A joint estimation method for heavy commercial vehicle load and road slope

Abstract

The present invention proposes a method for jointly estimating the load and road slope of a heavy-duty commercial vehicle, comprising the following steps: S1, collecting data information generated by the vehicle during driving and transmitting the data information to a data storage and processing platform via a wireless network; S2, segmenting the driving data into segments; S3, screening valid driving data: when the vehicle speed, acceleration, and engine output power at a certain moment are all greater than corresponding thresholds, the data at that moment is screened as valid driving data; S4, establishing a mathematical model for jointly estimating the load and road slope that takes into account the driving resistance coefficient of the heavy-duty commercial vehicle; S5, jointly estimating the load and road slope, thereby analyzing the vehicle mass of the heavy-duty commercial vehicle and the slope conditions at each moment. The present invention can jointly estimate the load and road slope of a heavy-duty commercial vehicle when the vehicle's entire vehicle and assembly design parameters are unknown; and has the characteristics of high real-time performance and a wide monitoring area.

Description

Technical Field

[0001] The present invention relates to the technical field of estimating load and road slope, and in particular to a method for jointly estimating the load and road slope of a heavy commercial vehicle. Background Art

[0002] With the rapid development of the Chinese economy and the accelerated pace of urbanization, the demand for freight transportation is high. In pursuit of economic benefits, overloading is becoming increasingly serious. Overloaded vehicles can easily lead to serious traffic accidents, severely damage transportation infrastructure such as roads and bridges, and promote vicious competition in the transportation market. Therefore, it is necessary to address the issue of vehicle overloading from a regulatory perspective.

[0003] There are two main methods for measuring vehicle load. The first involves building monitoring stations and using weighing equipment such as scales to measure vehicle weight. This method suffers from low efficiency, fixed monitoring points, and a limited monitoring area. The second method involves installing sensors on the vehicle's suspension system and using the information collected by the sensors to determine vehicle weight. However, this method suffers from high sensor costs, labor-intensive calibration processes, limited sensor installation locations, and difficulty maintaining measurement accuracy over long periods of use. Summary of the Invention

[0004] The present invention aims to at least solve the technical problems existing in the prior art, and in particular innovatively proposes a method for jointly estimating the load of a heavy commercial vehicle and the road slope.

[0005] To achieve the above-mentioned object of the present invention, the present invention provides a method for jointly estimating the load of a heavy commercial vehicle and the road slope, comprising the following steps:

[0006] S1, collects data information generated by the vehicle during driving and transmits the data information to the data storage and processing platform via a wireless network;

[0007] S2, segmenting driving data segments: Based on the vehicle speed and data collection time, the parking status of the vehicle is judged, and the length of the vehicle's parking time is calculated. The driving data segments between two long parking states that meet the medium- and long-distance driving conditions are screened out; generally, more than 0.5 hours is considered a long parking time, and more than 50 kilometers is considered a medium- and long-distance driving time.

[0008] S3, screening valid driving data: when the vehicle speed, acceleration and engine output power at a certain moment are all greater than the corresponding thresholds, the data at that moment is screened as valid driving data;

[0009] S4, establishing a mathematical model for joint estimation of load and road slope considering the driving resistance coefficient of heavy commercial vehicles;

[0010] S5, joint estimation of load and road slope: The system state variables are selected as vehicle speed, vehicle mass and road slope, and the system observation variable is selected as vehicle speed. It is assumed that the mass of the vehicle does not change during driving and the slope changes slowly. The system state equation and system measurement equation are established to analyze the vehicle mass of the heavy commercial vehicle and the slope conditions at each moment.

[0011] Furthermore, the data information includes: time, vehicle speed, engine speed, engine net output torque percentage, and engine friction torque percentage.

[0012] Furthermore, the engine output power at each moment is calculated as follows:

[0013]

[0014] P e Indicates engine output power;

[0015] n e Indicates engine speed;

[0016] C te_a Indicates the percentage of net engine output torque;

[0017] C te_f Indicates the percentage of engine friction torque;

[0018] T e_n Indicates the maximum base engine torque.

[0019] Furthermore, the mathematical model for joint estimation of load and road slope is as follows:

[0020] a k =[3600P e_k η t -(C A +C B v k +C C v k 2 +m k gi k )v k ] / δm k v k (4)

[0021]

[0022] Where a k is the acceleration at the kth moment;

[0023] P e_k is the engine output power at the kth moment;

[0024] η t is the mechanical efficiency of the transmission system, which is taken as 85%;

[0025] v k is the vehicle speed at the kth moment;

[0026] m k is the estimated value of mass at the kth moment;

[0027] g is the acceleration due to gravity, which is 9.8m / s2;

[0028] i k is the estimated value of the road slope at the kth moment;

[0029] δ is the equivalent translation mass;

[0030] C A 、C B and C C are all constant coefficients, and the mass estimate m at the kth moment is k The recommended values ​​of driving resistance coefficient for medium and heavy-duty commercial vehicles in GB / T 27840-2021 are obtained by linear interpolation.

[0031] Furthermore, the system state equation is:

[0032]

[0033] The system measurement equation is:

[0034] Z k =v k =[1 0 0][v k m k i k ] T +G k (6)

[0035] Where, X k is the system state variable at the kth moment;

[0036] Z k is the system measurement variable at the kth moment;

[0037] v k is the vehicle speed at the kth moment;

[0038] m k is the estimated value of mass at the kth moment;

[0039] i k is the estimated value of the road slope at the kth moment;

[0040] Δt is the time interval;

[0041] a k-1 is the acceleration at the k-1th moment, calculated by formula (4);

[0042] W k-1 is the process noise;

[0043] [] T is the transpose of the matrix;

[0044] G k To measure noise.

[0045] Furthermore, the optimal estimation algorithm is used to estimate the system state value X of the effective driving data. k The optimal estimation algorithm is an extended Kalman filter or an unscented Kalman filter. The extended Kalman filter expands the nonlinear state equation into a Taylor series around the filter value and omits the second-order and higher terms to obtain an approximate linear model of the nonlinear system. Then, an algorithm similar to the linear Kalman filter is used for state estimation. The unscented Kalman filter uses an unscented transform to deal with the nonlinear transmission problem of the mean and covariance, approximating the probability density distribution of the nonlinear function, and thus performing state estimation.

[0046] Furthermore, during the data transmission process, encryption is performed in the following manner:

[0047] The data to be transmitted is encrypted using an encryption key, the expression of the key is:

[0048] sk=α×log(f(x)+ζ)

[0049] Where sk is the secret key;

[0050] α is the learning rate;

[0051] log is the logarithmic function;

[0052] f(x) is the output value of the neural network, and x is the data information converted into binary;

[0053] ζ is the disturbance factor;

[0054] Then the obtained sk is normalized to obtain the final encrypted information sk′:

[0055] sk′=Nor(sk)+x

[0056] Nor() is the normalization process.

[0057] The general method is that the vehicle sender and the processing platform as the receiver use the same secret key. However, this situation is prone to leaks. Therefore, generating a dynamic secret key through the above method can greatly improve security.

[0058] In summary, due to the adoption of the above-mentioned technical solution, the present invention can jointly estimate the load and road slope of heavy-duty commercial vehicles when the vehicle's complete vehicle and assembly design parameters are unknown; it has the characteristics of high real-time performance and a wide monitoring area; it can provide regulatory assistance on vehicle overloading issues and reduce monitoring costs; and it can assist in the implementation of freight efficiency (i.e., fuel consumption per unit cargo load mass, expressed in liters per ton per hundred kilometers [L / (t*100km)]) evaluation of logistics transport fleets.

[0059] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0061] Figure 1 The present invention is a flowchart of a method for jointly estimating the load of a heavy commercial vehicle and the road slope.

[0062] Figure 2 It is the combined estimation result of load and road slope of the present invention.

[0063] Figure 3 It is the combined estimation result of load and road slope of the present invention. DETAILED DESCRIPTION

[0064] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0065] A flowchart of a joint estimation method for heavy commercial vehicle load and road slope is shown in the figure. Figure 1 As shown, the following steps are included:

[0066] Step 1. Collect vehicle driving data. Using a data acquisition terminal, collect driving data such as time (s), speed (km / h), engine speed (rpm), engine net output torque percentage (%), and engine friction torque percentage (%), and transmit it to a data storage and processing platform via a wireless network.

[0067] Step 2. Segment the driving data. Based on vehicle speed and data collection time, determine the vehicle's parking status and calculate the length of parking time. Define vehicles traveling more than 50 km as being in a medium- to long-distance driving state, and vehicles parked for more than 0.5 hours as being in a long-term parking state. Filter out driving data segments between two long-term parking periods that meet the medium- to long-distance driving criteria.

[0068] Step 3. Filter valid driving data. For each driving data segment, the acceleration at each moment can be calculated by differentiating the vehicle speed over time. The engine output power at each moment can be calculated using the engine speed, engine net output torque percentage, engine friction torque percentage, and engine maximum reference torque data, as shown in the following formula:

[0069]

[0070] Where, P e Indicates engine output power (kW), n e Indicates engine speed (rpm), C te_a Indicates the percentage of engine net output torque (%), C te_f Indicates the percentage of engine friction torque (%), T e_n Indicates the maximum base engine torque (Nm).

[0071] The vehicle speed threshold, acceleration threshold, and engine output power threshold are set to 40 km / h, 0.15 m / s2, and 10 kW, respectively. In the driving data segment, when the vehicle speed, acceleration, and engine output power at a certain moment are all greater than the corresponding thresholds, the data at that moment are filtered as valid driving data.

[0072] Step 4. Establish a mathematical model for the joint estimation of load and road slope. Assuming that the rolling resistance system of a truck tire has an approximately linear relationship with vehicle speed, its longitudinal dynamics equation is shown below.

[0073]

[0074] Among them, F t is the driving force (N), P e is the engine output power (kW), η t is the mechanical efficiency of the transmission system, v is the vehicle speed (km / h), F f is the rolling resistance (N), f0 and f1 are the velocity fitting constant and linear term of the rolling resistance system, m is the vehicle mass (kg), including the loaded mass and the curb mass, g is the acceleration of gravity (m / s2), α is the ramp angle, and F wis the air resistance (N), C D is the air resistance coefficient, A is the frontal area (m2), F i is the slope resistance (N), F j is the acceleration resistance (N), δ is the translational mass equivalent coefficient, and a is the acceleration (m / s2).

[0075] Assume that the road slope i = tanα. Since the actual road slope is small, we can assume that sinα≈tanα=i and cosα≈1. Therefore, equation (3) can be further derived from equation (2).

[0076] F t ≈mgf0+mgf1v+C D Av 2 / 21.15+mgi+δma (3)

[0077] Since the driving resistance of different types of heavy-duty commercial vehicles varies greatly and it is difficult to obtain the actual driving resistance of the estimated object, a mathematical model for joint estimation of load and road slope was established based on the vehicle longitudinal driving dynamics equation and combined with the driving resistance coefficient of heavy-duty commercial vehicles or the driving resistance coefficient obtained from road coasting tests, as shown in the following formula:

[0078] a k =[3600P e_k η t -(C A +C B v k +C C v k 2 +m k gi k )v k ] / δm k v k (4)

[0079]

[0080] Where a k is the acceleration at the kth moment (m / s2); v k is the vehicle speed at the kth moment (km / h), m k is the estimated mass value at the kth moment (kg); i k is the estimated value of the road slope at the kth moment; P e_k is the engine output power (kW) at the kth moment, calculated by formula (1); η t is the mechanical efficiency of the transmission system, which is taken as 85%; δ is the equivalent translational mass.

[0081] In the mathematical model for joint estimation of load and road slope, C A 、C B and C C are all constant coefficients, and the mass estimate m at the kth moment is k The recommended values ​​of driving resistance coefficient for medium and heavy-duty commercial vehicles in GB / T 27840-2021 are obtained by linear interpolation.

[0082] Step 5. Joint estimation of load and road slope. Select the system state variables as vehicle speed, vehicle mass, and road slope, select the system observation variable as vehicle speed, and assume that the mass does not change during the vehicle's driving and the slope changes slowly. Establish the system state equation and system measurement equation, as shown in Equation (5) and Equation (6), respectively:

[0083]

[0084] Z k =v k =[1 0 0][v k m k i k ] T +G k (6)

[0085] Where, X k is the system state variable at the kth moment; Z k is the system measurement variable at the kth moment; a k-1 is the acceleration at the k-1th moment (m / s2), calculated by formula (4); Δt is the time interval (s); W k-1 is the process noise, and its error covariance matrix is ​​Q; G k is the measurement noise, and its error covariance matrix R.

[0086] Based on the mathematical model of joint estimation of load and road slope, the optimal estimation algorithm is used to estimate the system state value X of the effective driving data. k Make an estimate, X k Including the vehicle speed v at the kth moment k , mass estimate m k and the estimated road slope i k For invalid data segments, the current system state value X k Equal to the system state value X at the previous moment k-1 .

[0087] The mathematical model for the joint estimation of load and road slope indicates that the system is nonlinear, so either the extended Kalman filter or the unscented Kalman filter can be chosen as the optimal estimation method. The extended Kalman filter expands the nonlinear state equation into a Taylor series around the filter value and ignores second-order and higher terms to obtain an approximate linear model of the nonlinear system. State estimation is then performed using an algorithm similar to the linear Kalman filter. The unscented Kalman filter utilizes an unscented transform to address the nonlinear transfer of mean and covariance, approximating the probability density distribution of the nonlinear function for state estimation.

[0088] Test 1: Taking a truck with a curb mass of 3290kg and an actual test mass of 4400kg as an example, for a selected driving data segment between two long-term stops and meeting the medium-to-long distance driving conditions, the speed curve is as follows: Figure 2 (a) shows the total mileage of 119.4 km, with a total of 12629 seconds of data. The valid driving data with speed, acceleration and engine output power all exceeding the corresponding thresholds is 2379 seconds. Set the initial state variable X(0) = [1 5000 0] T , process noise error covariance matrix Q = diag (0.04, 1600, 0.01), measurement noise error covariance matrix R = 0.1, C A 、C B and C C The initial values ​​are 571.95, 2.795 and 0.112 respectively.

[0089] For the valid driving data segment, based on the mathematical model of joint estimation of load and road slope, the unscented Kalman filter method is used as the optimal estimation method. The weight estimation results and slope estimation results are as follows: Figure 2 (b) and Figure 2 As shown in (c), the average weight estimate of the valid data segment is 4560 kg, which has an estimated error of 3.6% compared to the actual mass of 4400 kg.

[0090] Test 2: Taking a commercial vehicle with a vehicle test mass of 7200kg as an example, for a selected driving data segment between two long-term stops and meeting the medium-to-long distance driving conditions, the vehicle speed curve is as follows: Figure 3 (a) shows the total mileage of 113.8 km, with a total of 12557 seconds of data. The valid driving data with speed, acceleration and engine output power all exceeding the corresponding thresholds is 2332 seconds. Set the initial state variable X(0) = [1 5000 0] T, the process noise error covariance matrix Q = diag (0.04, 1600, 0.01), the measurement noise error covariance matrix R = 0.1. Based on the mathematical model of joint estimation of load and road slope, the unscented Kalman filter algorithm is used to jointly estimate the load and road slope. The weight estimation results and slope estimation results of the effective driving data segment are as follows: Figure 3 (b) and Figure 3 As shown in (c), the average weight estimate of the valid data segment is 7478 kg, which has an estimated error of 2.4% compared to the actual mass of 7200 kg.

[0091] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

Claims

1. A joint estimation method for heavy commercial vehicle load and road slope, characterized in that: The following steps are involved: S1, collects data information generated by the vehicle during driving and transmits the data information to the data storage and processing platform via a wireless network; S2, segmenting driving data segments: judging the vehicle's parking state based on the vehicle speed and data collection time, calculating the length of the vehicle's parking time, and filtering out driving data segments between two long parking states that meet the medium- and long-distance driving conditions; S3, screening valid driving data: when the vehicle speed, acceleration and engine output power at a certain moment are all greater than the corresponding thresholds, the data at that moment is screened as valid driving data; S4, establishing a mathematical model for jointly estimating the load and road slope taking into account the driving resistance coefficient of heavy commercial vehicles; the mathematical model for jointly estimating the load and road slope is as follows: Where a k is the acceleration at the kth moment; P e_k is the engine output power at the kth moment; η t is the mechanical efficiency of the transmission system; v k is the vehicle speed at the kth moment; m k is the estimated value of mass at the kth moment; g is the acceleration due to gravity, which is 9.8 m / s 2 ; i k is the estimated value of the road slope at the kth moment; δ is the equivalent translation mass; C A 、C B and C C All are constant coefficients; f0 and f1 are the velocity fitting constant and linear term of the rolling resistance system respectively; C D is the air resistance coefficient; A is the windward area; S5, joint estimation of load and road slope: The system state variables are selected as vehicle speed, vehicle mass and road slope, and the system observation variable is selected as vehicle speed. It is assumed that the mass of the vehicle does not change during driving and the slope changes slowly. The system state equation and system measurement equation are established to analyze the vehicle mass of the heavy commercial vehicle and the slope conditions at each moment.

2. A method for jointly estimating the load and road slope of a heavy commercial vehicle according to claim 1, characterized in that: The data information includes: time, vehicle speed, engine speed, engine net output torque percentage and engine friction torque percentage.

3. The method for jointly estimating the load and road slope of a heavy commercial vehicle according to claim 1, characterized in that: The engine output power at each moment is calculated as follows: P e Indicates engine output power; n e Indicates engine speed; C te_a Indicates the percentage of net engine output torque; C te_f Indicates the percentage of engine friction torque; T e_n Indicates the maximum base engine torque.

4. The method for jointly estimating the load and road slope of a heavy commercial vehicle according to claim 1, characterized in that: The system state equation is: The system measurement equation is: Z k =v k =[100][v k m k i k ] T +G k (6) Where, X k is the system state variable at the kth moment; Z k is the system measurement variable at the kth moment; v k is the vehicle speed at the kth moment; m k is the estimated value of mass at the kth moment; i k is the estimated value of the road slope at the kth moment; Δt is the time interval; a k-1 is the acceleration at the k-1th moment, calculated by formula (4); W k-1 is the process noise; [] T is the transpose of the matrix; G k To measure noise.

5. A method for jointly estimating the load and road slope of a heavy commercial vehicle according to claim 4, characterized in that: The system state value X of the effective driving data is estimated by using the optimal estimation algorithm k Estimation is performed, and the optimal estimation algorithm is an extended Kalman filter or an unscented Kalman filter.

Images