CN-121986649-A - Intelligent prediction system for threshing and separating quality of soybean combine harvester

CN121986649ACN 121986649 ACN121986649 ACN 121986649ACN-121986649-A

Abstract

The invention discloses an intelligent prediction system for threshing separation quality of a soybean combine harvester, which particularly relates to the technical field of intelligent prediction of separation quality, and is characterized in that an axial non-uniformity coefficient and a load peak value offset are extracted, a threshing strength axial difference factor is constructed by combining a drum rotating speed, an inertial interference quantity is constructed by coupling the threshing strength axial difference factor with a pitch angle speed, a low-frequency rigid body motion and a high-frequency threshing characteristic component are separated by utilizing a vibration signal time-frequency decoupling method, a time sequence prediction model is adopted to perform distributed prediction on the entrainment loss rate of each axial section, an on-line self-correction of the model is realized by matching a dynamic correction factor constructed based on actual measurement deviation with a history working condition fusion mechanism, and the drum rotating speed is regulated in a parallel operation mode by performing closed loop comparison on a comprehensive quality index and a preset threshold value, so that the self-adaptive control on threshing energy input is realized.

Inventors

KANG JIAXIN
CHEN GUOQIANG
CHEN LINGPING
LI YAN

Assignees

湖南工程学院

Dates

Publication Date: 20260508
Application Date: 20260410

Claims (10)

1. The intelligent prediction system for threshing and separating quality of the soybean combine harvester is characterized by comprising a data acquisition and description module, an axial difference calculation module, a disturbance decoupling analysis module, a loss distribution prediction module, a dynamic correction evaluation module and a closed-loop control output module; The data acquisition description module is used for acquiring a vehicle body attitude angle, an attitude angular speed, a threshing cylinder rotating speed, a cylinder torque and vibration acceleration signals acquired by a plurality of vibration measuring points axially arranged along the cylinder, and constructing axial load distribution characteristics of the cylinder through axial sampling to form a load attitude coupling space-time description set; The axial difference calculation module is used for extracting an axial unevenness coefficient and a load peak value offset according to the load attitude coupling space-time description set and calculating a threshing strength axial difference factor by combining the rotating speed of the roller; The disturbance decoupling analysis module is used for constructing the disturbance quantity of the inertia force on the grain sedimentation track according to the threshing strength axial difference factor and the vehicle body pitch angle speed, and decoupling the low-frequency rigid body motion component and the high-frequency threshing characteristic component by combining the time-frequency analysis result of the vibration signal to obtain an inertia threshing coupling disturbance coefficient; The loss distribution prediction module is used for predicting the entrainment loss rate along each axial section of the roller by adopting a time sequence prediction model according to the inertia threshing coupling disturbance coefficient and the real-time axial load distribution to obtain an entrainment loss rate distribution prediction value; The dynamic correction evaluation module is used for calculating a dynamic correction factor according to the entrainment loss rate distribution predicted value and the loss rate deviation value of the actual measurement value of the loss sensor at the outlet of the roller, and obtaining the overall threshing separation quality comprehensive index by fusing historical working condition data; And the closed-loop control output module is used for comparing the integral threshing and separating quality comprehensive index with a preset separating quality threshold value, and calculating the rotating speed regulating quantity of the threshing cylinder if the integral threshing and separating quality comprehensive index exceeds the preset separating quality threshold value.
2. The intelligent prediction system for threshing and separating quality of a soybean combine harvester according to claim 1, wherein the process for forming the load attitude coupling space-time description set is as follows: A multisource sensing unit is arranged in a threshing cylinder area and a vehicle body of the soybean combine harvester, and dynamic response parameters related to threshing separation quality are continuously collected; The dynamic response parameters comprise a vehicle body roll angle, a pitch angle, an attitude angular speed, a roller real-time rotating speed, a roller driving torque and a vibration acceleration signal; All the sensing units are connected to the vehicle-mounted data acquisition system and then Zhong Tongbu is carried out; setting a uniform sampling frequency for the acquired dynamic response parameters; during operation of the harvester in a fixed time window Sliding buffering of dynamic response parameters for each time instant Pre-processing dynamic response parameters of the test piece; Sequencing the pretreated axial vibration The rotation speed of the roller corresponding to the moment Drum driving torque Fusing to construct a roller axial load distribution characteristic vector: Wherein, the first Load characterization value of measuring point For a normalized combination of vibration effective value and torque, Is the total number of axial positions; Distributing the axial load of the roller into characteristic vectors And the load attitude coupling space-time description set is formed by jointly mapping with the roll angle, the pitch angle and the corresponding attitude angular speed.
3. The intelligent prediction system for threshing and separating quality of soybean combine harvester according to claim 2, wherein the logic for obtaining the axial non-uniformity coefficient is characterized by extracting the axial load distribution characteristic vector of the roller from the load attitude coupling space-time description set And calculate the standard deviation of the axial load distribution And mean value of And calculating the ratio of the standard deviation to the average value of the axial load distribution to obtain the axial non-uniformity coefficient.
4. The intelligent prediction system for threshing and separating quality of a soybean combine harvester according to claim 2, wherein the load peak offset is obtained by logic for obtaining a load distribution characteristic vector from the axial direction of a roller Obtaining a load representation value Is the maximum value of (2) and the load representation value And (3) carrying out difference operation on the maximum value and the average value of the load peak value offset.
5. The intelligent prediction system for threshing and separating quality of soybean combine harvester according to claim 4, wherein the logic for obtaining the axial difference factor of threshing strength is as follows, according to the axial unevenness coefficient And peak load offset Calculating threshing strength axial difference factor by combining rotating speed of roller The formula is as follows: Wherein As a factor of the axial non-uniformity, In order to achieve the peak load offset, For the rotational speed of the drum, Is the standard rotation speed.
6. The intelligent prediction system for threshing and separating quality of a soybean combine harvester according to claim 5, wherein the interference quantity of inertia force on grain sedimentation tracks is constructed according to the threshing strength axial difference factor and the pitch angle speed of a vehicle body, and the calculation formula is as follows: Wherein As the interference of inertial force to the grain sedimentation track, For the axial difference factor of threshing strength, Is the pitch angle rate.
7. The intelligent prediction system for threshing and separating quality of soybean combine harvester according to claim 6, wherein the process of obtaining the inertial threshing and coupling disturbance coefficient is as follows: for vibration acceleration signals Performing time-frequency analysis, and acquiring each moment by adopting short-time Fourier transform Spectral representation of a lower vibration acceleration signal ; Dividing the vibration acceleration signal into a low-frequency rigid motion component and a high-frequency threshing characteristic component according to a preset frequency range, wherein the specific separation rule is as follows: Wherein As a low-frequency component of the light, As the high-frequency component of the wave, And The frequency range values for the low frequency and the high frequency respectively, Is the maximum frequency of the vibration acceleration signal; The disturbance quantity, the low-frequency component and the high-frequency component of the grain sedimentation track caused by the inertia force are normalized, and then the inertia threshing coupling disturbance coefficient is calculated, and the calculation formula is as follows: Wherein Is the inertia threshing coupling disturbance coefficient, 、 The preset weight coefficients are respectively the interference quantity of the inertia force on the grain sedimentation track and the ratio of the low-frequency component to the high-frequency component.
8. The intelligent prediction system for threshing and separating quality of a soybean combine harvester according to claim 1, wherein the process of obtaining the comprehensive index of the overall threshing and separating quality is as follows: carrying out difference operation on the entrainment loss rate predicted value and the actual measured value of the loss sensor at the outlet of the roller to obtain a loss rate deviation value; and calculating a dynamic correction factor according to the loss rate deviation value, wherein the calculation formula is as follows: Wherein Is a dynamic correction factor; And the historical working condition data are fused to obtain the integral threshing separation quality comprehensive index.
9. The intelligent prediction system for threshing and separating quality of a soybean combine harvester according to claim 8, wherein the integrated threshing and separating quality comprehensive index is compared with a preset separating quality threshold value, and if the integrated threshing and separating quality comprehensive index is larger than the separating quality threshold value, the existence of overrun in separating quality is indicated; if the integrated threshing and separating quality comprehensive index is smaller than or equal to the separating quality threshold, the separating quality is not overrun, and the current state is maintained.
10. The intelligent prediction system for threshing and separating quality of a soybean combine harvester according to claim 9, wherein if the separating quality is out of limit, the calculation formula for calculating the rotational speed adjustment quantity of the threshing cylinder according to the difference between the integral threshing and separating quality comprehensive index and the separating quality threshold value is as follows: Wherein For the rotational speed adjustment of the threshing cylinder, For the integral threshing and separating quality comprehensive index, For a preset separation quality threshold value, Is a rotation speed adjustment coefficient.

Description

Intelligent prediction system for threshing and separating quality of soybean combine harvester Technical Field The invention relates to the technical field of intelligent prediction of separation quality, in particular to an intelligent prediction system of threshing and separating quality of a soybean combine harvester. Background In the modern agricultural intelligent process, the intelligent prediction technology of threshing and separating quality of a soybean combine harvester is regarded as a core means for realizing accurate damage-reducing harvesting. However, the complexity of the field operating environment presents a serious challenge to the reliability of the technology, especially in the case of attitude disturbance caused by severe terrain variations. When the harvester suddenly enters a deep ditch or spans a ridge from stable running, the physical field distribution of the threshing system can be changed instantaneously by the violent pitching or rolling of the vehicle body, and the dynamic process brings a deep dilemma of a physical layer to the existing intelligent prediction method. Threshing drums generally adopt a long shaft structure, and the separation performance of the threshing drums is highly dependent on the uniformity of the distribution of internal materials along the axial direction. When the fuselage is tilted, the crop grain mixture in the drum will accumulate rapidly toward the lower end under the force of gravity, resulting in instantaneous reconstruction of the load distribution. The physical phenomenon directly breaks the lumped parameter assumption commonly adopted by the existing prediction model, namely, the state of the material in the roller is considered to be homogeneous and stable. At the moment of inclination, the high end is in danger of rapid increase of the breakage rate caused by excessive striking due to rare materials, and the low end is in embarrassment of entrainment loss caused by incomplete threshing due to serious accumulation of materials, so that the traditional prediction model established based on the uniform field theory is completely ineffective. Further complicating this is that the axial inertial forces generated by the severe pitching action can further interfere with the dynamics of the separation process. The inertia force changes the movement track of grains penetrating through the straw layer to sink towards the concave plate, normal separation can be completed by gravity and high-frequency vibration, and the grains can be pressed back to the straw layer again due to inertia impact and are discharged along with the straw discharging port, so that unexpected recessive loss is caused. However, vibration sensors configured by current mainstream prediction systems are mostly designed for high-frequency threshing characteristics, and it is difficult to accurately capture and decouple such low-frequency, large-scale rigid motion disturbances. Therefore, in the case of abrupt change of terrain, how to break through the limitation of the lumped parameter model and establish an intelligent perception system capable of simultaneously identifying the coupling effect of the gesture disturbance and the separation mechanism becomes a key technical problem to be solved in order to improve the self-adaptive control precision of the soybean combine harvester. Disclosure of Invention In order to overcome the above-mentioned drawbacks of the prior art, embodiments of the present invention provide an intelligent prediction system for threshing and separating quality of a soybean combine harvester, so as to solve the problems set forth in the background art. In order to achieve the above purpose, the present invention provides the following technical solutions: the intelligent prediction system for threshing and separating quality of the soybean combine harvester comprises a data acquisition and description module, an axial difference calculation module, a disturbance decoupling analysis module, a loss distribution prediction module, a dynamic correction evaluation module and a closed-loop control output module; The data acquisition description module is used for acquiring a vehicle body attitude angle, an attitude angular speed, a threshing cylinder rotating speed, a cylinder torque and vibration acceleration signals acquired by a plurality of vibration measuring points axially arranged along the cylinder, and constructing axial load distribution characteristics of the cylinder through axial sampling to form a load attitude coupling space-time description set; The axial difference calculation module is used for extracting an axial unevenness coefficient and a load peak value offset according to the load attitude coupling space-time description set and calculating a threshing strength axial difference factor by combining the rotating speed of the roller; The disturbance decoupling analysis module is used for constructing the disturbance quantity of the inerti