A rice harvester electric header control method and controller based on feed load estimation

Abstract

The present invention discloses a control method and controller for an electric harvester header of a rice harvester based on feed load estimation. The method comprises the following steps: 1: constructing a header dynamics model based on header inertia, transmission ratio, and other parameters; 2: collecting harvester travel speed, drive motor torque, and speed information; 3: establishing a fitting formula for the current vehicle speed and header drive torque; 4: calculating the root mean square value of the feed load torque disturbance during a sampling period using the filtering result combined with the header dynamics model; 5: updating the fuzzy PID control domain based on the root mean square value; and 6: performing feedback control of the header using the updated domain combined with the fuzzy inference results. The present invention addresses the need for tracking and controlling the speed of a harvester's electric harvester header under complex load fluctuations. The method uses the root mean square value of historical external torque fluctuations as an indicator to perform real-time corrections to the domain of the fuzzy PID controller. This method provides improved robustness in the face of load fluctuations under different conditions, thereby providing a method for controlling the speed of a harvester header.

Description

Technical Field

[0001] The invention discloses a control method and a controller for an electric harvester of a rice harvester based on feeding load estimation, and belongs to the field of intelligent control of agricultural machinery. Background Art

[0002] In recent years, agricultural mechanization has developed rapidly. Currently, mechanized rice harvesting has reached over 90%, and the market share of rice combine harvesters is approximately 2 million. This rapid increase in rice harvesting has also resulted in significant energy (diesel) consumption and environmental pollution. Consequently, electric rice harvesters have become a research focus in the rice harvesting field. Due to the complex and changing harvesting environment, influenced by factors such as operating mode and crop characteristics, the load on the harvester platform of a rice combine harvester fluctuates significantly during operation. Existing methods for controlling the motor speed of the harvester platform are unable to meet the harvesting requirements in various environments, resulting in frequent stalls and missed harvests. This poses a significant challenge to the drive control of the electric harvester platform of a rice harvester. Summary of the Invention

[0003] This paper proposes a control method and controller for an electric harvester's header based on feed load estimation. This method collects real-time data on the header motor's drive shaft load during operation, combines it with a header dynamics model to estimate load fluctuations within the sampling period, and optimizes header drive force control based on this information, improving the speed stability of the electric header under load fluctuations. The specific solution is as follows:

[0004] A method for controlling an electric header of a rice harvester based on load estimation comprises the following steps:

[0005] Step 1: Establish a parameter-based header dynamic model;

[0006] Step 2: Obtain the harvester's driving speed information, header speed information, and header torque information through sensors;

[0007] Step 3: Linearly fit the header torque and driving speed information;

[0008] Step 4: Calculate the external load torque fluctuation information and the torque root mean square value based on the fitting results and the header dynamics model;

[0009] Step 5: Construct a scaling factor based on the RMS value and update the domain;

[0010] Step 6: Use the expansion factor combined with the fuzzy reasoning results to control the header motor and track the header speed demand.

[0011] Furthermore, the header dynamics model established in step 1 is as follows:

[0012]

[0013] M q =(k c +k p +k m )*q

[0014] Among them, M G is the motor input torque; M pi M is the friction torque of the reel wheel during idling; ci M is the friction torque of the cutter during idling operation; mi is the idling friction torque of the cutting platform auger; i p 、i m are the transmission ratios from the driving shaft to the reel and the auger respectively; J p is the moment of inertia of the reel; ω ci is the input shaft speed; J m M is the moment of inertia of the auger; q Feed load torque; q is the feed amount of the header; k c 、k p 、k m are the feed load factors of the cutter, reel and auger respectively.

[0015] Furthermore, in step 2, the harvester driving speed information is collected by the driving shaft speed sensor of the driving component, and the calculation method is shown in the following formula:

[0016] v combine =ω drive_shaft *i shaft_tire *r tire

[0017] Among them, v combine Harvester travel speed; ω drive_shaft is the speed of the driving shaft of the walking component; i shaft_tire is the transmission ratio of the gearbox between the driving shaft and the driving wheel of the traveling component; r tire is the driving wheel radius.

[0018] Furthermore, the torque information and speed information of the harvester header in step 2 are collected by a torque sensor and a speed sensor installed on the header drive shaft;

[0019] Furthermore, in step 3, the minimum mean square (RMS) is used as the evaluation function to perform linear regression fitting on the torque information and driving speed information of the header:

[0020]

[0021]

[0022] Mq_v =k·v combine +b

[0023] Where k and b are linear regression fitting coefficients; v combinei is the harvester speed at the corresponding sampling time point; M qi is the input load torque at the corresponding sampling time node; is the average value of the feed load torque; is the average speed of the harvester at the sampling node; M q_v The estimated feed load value calculated by the fitting formula.

[0024] Furthermore, the specific method for calculating the feed torque disturbance and its root mean square value based on the fitting results in step 4 is:

[0025]

[0026]

[0027] in, is the mean value of the input load torque disturbance value, m represents the number of sampling points, M di Represents the torque value corresponding to the first i sampling points, R d is the RMS value of the input load torque disturbance.

[0028] Furthermore, the specific method for generating the domain based on the perturbation root mean square value in step 5 is:

[0029] The basic domain is generated as follows:

[0030]

[0031]

[0032] The discrete domain is generated as follows:

[0033]

[0034]

[0035] Among them, I p , I i , I d The domain of proportion, differentiation and integration; I e are the domains of the error in the feedback link and the differential value of the error; T pm 、T im 、T dm are the basic domain boundary values ​​of proportion, differentiation and integration respectively; kr p 、kr i、kr d are the domain expansion correction coefficients of the proportional, differential and integral links respectively; kr e and are the universe expansion correction coefficients of the feedback error and the differential value of the error respectively; are the discrete domain values ​​corresponding to the corresponding links.

[0036] Furthermore, the specific method of fuzzy reasoning for designing the fuzzy PID controller in step 6 is:

[0037] The fuzzy domain is converted into linguistic variables through the membership function. The membership function uses trigonometric functions. The fuzzy rule table used in the derivation process is as follows:

[0038]

[0039] Among them, NB, NM, NS, ZO, PS, PM, and PB represent the fuzzy subsets of negative large, negative medium, negative small, zero, positive small, positive medium, and positive large, respectively.

[0040] Furthermore, the specific method for feedback tracking control of the header speed in step 6 is:

[0041]

[0042] k p =k p0 +Δk p

[0043] k i =k i0 +Δk i

[0044] k d =k d0 +Δk d

[0045] Among them, k p 、k i 、k d are proportional, differential and integral coefficients; e(t i ) is the error between the desired speed and the current speed of the header; k p0 、k i0 、k d0 The proportional, differential, and integral coefficient values ​​at the previous sampling time point; Δk p , Δk i and Δk d The change values ​​of the proportional, differential and integral coefficients respectively, which are generated by fuzzy rules.

[0046] The present invention also proposes a controller capable of executing the contents of steps 1, 3, 4, 5, and 6.

[0047] Beneficial effects of the present invention:

[0048] (1) Based on historical load information, the effect of speed fluctuation on header load is separated by linear fitting, providing more realistic feedback on the current header torque fluctuation characteristics;

[0049] (2) The torque fluctuation is calculated based on the dynamic characteristics of the header, which can meet the needs of headers with different parameters;

[0050] (3) The expansion coefficient of the fuzzy PID controller is adjusted based on the root mean square value of the load fluctuation, and tracking control is performed so that the control system can adapt to different levels of external torque fluctuations. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a flow chart of the method of the present invention; DETAILED DESCRIPTION

[0052] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0053] like Figure 1 As shown, the implementation of the present invention includes the following steps:

[0054] Step 1: Establish a parameter-based header dynamic model. The modeling method is as follows:

[0055]

[0056] M q =(k c +k p +k m )*q

[0057] Among them, M G is the motor input torque, acquired by the sensor in step 2; M pi is the friction torque of the reel when idling, which is determined by the parameters of the reel itself; M ci M is the friction torque of the cutter when idling, which is determined by the cutter parameters; mi is the idling friction torque of the cutting platform auger, which is determined by the parameters of the cutting platform auger; i p 、i m are the transmission ratios from the driving shaft to the reel and the auger, and are the design parameters of the header; J p is the moment of inertia of the reel, which is determined by the reel parameters; ω ci is the input shaft speed, acquired by the sensor in step 2; J m M is the moment of inertia of the auger, which is determined by the parameters of the auger itself; qFeed load torque is determined by external feed fluctuations; q is the header feed amount, which is determined by the driving speed and external load fluctuations; k c 、k p 、k m are the feed load factors of the cutter, reel and auger, respectively, and are design experience factors.

[0058] Step 2: Use sensors to obtain harvesting speed information, header speed information, and header torque information. The harvester speed information is collected by the driving shaft speed sensor of the driving component. The calculation method is shown in the following formula:

[0059] v combine =ω drive_shaft *i shaft_tire *r tire

[0060] Among them, v combine Harvester travel speed; ω drive_shaft is the speed of the driving shaft of the walking component; i shaft_tire is the transmission ratio of the gearbox between the driving shaft and the driving wheel of the traveling component; r tire is the driving wheel radius.

[0061] The torque information of the harvester header is collected through the torque sensor and speed sensor installed on the header drive shaft.

[0062] Step 3: Using the data collected in step 2, perform linear regression fitting on the header's torque information and driving speed information using the minimum mean square root as the loss evaluation function:

[0063]

[0064]

[0065] M q_v =k·v combine +b

[0066] Where k and b are linear regression fitting coefficients; v combinei is the harvester speed at the corresponding sampling time point; M qi is the input load torque at the corresponding sampling time node; is the average value of the feed load torque; is the average speed of the harvester at the sampling node; M q_v The estimated feed load value calculated by the fitting formula.

[0067] Step 4: Calculate the external load torque fluctuation information and the torque RMS value based on the fitting results and the header dynamics model:

[0068]

[0069]

[0070] Among them, M d is the input load torque disturbance value, R d is the RMS value of the input load torque disturbance.

[0071] Furthermore, the specific method for generating the domain based on the perturbation root mean square value in step 5 is:

[0072]

[0073] Among them, I p , I i , I d The domain of proportion, differentiation and integration; T pm 、T im 、T dm are the basic domain boundary values ​​of proportion, differentiation and integration respectively; kr p 、kr i 、kr d They are the domain expansion correction coefficients of the proportional, differential and integral links respectively.

[0074] Furthermore, the specific method of fuzzy reasoning in designing the fuzzy PID controller in step 6 is:

[0075] The fuzzy domain is converted into linguistic variables through the membership function. The membership function uses trigonometric functions. The fuzzy rule table used in the derivation process is as follows:

[0076]

[0077] Furthermore, the specific method for feedback tracking control of the header speed in step 6 is:

[0078]

[0079] k p =k p0 +Δk p

[0080] k i =k i0 +Δk i

[0081] k d =k d0 +Δk d

[0082] Among them, k p 、k i 、kd are proportional, differential and integral coefficients; e(t i ) is the error between the desired speed and the current speed of the header; k p0 、k i0 、k d0 The proportional, differential, and integral coefficient values ​​at the previous sampling time point; Δk p , Δk i and Δk d The change values ​​of the proportional, differential and integral coefficients respectively, which are generated by fuzzy rules.

[0083] The embodiment of the present invention also includes a controller, which can execute the contents of steps 1, 3, 4, 5, and 6. The physical device of the controller can be installed in the harvester cab.

[0084] The series of detailed descriptions listed above are only specific descriptions of feasible implementation methods of the present invention. They are not intended to limit the scope of protection of the present invention. Any equivalent methods or changes that do not deviate from the technology of the present invention should be included in the scope of protection of the present invention.

Claims

1. A rice harvester electric header control method based on feed load estimation, characterized in that: include: Step 1: Establish a parameter-based header dynamic model; Step 2: Obtain the harvester's driving speed information, header speed information, and header torque information through sensors; Step 3: Use historical data to perform linear regression analysis on the header torque and travel speed information; Step 4: Calculate the external load torque fluctuation information and the torque root mean square value based on the linear regression results and the header dynamics model; Step 5: Construct the domain of the scaling factor generation based on the RMS value; Step 6: Use the expansion factor combined with fuzzy reasoning rules to control the header motor and track the header speed demand.

2. A rice harvester electric header control method based on feed load estimation according to claim 1, characterized in that: The header dynamics model established in step 1 is: M q =(k c +k p +k m )*q Among them, M G is the motor input torque; M pi M is the friction torque of the reel wheel during idling; ci M is the friction torque of the cutter during idling operation; mi is the idling friction torque of the cutting platform auger; i p 、i m are the transmission ratios from the driving shaft to the reel and the auger respectively; J p is the moment of inertia of the reel; ω ci is the input shaft speed; J m M is the moment of inertia of the auger; q Feed load torque; q is the feed amount of the header; k c 、k p 、k m are the feed load factors of the cutter, reel and auger respectively.

3. The method for controlling an electric header of a rice harvester based on feeding load estimation according to claim 1, wherein: In step 2, the harvester driving speed information is collected by the driving shaft speed sensor of the driving component, and the calculation method is shown in the following formula: v combine =ω drive_shaft *i shaft_tire *r tire Among them, v combine Harvester travel speed; ω drive_shaft is the speed of the driving shaft of the walking component; i shaft_tire is the transmission ratio of the gearbox between the driving shaft and the driving wheel of the traveling component; r tire is the driving wheel radius.

4. The method for controlling an electric header of a rice harvester based on feeding load estimation according to claim 1, wherein: In step 2, the torque information and speed information of the harvester header are collected by a torque sensor and a speed sensor installed on the header drive shaft.

5. The method for controlling an electric header of a rice harvester based on feeding load estimation according to claim 1, characterized in that: In step 3, linear regression is used to fit the torque information and driving speed information of the header using the minimum mean square root as the loss evaluation function: M q_v =k·v combine +b Where k and b are linear regression fitting coefficients; v combinei is the harvester speed at the corresponding sampling time point; M qi is the input load torque at the corresponding sampling time node; is the average value of the feed load torque; is the average speed of the harvester at the sampling node; M q_v The estimated feed load value calculated by the fitting formula.

6. The method for controlling an electric header of a rice harvester based on feeding load estimation according to claim 1, characterized in that: The specific calculation method of step 4 is as follows: Among them, M d is the input load torque disturbance value, R d is the RMS value of the input load torque disturbance.

7. The method for controlling an electric header of a rice harvester based on feeding load estimation according to claim 1, characterized in that: The domain of the step 5 includes the basic domain and the discrete domain. The basic domain is generated as follows: The method for generating discrete domains is as follows: Among them, I p , I i , I d The domain of proportion, differentiation and integration; I e are the domains of the error in the feedback link and the differential value of the error; T pm 、T im 、T dm are the basic domain boundary values ​​of proportion, differentiation and integration respectively; kr p 、kr i 、kr d are the domain expansion correction coefficients of the proportional, differential and integral links respectively; kr e and are the universe expansion correction coefficients of the feedback error and the differential value of the error respectively; are the discrete domain values ​​corresponding to the corresponding links.

8. The method for controlling an electric header of a rice harvester based on feeding load estimation according to claim 5, characterized in that: The fuzzy inference rules in step 6 are designed as follows 9. The method for controlling an electric header of a rice harvester based on feeding load estimation according to claim 8, characterized in that: The specific method for feedback tracking control of the header speed in step 6 is: k p =k p0 +△k p k i =k i0 +△k i k d =k d0 +△k d Among them, k p 、k i 、k d are proportional, differential and integral coefficients; e(t i ) is the error between the desired speed and the current speed of the header; k p0 、k i0 、k d0 The proportional, differential, and integral coefficient values ​​at the previous sampling time point; △k p , △k i and △k d The change values ​​of the proportional, differential and integral coefficients respectively, which are generated by fuzzy rules.

10. A controller, characterized in that: The controller can execute the contents of steps 1, 3, 4, 5, and 6 described in any one of claims 1 to 9.

Images