent variable and vehicle fuel consumption is not significant. From the PEMS data, we know that p=4 and n=20. Through SPSS analysis, we find that SSR is the regression sum of squares and SSE is the residual sum of squares, and the calculated F value is 5.
63. Taking the significance level α=0.01, by consulting the F distribution critical value table, we know that F(4,15)=4.
89. F=5.63> F(4,15)=4.89, indicating that the regression line obtained from the experimental data is highly significant. The significance test results of the fuel consumption correction model using SPSS statistical analysis software show that the model has good significance and meets the significance test requirements.
Citation Information
Patent Citations
Hadoop-based petroleum transport vehicle oil consumption prediction method
CN107239844A
An energy consumption reference correction method for transformer substation energy efficiency evaluation
CN109766517A
