Control method for driving motor system of electric tractor

Abstract

The present invention discloses a control method for a driving motor system of an electric tractor, which collects the driving motor speed ω in real time. m [n] and output torque T[n], calculate the vehicle acceleration a[n] and vehicle traction F[n] at time n, and calculate the estimated curb weight M*[n] and driving resistance F*[n] at time n based on the vehicle acceleration a[n] and vehicle traction F[n]. L [n], the estimated value of the curb mass and the estimated value of the driving resistance are introduced into the open-loop transfer function G(s) of the drive motor control loop. The present invention has the advantages of reducing the impact of changes in the traction mass and the driving resistance and ensuring the consistency of the acceleration characteristics.

Description

Technical Field

[0001] The present invention relates to the technical field of tractor driving control, in particular to a control method for a driving motor system of an electric tractor. Background Art

[0002] During actual traction, the tractor's traction mass is not fixed, especially when the vehicle is empty and fully loaded, with a significant difference in curb mass. This results in significant differences in acceleration characteristics corresponding to different traction masses, even when the accelerator pedal depth remains the same. This manifests itself as "acceleration overshoot" when lightly loaded or "acceleration undershoot" when heavily loaded, making operation more difficult for the driver. This problem becomes even more pronounced when driving with varying resistance, especially when driving on roads with varying slopes.

[0003] Therefore, there is an urgent need for a technical solution that can reduce the impact of changes in traction quality and driving resistance. Summary of the Invention

[0004] The purpose of the present invention is to address the deficiencies of the prior art and to provide a method for controlling a drive motor system of an electric tractor that can reduce the effects of variations in traction mass and driving resistance and ensure consistency in acceleration characteristics.

[0005] The technical solution adopted by the present invention to solve its technical problem is:

[0006] A method for controlling a drive motor system of an electric tractor, characterized by:

[0007] Step S1: Real-time acquisition of the driving motor speed ω m [n] and output torque T[n], calculate the vehicle acceleration a[n] and vehicle traction F[n] at time n, and calculate the estimated curb weight M*[n] and driving resistance F*[n] at time n based on the vehicle acceleration a[n] and vehicle traction F[n]. L [n];

[0008] Step S2: Introducing the estimated curb mass and driving resistance into the open-loop transfer function G(s) of the drive motor control loop. The tractor's curb mass and driving resistance are estimated online during driving, and these estimates are introduced into the drive motor control loop to mitigate the impact of changes in curb mass and driving resistance. This ensures consistent acceleration characteristics across varying tractor weights and slopes. The estimated curb mass and driving resistance are estimated solely from the motor speed and output torque, ensuring the accuracy of these estimates even on sloping roads without the need for additional sensors. Furthermore, changes in curb mass can be used to determine whether cargo has been left behind during driving, allowing for appropriate alerts.

[0009] The calculation method of the vehicle acceleration a[n] and the vehicle traction force F[n] at time n in step S1 of the present invention is:

[0010] a[n] = (ω m [n] - ω m [n-1])*K V / T S ;

[0011] F[n] = T[n]* K T ;

[0012] Among them, ω m [n] is the motor speed at time n, ω m [n-1] is the motor speed at time n-1, K V is the vehicle speed and motor speed coefficient, T S is the real-time sampling period, K T It is the conversion coefficient between the motor output torque and the vehicle driving force.

[0013] In step S1 of the present invention, the initialization of the estimated value of the curb mass M* is the vehicle's unloaded mass M min , estimated driving resistance F* L is 0, that is:

[0014] F* L = 0;

[0015] M*=M min ;

[0016] At rest, the estimated driving resistance F* at time n L [n] is the traction force output by the motor system, and the estimated value of the curb mass is M*, that is:

[0017] F* L = F* L [n]=F[n];

[0018] M*[n] =M*.

[0019] In step S1 of the present invention, the estimated value of the curb mass M*[n] and the estimated value of the driving resistance F*[n] at time n are calculated based on the vehicle acceleration a[n] and the vehicle traction force F[n]. L The method of [n] is:

[0020] F[n-1]-F L [n-1]=M[n-1]*a[n-1];

[0021] F[n]-F L [n] = M[n]*a[n];

[0022] Among them, F L [n] is the driving resistance at time n, M[n] is the curb weight at time n, F L [n-1] is the driving resistance at time n-1, M[n-1] is the curb weight at time n-1;

[0023] Let F* L [n]=F L [n-1] =F L [n],M * [n]=M[n-1]= M[n], solve the estimated value of driving resistance F* at time n L [n] and the estimated curb mass M*[n].

[0024] In step S1 of the present invention, the estimated value of the running resistance F* is calculated. L [n] and the estimated value of curb mass M*[n], F* L [n] and M*[n] are low-pass filtered to obtain M* and F* L .

[0025] The specific method for estimating the curb mass and the running resistance in step S1 of the present invention is:

[0026] Step S101: Start, execute S102;

[0027] Step S102: Parameter initialization, F * L = 0, M * =M min , then execute S103;

[0028] Step S103: According to the sampling period T s Sampling, real-time sampling of the driving motor speed ω m [n] and output torque T[n], and calculate the vehicle acceleration a[n] and vehicle traction F[n] at time n;

[0029] a[n] = (ω m [n] - ω m [n-1])*K V / T S ;

[0030] F[n] = T[n]* K T ;

[0031] Among them, ω m [n] is the motor speed at time n, ω m [n-1] is the motor speed at time n-1, K V is the vehicle speed and motor speed coefficient, T Sis the real-time sampling period, K T is the conversion coefficient between the motor output torque and the vehicle driving force;

[0032] Step S104: Determine whether the vehicle is stationary, if yes, execute S105, if no, execute S106;

[0033] Step S105: ω m [n] = 0, record the estimated driving resistance value F * L =T[n]* K T , then execute S103;

[0034] Step S106: Based on Newton's second law, a vehicle dynamics model is constructed. The two sets of data calculated based on the latest sampling form the following equations:

[0035] F[n-1] -F L [n-1] = M[n-1]*a[n-1];

[0036] F[n] -F L [n] = M[n]*a[n];

[0037] Let F* L [n]=F L [n-1] =F L [n],M * [n]=M[n-1]= M[n], the estimated value of curb mass M is obtained based on the calculated vehicle acceleration and vehicle traction * [n] and estimated driving resistance F* L [n];

[0038] After the calculation is completed, execute S107;

[0039] Step S107: Determine the current M * [n] and M * Whether the difference exceeds the set threshold, if it exceeds the set threshold, the quality change flag S_M is set to 1 and S108 is executed; otherwise, the quality change flags S_M and S_M_P are set to 0, that is, S_M=0, S_M_P=0, and S111 is executed;

[0040] S_M is a status indication that the estimated curb mass value exceeds the set threshold value compared with the historical estimated curb mass value, and S_M_P indicates whether the existing data after the curb mass change meets the estimated mass and resistance requirements;

[0041] Step S108: Determine whether the flags S_M and S_M_P are both equal to 1. If yes, execute S109; if not, execute S110.

[0042] Step S109: Determine M * Does the difference between [n] and R_M exceed the set threshold? If yes, execute S103; if no, execute S111;

[0043] When M * When the difference between [n] and R_M exceeds the set threshold:

[0044] R_M=M * [n];

[0045] R_M records the current estimated instantaneous quality;

[0046] When R_F * L With F * L When the difference of [n] is within the set threshold range, update F according to the following formula * L ,

[0047] F * L = K FIL * F * L [n]+ (1-K FIL )*F * L ;

[0048] R_F * L = F * L [n];

[0049] where K FIL is the filter coefficient, 0 <K FIL <1, R_F * L Record the current estimated instantaneous driving resistance;

[0050] Step S110: Record the current estimated curb mass and running resistance, and set the mass change flag S_M_P to 1, that is:

[0051] R_M=M * [n];

[0052] R_F * L = F * L [n];

[0053] S_M_P=1;

[0054] and execute S103;

[0055] Step S111: Correct M using the latest estimated curb mass and running resistance. * and F * L , the calculation formula is as follows:

[0056] F * L = K FIL * F * L [n]+ (1-K FIL )*F * L ;

[0057] M * = K FIL *M * [n] + (1-K FIL )*M * ;

[0058] And execute S103.

[0059] If the present invention enters S103 from S102, it is necessary to continuously sample multiple times and record the motor speed and motor output torque of each sampling respectively;

[0060] If S203 is entered from other steps, sampling is only required once to record the motor speed and motor output torque.

[0061] In step S106 of the present invention,

[0062] When F[n] = 0,

[0063] M * [n] =M * ;

[0064] F * L [n] = F[n]-M * [n] * a[n];

[0065] When a[n] = 0,

[0066] M * [n] =M * [n-1];

[0067] F * L [n] = F[n];

[0068] When a[n] ≠ 0 and a[n]=a[n-1], calculate M according to the following formulas: * [n] and F *L [n],

[0069] M * [n] =M * [n-1];

[0070] F * L [n] =F * L [n-1]+F[n]-F[n-1];

[0071] When a[n] ≠a[n-1], calculate M according to the following formulas: * [n] and F * L [n],

[0072] M * [n] = (F[n]-F[n-1] ) / (a[n]-a[n-1]);

[0073] F * L [n]= (a[n]*F[n-1]-a[n-1]*F[n]) / ((a[n]-a[n-1])).

[0074] In step S2 of the present invention, the method for introducing the estimated value of the curb mass and the estimated value of the running resistance into the open-loop transfer function G(s) of the drive motor control loop is:

[0075] The speed regulator in the vehicle drive system adopts PI regulator, and the transfer function of PI regulator is (K p +K i / s), K p is the proportional gain, Ki is the integral coefficient, and the estimated curb mass value M is introduced * and vehicle unladen mass M min , the ratio of the two adjusts the output value of the PI regulator;

[0076] Conversion coefficient K between motor output torque and vehicle driving force T It is known that the estimated driving resistance value F is introduced * L , by K T After conversion, the resistance torque T is obtained * L ;

[0077] The driving motor system is given a torque command T G Increase resistance torque T * L ;

[0078] The driving motor system is given a torque command T GThe transfer function of the actual output torque T is 1 / (1+sT m ), T m is the motor torque response time constant;

[0079] The actual total mass of the vehicle is M;

[0080] The actual value of driving resistance is F L ;

[0081] The open-loop transfer function G(s)= (((K p +K i / s)*M * / M min +T * L ) / (1+sT m )*K T -F L ) / M / s.

[0082] In step S2 of the present invention, due to the motor torque response time constant T m is relatively small, and the open-loop transfer function G(s) is simplified to

[0083] G(s)≈(((K p +K i / s)*M * / M min + T * L )*K T -F L ) / M / s;

[0084] = (K p +K i / s)*K T / s / M min *M * / M+(F * L -F L ) / M / s;

[0085] When the estimated value of the curb weight is close to the actual value and the estimated value of the driving resistance is close to the actual value, G(s)≈(Kp+Ki / s)*K T / M min / s.

[0086] The present invention has the following beneficial effects: the curb mass and driving resistance of the tractor can be estimated online during vehicle driving, and the estimated curb mass and driving resistance of the tractor are introduced into the drive motor control circuit to reduce the impact of changes in curb mass and driving resistance. This ensures the consistency of acceleration characteristics under different tractor traction masses and different slopes. The curb mass and driving resistance involved are estimated only by the motor speed and motor output torque. The accuracy of the estimated curb mass and driving resistance values ​​can also be maintained on sloping roads without adding additional sensors. At the same time, changes in curb mass can be used to determine whether cargo has been left behind during driving, and corresponding reminders can be issued. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 It is a flow chart of online estimation of curb mass and driving resistance of the present invention.

[0088] Figure 2 It is an open-loop transfer function in which the control system of the present invention does not introduce the estimated value of the curb mass and the estimated value of the running resistance.

[0089] Figure 3 It is an open-loop transfer function in which the control system of the present invention introduces the estimated value of the curb mass and the estimated value of the running resistance. DETAILED DESCRIPTION

[0090] The present invention will be described below with reference to the accompanying drawings and embodiments.

[0091] A control method for a drive motor system of an electric tractor,

[0092] Step S1: Real-time acquisition of the driving motor speed ω m [n] and output torque T[n], calculate the vehicle acceleration a[n] and vehicle traction F[n] at time n, and calculate the estimated curb weight M*[n] and driving resistance F*[n] at time n based on the vehicle acceleration a[n] and vehicle traction F[n]. L [n];

[0093] Step S2: Introducing the estimated curb mass and driving resistance into the open-loop transfer function G(s) of the drive motor control loop. The tractor's curb mass and driving resistance are estimated online during driving, and these estimates are introduced into the drive motor control loop to mitigate the impact of changes in curb mass and driving resistance. This ensures consistent acceleration characteristics across varying tractor weights and slopes. The estimated curb mass and driving resistance are estimated solely from the motor speed and output torque, ensuring the accuracy of these estimates even on sloping roads without the need for additional sensors. Furthermore, changes in curb mass can be used to determine whether cargo has been left behind during driving, allowing for appropriate alerts.

[0094] As attached Figure 1 As shown, the specific method for estimating the curb weight and driving resistance in step S1 is:

[0095] Step S101: Start, execute S102;

[0096] Step S102: Parameter initialization, the first power-on initialization curb mass estimate M* is the vehicle unloaded mass M min , estimated driving resistance F* L is 0, that is:

[0097] F* L = 0;

[0098] M*=M min ;

[0099] Then execute S103;

[0100] Step S103: According to the sampling period T s Sampling, real-time sampling of the driving motor speed ω m [n] and output torque T[n], and calculate the vehicle acceleration a[n] and vehicle traction F[n] at time n;

[0101] a[n] = (ω m [n] - ω m [n-1])*K V / T S ;

[0102] F[n] = T[n]* K T ;

[0103] Among them, ω m [n] is the motor speed at time n, ω m [n-1] is the motor speed at time n-1, K V is the vehicle speed and motor speed coefficient, TS is the real-time sampling period, K T is the conversion coefficient between the motor output torque and the vehicle driving force;

[0104] If S102 enters S103, it is necessary to continuously sample multiple times and record the motor speed and motor output torque of each sample. In this embodiment, the sample is continuously sampled three times and the motor speed ω of each sample is recorded. m [0]、ω m [1] and ω m [2], the motor output torques T[0], T[1] and T[2];

[0105] Calculate vehicle acceleration:

[0106] a[1]=(ω m [1]-ω m [0])*K V / T s ;

[0107] a[2]=(ω m [2]-ω m [1])*K V / T s ;

[0108] Calculate vehicle traction:

[0109] F[0] = T[0]*K T ;

[0110] F[1] = T[1]*K T ;

[0111] F[2] = T[2]*K T ;

[0112] If S103 is entered from other links, only one sampling is required to record the motor speed and motor output torque;

[0113] Then execute S104;

[0114] Step S104: Determine whether the vehicle is stationary, if yes, execute S105, if no, execute S106;

[0115] Step S105: In the static state, ω m [n] = 0, estimated driving resistance value F* at time n L [n] is the traction force output by the motor system, and the estimated value of the curb mass is M*, that is:

[0116] F* L = F* L [n]=F[n]= T[n]* KT ;

[0117] M*[n] =M*;

[0118] Then execute S103;

[0119] Step S106: Based on Newton's second law, a vehicle dynamics model is constructed. The two sets of data calculated based on the latest sampling form the following equations:

[0120] F[n-1] -F L [n-1] = M[n-1]*a[n-1];

[0121] F[n] -F L [n] = M[n]*a[n];

[0122] Among them, F L [n] is the driving resistance at time n, M[n] is the curb weight at time n, F L [n-1] is the driving resistance at time n-1, M[n-1] is the curb weight at time n-1;

[0123] Let F* L [n]=F L [n-1] =F L [n],M * [n]=M[n-1]= M[n], the estimated value of curb mass M is obtained based on the calculated vehicle acceleration and vehicle traction * [n] and estimated driving resistance F* L [n];

[0124] After the calculation is completed, execute S107;

[0125] In step S106, the following processing methods are:

[0126] When F[n] = 0,

[0127] M * [n] =M * ;

[0128] F * L [n] = F[n]-M * [n] * a[n];

[0129] When a[n] = 0,

[0130] M * [n] =M * [n-1];

[0131] F *L [n] = F[n];

[0132] When a[n] ≠ 0 and a[n]=a[n-1], calculate M according to the following formulas: * [n] and F * L [n],

[0133] M * [n] =M * [n-1];

[0134] F * L [n] =F * L [n-1]+F[n]-F[n-1];

[0135] When a[n] ≠a[n-1], calculate M according to the following formulas: * [n] and F * L [n],

[0136] M * [n] = (F[n]-F[n-1] ) / (a[n]-a[n-1]);

[0137] F * L [n]= (a[n]*F[n-1]-a[n-1]*F[n]) / ((a[n]-a[n-1]));

[0138] Step S107: Determine the current M * [n] and M * Whether the difference exceeds the set threshold, if it exceeds the set threshold, the quality change flag S_M is set to 1 and S108 is executed; otherwise, the quality change flags S_M and S_M_P are set to 0, that is, S_M=0, S_M_P=0, and S111 is executed;

[0139] S_M indicates whether the estimated curb mass value has changed by more than a set threshold compared to the historical estimated curb mass value. S_M_P indicates whether the existing data after the curb mass change meets the estimated mass and resistance requirements. Because the estimation result will have a large error if the two sets of data before and after the mass change are used for estimation, continued sampling is required after the mass change to ensure estimation accuracy.

[0140] Step S108: Determine whether the flags S_M and S_M_P are both equal to 1. If yes, execute S109; if not, execute S110.

[0141] Step S109: Determine M *Does the difference between [n] and R_M exceed the set threshold? If yes, execute S103; if no, execute S111;

[0142] When M * When the difference between [n] and R_M exceeds the set threshold:

[0143] R_M=M * [n];

[0144] R_M records the current estimated instantaneous quality, which is mainly used in the next cycle and is equivalent to temporary storage of historical data;

[0145] When R_F * L With F * L When the difference of [n] is within the set threshold range, update F according to the following formula * L ,

[0146] F * L = K FIL * F * L [n]+ (1-K FIL )*F * L ;

[0147] R_F * L = F * L [n];

[0148] where K FIL is the filter coefficient, 0 <K FIL <1, R_F * L Record the currently estimated instantaneous driving resistance, which is mainly used for the next cycle and is equivalent to temporarily storing historical data;

[0149] Then execute S103;

[0150] Step S110: Record the current estimated curb mass and running resistance, and set the mass change flag S_M_P to 1, that is:

[0151] R_M=M * [n];

[0152] R_F * L = F * L [n];

[0153] S_M_P=1;

[0154] and execute S103;

[0155] Step S111: Correct M using the latest estimated curb mass and running resistance. * and F * L , the calculation formula is as follows:

[0156] F * L = K FIL * F * L [n]+ (1-K FIL )*F * L ;

[0157] M * = K FIL *M * [n] + (1-K FIL )*M * ;

[0158] And execute S103.

[0159] like Figure 2 As shown in the figure, without introducing the estimated value of curb weight and driving resistance, the open-loop transfer function of the drive motor control loop is G(s) = ((Kp+Ki / s) / (1+sT m )*K T -F L ) / M / s.

[0160] like Figure 3 As shown, in step S2, the method of introducing the estimated value of the curb mass and the estimated value of the driving resistance into the open-loop transfer function G(s) of the drive motor control loop is:

[0161] The speed regulator in the vehicle drive system adopts PI regulator, and the transfer function of PI regulator is (K p +K i / s), K p is the proportional gain, Ki is the integral coefficient, and the estimated curb mass value M is introduced * and vehicle unladen mass M min , the ratio of the two adjusts the output value of the PI regulator;

[0162] Conversion coefficient K between motor output torque and vehicle driving force T It is known that the estimated driving resistance value F is introduced * L , by K T After conversion, the resistance torque T is obtained *L =F * L / K T ;

[0163] The driving motor system is given a torque command T G Increase resistance torque T * L ;

[0164] The driving motor system is given a torque command T G The transfer function of the actual output torque T is 1 / (1+sT m ), T m is the motor torque response time constant;

[0165] The actual total mass of the vehicle is M;

[0166] The actual value of driving resistance is F L ;

[0167] The open-loop transfer function G(s)= (((K p +K i / s)*M * / M min +T * L ) / (1+sT m )*K T -F L ) / M / s.

[0168] Due to the motor torque response time constant T m is relatively small, and the open-loop transfer function G(s) is simplified to

[0169] G(s)≈(((K p +K i / s)*M * / M min + T * L )*K T -F L ) / M / s;

[0170] = (K p +K i / s)*K T / s / M min *M * / M+(F * L -F L ) / M / s;

[0171] When the estimated value of the curb weight is close to the actual value and the estimated value of the driving resistance is close to the actual value, G(s)≈(Kp+Ki / s)*KT / M min / s, at this time, G(s) is only related to Kp, Ki, K T and M min Related, M min is the unloaded mass of the vehicle. These variables are all constants. The transient and steady-state indicators of the entire drive motor control loop are basically not affected by changes in the curb mass and driving resistance.

[0172] In this embodiment, changes in the curb mass of the tractor generally occur in the following two situations: one is when the towed cargo is increased or decreased when the vehicle is parked; the other is when cargo is left behind during driving. Therefore, the curb mass of the tractor will not change frequently during driving, that is, the curb mass of the tractor will remain unchanged for a short period of time. Once the curb mass changes, the changed curb mass will also remain unchanged for a period of time.

[0173] In this embodiment, the tractor's driving resistance includes at least one of wind resistance, friction resistance, and gravity on a slope. The change in the tractor's driving resistance depends on changes in speed and slope. When the time interval between two adjacent motor speed measurements is short enough, it can be considered that the driving resistance corresponding to the two samples is approximately equal.

[0174] In order to improve the estimation accuracy in this embodiment, the sampling period of the motor speed and output torque can be shortened.

[0175] This embodiment is not limited to being applicable to the control of the drive motor system of an electric tractor, but is also applicable to the control of the drive motor system of a truck, a passenger bus, etc.

Claims

1. A method for controlling a drive motor system of an electric tractor, characterized in that: Step S1: Real-time acquisition of the driving motor speed ω m [n] and output torque T[n], calculate the vehicle acceleration a[n] and vehicle traction F[n] at time n, and calculate the estimated curb weight M*[n] and driving resistance F*[n] at time n based on the vehicle acceleration a[n] and vehicle traction F[n]. L [n]; Step S2: Introducing the estimated curb mass value and the estimated driving resistance value into the open-loop transfer function G(s) of the drive motor control loop; The specific method is: The speed regulator in the vehicle drive system adopts PI regulator, and the transfer function of PI regulator is (K p +K i / s), K p is the proportional gain, Ki is the integral coefficient, and the estimated curb mass value M is introduced * and vehicle unladen mass M min , the ratio of the two adjusts the output value of the PI regulator; Conversion coefficient K between motor output torque and vehicle driving force T It is known that the estimated driving resistance value F is introduced * L , by K T After conversion, the resistance torque T is obtained * L ; The driving motor system is given a torque command T G Increase resistance torque T * L ; The driving motor system is given a torque command T G The transfer function of the actual output torque T is 1 / (1+sT m ), T m is the motor torque response time constant; The actual total mass of the vehicle is M; The actual value of driving resistance is F L ; The open-loop transfer function G(s)= (((K p +K i / s)*M * / M min +T * L ) / (1+sT m )*K T -F L ) / M / s.

2. The electric tractor drive motor system control method according to claim 1, characterized in that: The calculation method of the vehicle acceleration a[n] and the vehicle traction force F[n] at time n in step S1 is: a[n] = (ω m [n] - oh m [n-1])*K V / T S ; F[n] = T[n] * K T ; Among them, ω m [n] is the motor speed at time n, ω m [n-1] is the motor speed at time n-1, K V is the vehicle speed and motor speed coefficient, T S is the real-time sampling period, K T It is the conversion coefficient between the motor output torque and the vehicle driving force.

3. The electric tractor drive motor system control method according to claim 1 or 2, characterized in that: In step S1, the initialization of the estimated curb mass M* is the vehicle's unladen mass M min , estimated driving resistance F* L is 0, that is: F* L = 0； M*=M min ; At rest, the estimated driving resistance F* at time n L [n] is the traction force output by the motor system, and the estimated value of the curb mass is M*, that is: F* L = F* L [n]=F[n]; M*[n] =M*.

4. The electric tractor drive motor system control method according to claim 3, characterized in that: In step S1, the estimated value of the curb weight M*[n] and the estimated value of the driving resistance F*[n] at time n are calculated based on the vehicle acceleration a[n] and the vehicle traction force F[n]. L The method of [n] is: F[n-1]-F L [n-1]=M[n-1]*a[n-1]; F[n]-F L [n] = M[n]*a[n]; Among them, F L [n] is the driving resistance at time n, M[n] is the curb weight at time n, F L [n-1] is the driving resistance at time n-1, M[n-1] is the curb weight at time n-1; Let F* L [n] =F L [n-1] =F L [n], M*[n] =M[n-1] = M[n], solve the estimated value of driving resistance F* at time n L [n] and the estimated curb mass M*[n].

5. The electric tractor drive motor system control method according to claim 1, 2 or 4, characterized in that: In step S1, the estimated value of running resistance F* is calculated. L [n] and the estimated value of curb mass M*[n], F* L [n] and M*[n] are low-pass filtered to obtain M* and F* L .

6. The electric tractor drive motor system control method according to claim 1, characterized in that: The specific method for estimating the curb weight and driving resistance in step S1 is: Step S101: Start, execute S102; Step S102: Parameter initialization, F * L = 0, M * =M min , then execute S103; Step S103: According to the sampling period T s Sampling, real-time sampling of the driving motor speed ω m [n] and output torque T[n], and calculate the vehicle acceleration a[n] and vehicle traction F[n] at time n; a[n] = (ω m [n] - oh m [n-1])*K V / T S ; F[n] = T[n] * K T ; Among them, ω m [n] is the motor speed at time n, ω m [n-1] is the motor speed at time n-1, K V is the vehicle speed and motor speed coefficient, T S is the real-time sampling period, K T is the conversion coefficient between the motor output torque and the vehicle driving force; Step S104: Determine whether the vehicle is stationary, if yes, execute S105, if no, execute S106; Step S105: ω m [n] = 0, record the estimated driving resistance F * L =T[n]* K T , then execute S103; Step S106: Based on Newton's second law, a vehicle dynamics model is constructed. The two sets of data calculated based on the latest sampling form the following equations: F[n-1] -F L [n-1] = M[n-1]*a[n-1]; F[n] -F L [n] = M[n]*a[n]; Let F* L [n]=F L [n-1] =F L [n],M * [n]=M[n-1] = M[n], the estimated value of curb mass M is obtained based on the calculated vehicle acceleration and vehicle traction * [n] and estimated driving resistance F* L [n]; After the calculation is completed, execute S107; Step S107: Determine the current M * [n] and M * Whether the difference exceeds the set threshold, if it exceeds the set threshold, the quality change flag S_M is set to 1 and S108 is executed; otherwise, the quality change flags S_M and S_M_P are set to 0, that is, S_M=0, S_M_P=0, and S111 is executed; S_M is a status indication that the estimated curb mass value exceeds the set threshold value compared with the historical estimated curb mass value, and S_M_P indicates whether the existing data after the curb mass change meets the estimated mass and resistance requirements; Step S108: Determine whether the flags S_M and S_M_P are both equal to 1. If yes, execute S109; if not, execute S110. Step S109: Determine M * Does the difference between [n] and R_M exceed the set threshold? If yes, execute S103; if no, execute S111; When M * When the difference between [n] and R_M exceeds the set threshold: R_M=M * [n]; R_M records the current estimated instantaneous quality; When R_F * L With F * L When the difference of [n] is within the set threshold range, update F according to the following formula * L , F * L = K FIL * F * L [n] + (1-K FIL )*F * L ; R_F * L = F * L [n]; where K FIL is the filter coefficient, 0 <K FIL <1, R_F * L Record the current estimated instantaneous driving resistance; Step S110: Record the current estimated curb mass and running resistance, and set the mass change flag S_M_P to 1, that is: R_M=M * [n]; R_F * L = F * L [n]; S_M_P=1; and execute S103; Step S111: Correct M using the latest estimated curb mass and running resistance. * and F * L , the calculation formula is as follows: F * L = K FIL * F * L [n] + (1-K FIL )*F * L ; M * = K FIL *M * [n] + (1-K FIL )*M * ; And execute S103.

7. The electric tractor drive motor system control method according to claim 6, characterized in that: If S102 enters S103, it is necessary to continuously sample multiple times and record the motor speed and motor output torque of each sample; If S203 is entered from other steps, sampling is only required once to record the motor speed and motor output torque.

8. The electric tractor drive motor system control method according to claim 6 or 7, characterized in that: In step S106, When F[n] = 0, M * [n] =M * ; F * L [n] = F[n] -M * [n] * a[n]; When a[n] = 0, M * [n] =M * [n-1]; F * L [n] = F[n]; When a[n] ≠ 0 and a[n] = a[n-1], calculate M according to the following formulas: * [n] and F * L [n], M * [n] =M * [n-1]; F * L [n] =F * L [n-1]+F[n]-F[n-1]; When a[n] ≠a[n-1], calculate M according to the following formulas: * [n] and F * L [n], M * [n] = (F[n] -F[n-1] ) / (a[n]-a[n-1]); F * L [n]= (a[n]*F[n-1]-a[n-1]*F[n]) / ((a[n]-a[n-1]))。 9. The method for controlling a drive motor system of an electric tractor according to claim 1, 2, 4, 6 or 7, characterized in that: In step S2, due to the motor torque response time constant T m is relatively small, and the open-loop transfer function G(s) is simplified to G(s)≈(((K p +K i / s)*M * / M min + T * L )*K T -F L ) / M / s; = (K p +K i / s)*K T / s / M min *M * / M+(F * L -F L ) / M / s; When the estimated value of the curb weight is close to the actual value and the estimated value of the driving resistance is close to the actual value, we can get G(s) ≈(Kp+Ki / s)*K T / M min / s.

Images