CN110942146B - Method and device for measuring feeding quantity of self-propelled silage harvester - Google Patents
Method and device for measuring feeding quantity of self-propelled silage harvester Download PDFInfo
- Publication number
- CN110942146B CN110942146B CN201910485662.5A CN201910485662A CN110942146B CN 110942146 B CN110942146 B CN 110942146B CN 201910485662 A CN201910485662 A CN 201910485662A CN 110942146 B CN110942146 B CN 110942146B
- Authority
- CN
- China
- Prior art keywords
- feeding
- self
- measuring
- feeding roller
- silage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01D—HARVESTING; MOWING
- A01D45/00—Harvesting of standing crops
- A01D45/02—Harvesting of standing crops of maize, i.e. kernel harvesting
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P60/00—Technologies relating to agriculture, livestock or agroalimentary industries
- Y02P60/80—Food processing, e.g. use of renewable energies or variable speed drives in handling, conveying or stacking
- Y02P60/87—Re-use of by-products of food processing for fodder production
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Environmental Sciences (AREA)
- Feeding And Watering For Cattle Raising And Animal Husbandry (AREA)
Abstract
A method and a device for measuring the feeding quantity of a self-propelled silage harvester, the method comprises the steps of acquiring feeding roller displacement information of the self-propelled silage harvester in real time, and transmitting the feeding roller displacement information to a vehicle-mounted computer; acquiring the rotating speed information of a feeding roller of the harvester in real time, and transmitting the rotating speed information of the feeding roller to the vehicle-mounted computer; acquiring the water content information of the harvested silage in real time, and transmitting the water content information of the silage to the vehicle-mounted computer; acquiring the advancing walking speed information of the harvester in real time, and transmitting the walking speed information to the vehicle-mounted computer; and carrying out data processing on the input feeding roller displacement information, feeding roller rotating speed information, silage water content information and walking speed information by the vehicle-mounted computer by adopting a feeding amount neural network to obtain the current feeding amount of the harvester. The invention also provides a measuring device adopting the measuring method.
Description
Technical Field
The invention relates to a precise agricultural machinery equipment detection and measurement technology, in particular to a method and a device for measuring the feeding quantity of a self-propelled silage harvester.
Background
The feeding amount is critical to the working quality of the self-propelled silage harvester and the drawing of silage yield maps. In the harvesting season, on one hand, the feeding component of the self-propelled silage harvester is frequently blocked due to the work of the overfeeding quantity, the operation quality is reduced, the yield of the cutting length is insufficient, and the fermentation of silage is affected; on the other hand, the operation performance of the corn green feeding harvester is reduced due to the fact that the rated feeding quantity is not achieved, and the working efficiency is low. In addition, in the large background of precise agriculture, in order to construct an accurate corn silage yield map, measurement of feeding amount is also necessary. The foreign research on the feeding quantity of the self-propelled silage harvester starts earlier. Related researchers have proposed that a certain result is obtained by measuring the torque of the driving shaft of the blower blowing the material and the driving shaft of the basic working member, by means of a radiation measuring device and a capacitive oscillation sensor mounted at the outlet of the throwing barrel, by means of a momentum sensor mounted on the wall of the throwing barrel, by means of a change in momentum of a springboard at the outlet of the throwing barrel to reflect the feeding amount, and by measuring the feeding amount by means of the opening between the feeding rollers in the feeding device. Domestic research on a corn silage harvester is also focused on the mechanical structure design of the corn silage harvester, and the feeding amount of the corn silage harvester is not systematically analyzed, and basically still belongs to a test research stage and cannot reach the quantitative degree due to the influence of various factors.
The method of installing the related sensor at the outlet of the material throwing barrel or on the barrel wall to measure the feeding quantity is convenient to install and accurate to measure, but the measured value can only be used for drawing a silage yield graph, the working state of the feeding device of the self-propelled silage harvester is difficult to express, and the impending blockage of the feeding device during the working of the self-propelled silage harvester cannot be predicted. The method of measuring the feeding amount by measuring the feeding roller displacement can be exactly combined with the method of measuring the feeding amount. When the silage harvester is operated, the main factors influencing the feeding amount are the field crop density (kg/m) 3 ) The larger the field crop density is, the wider the cutting width is, the higher the harvester walking speed is, the larger the feeding amount is, and the larger the feeding roller displacement is. The density of the field crops is limited by the moisture content of the crops, the harvesting time is different, the moisture content of the crops is also changed, and the higher the moisture content of the crops is, the higher the density of the field crops is under the condition that the number of the plants of the unit area is the same. In the prior art, the feeding quantity is calculated according to the calculated feeding roller displacement by using the following formula:
m＝δων fr cη nom
wherein: m is the feeding quantity (kg/s), delta is the feeding roller displacement value (m), omega is the feeding roller width (m), and v fr For the feed roll speed (m/s), η nom The compaction density of the material entering the feeding device after being extruded by the feeding roller is influenced by the characteristics of the silage material, and c is the compression coefficient.
As can be seen from the above: when the type of the silage harvester is determined and the characteristics of silage materials are determined, the feeding roller displacement and the feeding amount have a one-to-one correspondence, and the theoretical equation can better reflect the feeding amount. But eta nom The determination of the size is influenced by various factors, and different materials have different eta nom It is difficult to determine experimentally, which limits the practical application of using feed roll displacement to determine feed.
Disclosure of Invention
The invention provides a self-propelled silage harvester feeding amount measuring method and a measuring device aiming at the defects of the existing silage harvester feeding amount testing technology.
In order to achieve the above purpose, the invention provides a method for measuring the feeding amount of a self-propelled silage harvester, which comprises the following steps:
s100, measuring the displacement of a feeding roller, acquiring feeding roller displacement information of a self-propelled silage harvester in real time, and transmitting the feeding roller displacement information to a vehicle-mounted computer;
s200, measuring the rotation speed of a feeding roller, acquiring the rotation speed information of the feeding roller of the self-propelled silage harvester in real time, and transmitting the rotation speed information of the feeding roller to the vehicle-mounted computer;
s300, measuring the silage water content, acquiring the harvested silage water content information in real time, and transmitting the silage water content information to the vehicle-mounted computer;
s400, measuring the walking speed of the harvester, acquiring the walking speed information of the self-propelled silage harvester in real time, and transmitting the walking speed information to the vehicle-mounted computer; and
s500, acquiring feeding quantity, and carrying out data processing on the input feeding roller displacement information, feeding roller rotating speed information, silage water content information and walking speed information by using the vehicle-mounted computer by adopting a feeding quantity neural network model to obtain the current feeding quantity of the self-propelled silage harvester.
The feeding amount measuring method of the self-propelled silage harvester, wherein the feeding amount neural network is a BP neural network and comprises an input layer, an output layer and an hidden layer, the input layer comprises three neurons, and the three neurons are respectively a feeding roller displacement delta, a feeding roller rotating speed v and a silage water content omicron, and the feeding amount neural network is recorded as (P1) T= [ delta v omicron]The method comprises the steps of carrying out a first treatment on the surface of the The hidden layer includes a plurality of neurons; the output layer comprises a neuron, and the input of the output layer is the output a of the hidden layer 1 The output of the output layer is real-time feeding quantity a 2 The connection weight matrix between each neuron of the input layer and the hidden layer is IW 1,1 The method comprises the steps of carrying out a first treatment on the surface of the The neuron bias value of the hidden layer is b 1 The method comprises the steps of carrying out a first treatment on the surface of the The number of neurons of the hidden layer is S; neurons of the output layer and a 1 The connection weight matrix between the two is LW 2,1 The neuron bias value of the output layer is b 2 The method comprises the steps of carrying out a first treatment on the surface of the The transfer functions of the hidden layer and the output layer are f respectively 1 、f 2 ，n 1 ，n 2 Transfer functions f of hidden layer and output layer respectively 1 、f 2 Is a net input to the computer.
According to the feeding quantity measuring method of the self-propelled silage harvester, the number S of neurons in the hidden layer is determined by simulation experiment results, and S is more than or equal to 5.
The feeding amount measuring method of the self-propelled silage harvester, wherein b 1 1, b 2 1.
The method for measuring the feeding amount of the self-propelled silage harvester, wherein the transfer function f of the hidden layer and the output layer 1 、f 2 Are hyperbolic tangent transfer functions.
The feeding amount measuring method of the self-propelled silage harvester comprises the steps that the input of the output layer is a 1 ＝tagsig(IW 1,1 +b 1 ) The method comprises the steps of carrying out a first treatment on the surface of the The output of the output layer is a 2 ＝purelin(LW 2,1 +b 2 )。
The method for measuring the feeding amount of the self-propelled silage harvester comprises the steps of connecting weights IW between neurons of the input layer and neurons of the hidden layer 1,1 For determination through network training.
The method for measuring the feeding amount of the self-propelled silage harvester, wherein the connection weight IW 1,1 Normalizing the sample data to interval [ -1,1 ] by adopting an advanced network training stopping method]The N groups of sample data are sequenced according to the feeding quantity, three parts of sample data are extracted at equal intervals to be respectively used as training samples, verification samples and test samples, and the training samples are used for calculating gradient values of a network model and updating weights IW of the network model 1,1 And LW (LW) 2,1 The method comprises the steps of carrying out a first treatment on the surface of the The verification sample is used for stopping training, and training is stopped when errors on the verification set rise in several iterations, and the weight and the bias value which generate extremely small errors on the verification set are used as the final trained network connection weight and the neuron bias value; the test samples are used to compare different models and data packets tested.
The method for measuring the feeding amount of the self-propelled silage harvester can further comprise the following steps:
s600, controlling the advancing speed of the self-propelled silage harvester according to the feeding quantity displayed in real time by the vehicle-mounted computer so as to obtain the optimal feeding quantity.
In order to better achieve the above purpose, the invention also provides a feeding amount measuring device of a self-propelled silage harvester, wherein the feeding amount measuring method of the self-propelled silage harvester is adopted for feeding amount detection, and the feeding amount measuring device comprises the following steps:
the feeding roller displacement measuring mechanism is used for measuring feeding roller displacement information of the self-propelled silage harvester in real time and comprises a connecting spring and a linear displacement sensor, wherein the connecting spring is sleeved on the linear displacement sensor, one end of the connecting spring is connected with one end of the linear displacement sensor and is connected with a rotating shaft head of a front feeding floating roller of the self-propelled silage harvester, and the other end of the connecting spring is connected with the other end of the linear displacement sensor and is connected with a rotating shaft head of a front feeding fixed roller of the self-propelled silage harvester;
the feeding roller rotating speed measuring mechanism is used for acquiring rotating speed information of the feeding roller in real time and comprises a rotating speed measuring encoder and a coupler, wherein one end of the rotating speed measuring encoder is connected with a feeding roller shaft of the self-propelled silage harvester through the coupler;
the silage moisture measuring mechanism is used for acquiring the silage moisture content information in real time and is arranged at the front part of the cooperative transportation vehicle of the self-propelled silage harvester;
the walking speed measuring mechanism is used for acquiring walking speed information of the self-propelled silage harvester in real time, and comprises a proximity switch and a speed measuring gear, wherein the speed measuring gear is arranged on a walking driving shaft of the self-propelled silage harvester, and the proximity switch is arranged corresponding to the speed measuring gear; and
the vehicle-mounted computer is respectively connected with the feeding roller displacement measuring mechanism, the feeding roller rotating speed measuring mechanism, the silage moisture measuring mechanism and the walking speed measuring mechanism, receives feeding roller displacement information, feeding roller rotating speed information, silage moisture content information and walking speed information, performs data fusion through a feeding amount neural network model, takes the feeding roller displacement information, the feeding roller rotating speed information and the silage moisture content information as network input, takes real-time feeding amount as network output, establishes a nonlinear mapping relation between input and output to reflect the real-time feeding amount, and adjusts the real-time feeding amount by controlling the walking speed of the self-propelled silage harvester.
The invention has the technical effects that:
the measuring device is convenient to install, and the vehicle-mounted computer can obtain a plurality of parameters at the same time, so that a driver can know the current feeding amount, the running speed and the feed water content in real time, and the running speed is timely adjusted according to the change of the feeding amount, so that the harvester works in a stable and efficient state, the working efficiency is improved, and the cutting quality of the feed is ensured. In practice, the change of the humidity of the material can influence the friction coefficient, the density and the compression coefficient, thereby influencing the compaction density of the material extruded by the feeding roller, so the detection method of the invention selects the water content of the material to be used as the influence eta nom The main factors of the device are feeding roller displacement delta, feeding roller rotating speed v and material water content o, and the predicted value of feeding quantity is automatically given through the trained artificial neural network.
The invention utilizes the artificial neural network to obtain the real-time feeding quantity, breaks through the limitation that the displacement of the feeding roller is converted into the feeding quantity by utilizing the traditional formula in the prior art, takes the moisture content of silage into consideration, so that the mode of converting the displacement of the feeding roller into the feeding quantity can be applied to various silage harvests, the measurement result is more accurate, and the feeding quantity measurement method can be suitable for field operation environments with variable moisture content.
The invention will now be described in more detail with reference to the drawings and specific examples, which are not intended to limit the invention thereto.
Drawings
FIG. 1 is a schematic view of an installation position of a measuring device according to an embodiment of the invention;
FIG. 2 is a schematic view of a feed roll displacement measuring mechanism according to an embodiment of the present invention;
FIG. 3 is a schematic view of a feed roller rotational speed measurement mechanism according to an embodiment of the present invention;
FIG. 4 is a schematic view of a walking speed measuring mechanism according to an embodiment of the present invention;
FIG. 5 is a right side view of FIG. 4;
FIG. 6 is a schematic diagram of a neural network model of feeding according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of the operation of a measuring device according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of a neural network model of feed rate according to an embodiment of the present invention.
Wherein reference numerals are used to refer to
1 feeding roller displacement measuring mechanism
11 front feeding floating roller
12 front feeding fixed roller
13 linear displacement sensor
14 connection spring
2 feeding roller rotating speed measuring mechanism
21-speed measuring encoder
22 coupling
23 feeding roller shaft
24 feed roller
3 silage moisture measuring mechanism
31 microwave moisture measuring sensor
4 walking speed measuring mechanism
41 proximity switch
42 speed measuring gear
43 brake pad
44 walking driving axle
5 vehicle-mounted computer
6 cooperative transport vehicle
7 self-propelled silage harvester
Detailed Description
The structural and operational principles of the present invention are described in detail below with reference to the accompanying drawings:
referring to fig. 1 and 7, fig. 1 is a schematic diagram illustrating an installation position of a measuring device according to an embodiment of the invention, and fig. 7 is a schematic diagram illustrating an operation of the measuring device according to an embodiment of the invention. The self-propelled silage harvester 7 to which the present invention is applicable comprises at least a pair of feeding rollers 24 of variable relative displacement. The invention relates to a feeding quantity measuring device of a self-propelled silage harvester, which comprises a feeding roller displacement measuring mechanism 1, a feeding roller rotating speed measuring mechanism 2, a silage moisture measuring mechanism 3, a walking speed measuring mechanism 4 and a vehicle-mounted computer 5. The vehicle-mounted computer 5 is respectively connected with the feeding roller displacement measuring mechanism 1, the feeding roller rotating speed measuring mechanism 2, the silage moisture measuring mechanism 3 and the walking speed measuring mechanism 4, receives the feeding roller displacement information, the feeding roller rotating speed information, the silage moisture content information and the walking speed information, performs data fusion through a feeding neural network model, takes the feeding roller displacement information, the feeding roller rotating speed information and the silage moisture content information as network input, takes real-time feeding amount as network output, establishes a nonlinear mapping relation between input and output to reflect the real-time feeding amount, and adjusts the real-time feeding amount by controlling the advancing walking speed of the self-propelled silage harvester 7.
Referring to fig. 2, fig. 2 is a schematic view of a feeding roller displacement measuring mechanism according to an embodiment of the present invention. The feeding roller displacement measuring mechanism 1 provided by the invention can be used for measuring the relative displacement change of the feeding roller 24 in real time and is used for measuring the feeding roller displacement information of the self-propelled silage harvester 7 in real time, and comprises a connecting spring 14 and a linear displacement sensor 13, wherein the connecting spring 14 is arranged between the front feeding floating roller 11 and the front feeding fixed roller 12, the connecting spring 14 is sleeved on the linear displacement sensor 13, one end of the connecting spring 14 is connected with one end of the linear displacement sensor 13 and is connected with a rotating shaft head of the front feeding floating roller 11 of the self-propelled silage harvester 7, and the other end of the connecting spring 14 is connected with the other end of the linear displacement sensor 13 and is connected with a rotating shaft head of the front feeding fixed roller 12 on the same side of the self-propelled silage harvester 7. When the feeding amount changes, the relative displacement of the feeding roller, namely the relative displacement between the front feeding floating roller 11 and the front feeding fixed roller 12, changes, and then the connecting spring 14 is deformed, and the deformation can be measured by the linear displacement sensor 13.
Referring to fig. 3, fig. 3 is a schematic view of a feeding roller rotation speed measuring mechanism according to an embodiment of the present invention. The feeding roller rotating speed measuring mechanism 2 of the embodiment is used for acquiring the rotating speed information of the feeding roller in real time and comprises a rotating speed measuring encoder 21 and a coupler 22, wherein one end of the rotating speed measuring encoder 21 is connected with a feeding roller shaft 23 of the self-propelled silage harvester 7 through the coupler 22. When the feed roller 24 rotates, the rotational speed measuring encoder 21 obtains the rotational speed value of the feed roller 24.
The silage moisture measuring mechanism 3 of this embodiment can acquire the change of the moisture of the fodder in real time, is used for acquiring the moisture content information of the silage in real time, comprises a microwave moisture measuring sensor 31, and is installed in the front part of the cooperative transportation vehicle 6 of the self-propelled silage harvester 7. When the chopped fodder is thrown to the cooperative transportation vehicle 6 by the throwing cylinder of the self-propelled silage harvester 7, the fodder covers the upper surface of the silage moisture measuring mechanism 3, the microwave moisture measuring sensor 31 can immediately measure the moisture content of the silage at the moment, and data are transmitted to the vehicle-mounted computer 5 in a wireless transmission mode.
Referring to fig. 4 and 5, fig. 4 is a schematic diagram of a walking speed measuring mechanism according to an embodiment of the invention, and fig. 5 is a right side view of fig. 4. The walking speed measuring mechanism 4 of this embodiment is configured to obtain, in real time, walking speed information of the walking of the self-propelled silage harvester 7, and includes a proximity switch 41 and a tachometer gear 42, where the tachometer gear 42 is installed on a walking driving shaft 44 of the self-propelled silage harvester 7, the tachometer gear 42 may be parallel to a brake pad 43 installed on the walking driving shaft 44, and the proximity switch 41 is set corresponding to teeth of the tachometer gear 42. When the self-propelled silage harvester 7 walks, the walking driving shaft 44 rotates, the speed measuring gear 42 rotates synchronously with the walking driving shaft 44, the proximity switch 41 counts the number of teeth of the speed measuring gear 42, namely the number of pulses, and the current walking speed value of the self-propelled silage harvester 7 is obtained through the number of pulses. And finally, transmitting the measured values of all the measuring mechanisms to the vehicle-mounted computer 5 in a serial port communication or wireless communication mode, carrying out data fusion on the measured values in the vehicle-mounted computer 5 through a trained artificial neural network model, and finally obtaining the real-time feeding quantity of the self-propelled silage harvester 7.
The invention relates to a feeding amount measuring method of a self-propelled silage harvester, which comprises the following steps:
step S100, measuring the displacement of a feeding roller, acquiring feeding roller displacement information of the self-propelled silage harvester 7 in real time, and transmitting the feeding roller displacement information to the vehicle-mounted computer 5; in the embodiment, a linear displacement sensor 13 arranged at a feeding roller of a feeding device of the silage harvester is used for acquiring a displacement value of the feeding roller in the vertical direction in real time, and the acquired displacement value is transmitted to the vehicle-mounted computer 5 through serial communication;
step 200, measuring the rotation speed of a feeding roller, acquiring the rotation speed information of the feeding roller of the self-propelled silage harvester 7 in real time, and transmitting the rotation speed information of the feeding roller to the vehicle-mounted computer 5; the rotation speed of the feeding roller 24 is measured in real time through a rotation speed measuring encoder 21 arranged at the head of the feeding roller shaft 23, and the obtained rotation speed value is transmitted to the vehicle-mounted computer 5 through serial communication;
step S300, measuring the silage water content, acquiring the harvested silage water content information in real time, and transmitting the silage water content information to the vehicle-mounted computer 5; the silage water content can be obtained in real time through the silage water content measuring mechanism 3 arranged in the cooperative transportation vehicle 6, and the obtained water content is transmitted to the vehicle-mounted computer 5 through serial communication;
step 400, measuring the walking speed of the harvester, acquiring the walking speed information of the self-propelled silage harvester 7 in real time, and transmitting the walking speed information to the vehicle-mounted computer 5; the advancing speed value of the harvester is obtained in real time through a walking speed measuring mechanism 4 arranged on the self-propelled silage harvester 7, and the measured walking speed information of the harvester is transmitted into a vehicle-mounted computer 5 in a serial port communication mode; and
and S500, acquiring feeding quantity, and carrying out data processing on the input feeding roller displacement information, feeding roller rotating speed information, silage water content information and walking speed information by the vehicle-mounted computer 5 by adopting a feeding quantity neural network model to obtain the current feeding quantity of the self-propelled silage harvester 7. The feeding roller displacement value, the feeding roller rotating speed value and the silage water content can be used as input layer data of the artificial neural network, the feeding amount is used as output layer data, the hidden layer and the output layer of the artificial neural network all adopt hyperbolic tangent transfer functions, the input layer data are transmitted to the untrained artificial neural network after being normalized, the network is trained, and therefore internal parameters of the artificial neural network are determined, and the nonlinear mapping relation of input and output is established. And (3) inputting a group of data by using the trained artificial neural network, so that the real-time feeding amount under the input can be obtained.
The method can further comprise a step S600 of controlling the advancing speed of the self-propelled silage harvester 7 according to the feeding amount displayed in real time by the vehicle-mounted computer 5 so as to obtain an optimal feeding amount.
Referring to fig. 6, fig. 6 is a schematic diagram of a feeding neural network according to an embodiment of the present invention. Wherein the feeding quantity neural network is a BP neural network and comprises an input layer, an output layer and an hidden layer, wherein the input layer comprises three neurons which are respectively feeding roller displacement delta, feeding roller rotating speed v and silage water content omicron, namely (P1) T= [ delta v omicron]The method comprises the steps of carrying out a first treatment on the surface of the The hidden layer comprises a plurality of neurons, the number S of the neurons is determined by simulation experiment results and is generally not lower than five, namely S is not less than 5; the output layer comprises a neuron as a feeding quantity, the number of neurons of the hidden layer is determined by a simulation experiment result, and the input of the output layer is the output a of the hidden layer 1 The output of the output layer is real-time feeding quantity a 2 The feeding neural network model is shown in figure 8;
wherein IW 1,1 Representing a connection weight matrix between each neuron of the input layer and each neuron of the hidden layer, wherein the determination of the connection weight is required to be trained by a network; b 1 A neuron bias value, b, for the hidden layer 1 Preferably 1; s is the number of neurons of the hidden layer; LW (LW) 2,1 Is a as 1 A connection weight matrix between the output layer and the neuron, b 2 A neuron bias value b for the output layer 2 Preferably 1; f (f) 1 、f 2 Transfer functions of the hidden layer and the output layer, f 1 、f 2 Are all preferably hyperbolic tangent transfer functions; n is n 1 ，n 2 Representing implicit layer and output layer transfer functions f, respectively 1 、f 2 Is a net input to the computer. The input of the output layer is a 1 ＝tagsig(IW 1,1 +b 1 ) The method comprises the steps of carrying out a first treatment on the surface of the The output of the output layer is the real-time feeding quantity a 2 ，a 2 ＝purelin(LW 2,1 +b 2 )。
Wherein, the connection weight IW between each neuron of the input layer and the hidden layer 1,1 For determination through network training. The connection weight IW 1,1 Adopting an advanced stop network training method, wherein the sample is verified to avoid the phenomenon of overfitting, and sample data is normalized to interval [ -1,1]The sample data of N groups (such as 130 groups) are sequenced according to the feeding amount, three parts of the sample data are extracted at equal intervals to be respectively used as a training sample (60%), a verification sample (20%) and a test sample (20%), and the training sample is used for calculating the gradient value of a network model and updating the weight IW of the network model 1,1 And LW (LW) 2,1 The method comprises the steps of carrying out a first treatment on the surface of the The verification sample is used for determining when to terminate training, the training is terminated when the error on the verification set rises in several iterations, and the weight and the bias value which generate extremely small error on the verification set are used as the final trained network connection weight and the neuron bias value; the test samples are used to compare different models and to test whether bad data packets are present.
The invention displays the displacement of the feeding roller, the rotating speed of the feeding roller, the water content of silage and the running speed in real time by a vehicle-mounted computer, automatically transmits the displacement value of the feeding roller, the rotating speed value of the feeding roller and the water content of silage to an artificial neural network model, and judges the feeding amount of a feeding harvester in real time. When the feeding amount is too large, the feeding mechanism is easy to be blocked, and the vehicle-mounted computer immediately gives an alarm to remind a driver of reducing the vehicle speed; when the feeding amount is too small, the harvester does not exert the maximum working efficiency, and the vehicle-mounted computer makes a schematic to remind a driver to increase the vehicle speed. The harvester achieves the best working efficiency and working quality, and also helps the harvester driver to control the vehicle speed without depending on experience.
Of course, the present invention is capable of other various embodiments and its several details are capable of modification and variation in light of the present invention, as will be apparent to those skilled in the art, without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (9)
1. The feeding amount measuring method of the self-propelled silage harvester is characterized by comprising the following steps of:
s100, measuring the displacement of a feeding roller, acquiring feeding roller displacement information of a self-propelled silage harvester in real time, and transmitting the feeding roller displacement information to a vehicle-mounted computer;
s200, measuring the rotation speed of a feeding roller, acquiring the rotation speed information of the feeding roller of the self-propelled silage harvester in real time, and transmitting the rotation speed information of the feeding roller to the vehicle-mounted computer;
s300, measuring the silage water content, acquiring the harvested silage water content information in real time, and transmitting the silage water content information to the vehicle-mounted computer;
s400, measuring the walking speed of the harvester, acquiring the walking speed information of the self-propelled silage harvester in real time, and transmitting the walking speed information to the vehicle-mounted computer; and
s500, acquiring feeding quantity, and carrying out data processing on the input feeding roller displacement information, feeding roller rotating speed information, silage water content information and walking speed information by using a feeding quantity neural network model through the vehicle-mounted computer to obtain the current feeding quantity of the self-propelled silage harvester;
wherein the feeding quantity neural network is a BP neural network and comprises an input layer, an output layer and an hidden layer, wherein the input layer comprises three neurons which are respectively feeding roller displacement delta, feeding roller rotating speed v and silage water content omicron, and the feeding quantity neural network is recorded as (P1) T= [ delta ]νο]The method comprises the steps of carrying out a first treatment on the surface of the The hidden layer includes a plurality of neurons; the output layer comprises a neuron, and the input of the output layer is the output a of the hidden layer 1 The output of the output layer is real-time feeding quantity a 2 The connection weight matrix between each neuron of the input layer and the hidden layer is IW 1,1 The method comprises the steps of carrying out a first treatment on the surface of the The neuron bias value of the hidden layer is b 1 The method comprises the steps of carrying out a first treatment on the surface of the The number of neurons of the hidden layer is S; neurons of the output layer and a 1 The connection weight matrix between the two is LW 2,1 The neuron bias value of the output layer is b 2 The method comprises the steps of carrying out a first treatment on the surface of the The transfer functions of the hidden layer and the output layer are f respectively 1 、f 2 ，n 1 ，n 2 Transfer functions f of hidden layer and output layer respectively 1 、f 2 Is a net input to the computer.
2. The method for measuring the feeding quantity of the self-propelled silage harvester according to claim 1, wherein the neuron number S of the hidden layer is determined by simulation experiment results, and S is more than or equal to 5.
3. The method for measuring the feed quantity of a self-propelled silage harvester according to claim 1 or 2, wherein b 1 1, b 2 1.
4. A method for measuring the feed rate of a self-propelled silage harvester as set forth in claim 3, wherein the transfer function f of the hidden layer and the output layer 1 、f 2 Are hyperbolic tangent transfer functions.
5. The method for measuring the feeding amount of a self-propelled silage harvester according to claim 4, wherein the input of the output layer is a 1 ＝tagsig(IW 1,1 +b 1 ) The method comprises the steps of carrying out a first treatment on the surface of the The output of the output layer is a 2 ＝purelin(LW 2,1 +b 2 )。
6. The self-propelled silage harvester feed measurement of claim 1 or 2The method is characterized in that the connection weight IW between each neuron of the input layer and the hidden layer 1,1 For determination through network training.
7. The method for measuring feeding amount of self-propelled silage harvester as set forth in claim 6, wherein the connection weight IW is 1,1 Normalizing the sample data to interval [ -1,1 ] by adopting an advanced network training stopping method]The N groups of sample data are sequenced according to the feeding quantity, three parts of sample data are extracted at equal intervals to be respectively used as training samples, verification samples and test samples, and the training samples are used for calculating gradient values of a network model and updating weights IW of the network model 1,1 And LW (LW) 2,1 The method comprises the steps of carrying out a first treatment on the surface of the The verification sample is used for stopping training, and training is stopped when errors on the verification set rise in several iterations, and the weight and the bias value which generate extremely small errors on the verification set are used as the final trained network connection weight and the neuron bias value; the test samples are used to compare different models and data packets tested.
8. The method for measuring the feed amount of a self-propelled silage harvester of claim 7, further comprising the steps of:
s600, controlling the advancing speed of the self-propelled silage harvester according to the feeding quantity displayed in real time by the vehicle-mounted computer so as to obtain the optimal feeding quantity.
9. A self-propelled silage harvester feed amount measuring device, characterized in that the self-propelled silage harvester feed amount measuring method according to any one of the claims 1-8 is adopted for feed amount detection, comprising:
the feeding roller displacement measuring mechanism is used for measuring feeding roller displacement information of the self-propelled silage harvester in real time and comprises a connecting spring and a linear displacement sensor, wherein the connecting spring is sleeved on the linear displacement sensor, one end of the connecting spring is connected with one end of the linear displacement sensor and is connected with a rotating shaft head of a front feeding floating roller of the self-propelled silage harvester, and the other end of the connecting spring is connected with the other end of the linear displacement sensor and is connected with a rotating shaft head of a front feeding fixed roller of the self-propelled silage harvester;
the feeding roller rotating speed measuring mechanism is used for acquiring rotating speed information of the feeding roller in real time and comprises a rotating speed measuring encoder and a coupler, wherein one end of the rotating speed measuring encoder is connected with a feeding roller shaft of the self-propelled silage harvester through the coupler;
the silage moisture measuring mechanism is used for acquiring the silage moisture content information in real time and is arranged at the front part of the cooperative transportation vehicle of the self-propelled silage harvester;
the walking speed measuring mechanism is used for acquiring walking speed information of the self-propelled silage harvester in real time, and comprises a proximity switch and a speed measuring gear, wherein the speed measuring gear is arranged on a walking driving shaft of the self-propelled silage harvester, and the proximity switch is arranged corresponding to the speed measuring gear; and
the vehicle-mounted computer is respectively connected with the feeding roller displacement measuring mechanism, the feeding roller rotating speed measuring mechanism, the silage moisture measuring mechanism and the walking speed measuring mechanism, receives feeding roller displacement information, feeding roller rotating speed information, silage moisture content information and walking speed information, performs data fusion through a feeding amount neural network model, takes the feeding roller displacement information, the feeding roller rotating speed information and the silage moisture content information as network input, takes real-time feeding amount as network output, establishes a nonlinear mapping relation between input and output to reflect the real-time feeding amount, and adjusts the real-time feeding amount by controlling the walking speed of the self-propelled silage harvester.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910485662.5A CN110942146B (en) | 2019-06-05 | 2019-06-05 | Method and device for measuring feeding quantity of self-propelled silage harvester |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910485662.5A CN110942146B (en) | 2019-06-05 | 2019-06-05 | Method and device for measuring feeding quantity of self-propelled silage harvester |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110942146A CN110942146A (en) | 2020-03-31 |
| CN110942146B true CN110942146B (en) | 2023-06-16 |
Family
ID=69905678
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910485662.5A Active CN110942146B (en) | 2019-06-05 | 2019-06-05 | Method and device for measuring feeding quantity of self-propelled silage harvester |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110942146B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111656951B (en) * | 2020-05-13 | 2021-11-12 | 中国农业大学 | Harvester speed control method and system |
| CN112631279B (en) * | 2020-12-09 | 2023-03-14 | 甘肃省机械科学研究院有限责任公司 | Silage harvester feeding amount control system and method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107660370A (en) * | 2016-07-28 | 2018-02-06 | 中国农业机械化科学研究院 | A kind of harvester for peanut feeding quantity monitoring method and monitoring system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6460008B1 (en) * | 2000-07-19 | 2002-10-01 | Ivan E. Hardt | Yield monitoring system for grain harvesting combine |
-
2019
- 2019-06-05 CN CN201910485662.5A patent/CN110942146B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107660370A (en) * | 2016-07-28 | 2018-02-06 | 中国农业机械化科学研究院 | A kind of harvester for peanut feeding quantity monitoring method and monitoring system |
Non-Patent Citations (1)
| Title |
|---|
| 联合收割机自动监测系统研究进展;夏连庆等;《农用机械》;20130731;141-144页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110942146A (en) | 2020-03-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200337235A1 (en) | Information inference for agronomic data generation in sugarcane applications | |
| US10188025B2 (en) | Yield monitor for windrow-collected materials | |
| CN110942146B (en) | Method and device for measuring feeding quantity of self-propelled silage harvester | |
| EP1196022A2 (en) | Yield monitor for forage crops | |
| Reyns et al. | A review of combine sensors for precision farming | |
| US9372109B2 (en) | Harvester elevator in-flow weight sensor and methods for the same | |
| CN102379188B (en) | Feed quantity measuring method and feed quantity monitoring device for combine harvester | |
| CN100514029C (en) | Stalk cutting test device and method | |
| CN101663938B (en) | Real-time yield monitoring system of spicate corn | |
| EP3314996B1 (en) | A crop management system for processing crop material | |
| CN108934416A (en) | One kind being based on BP neural network combined harvester multi-parameter operation-control system and method | |
| CN104322203A (en) | On-line monitoring system for feed quantity of combine harvester and monitoring method thereof | |
| CN104782318A (en) | Test system and method for testing field performance of combine harvester | |
| Mathanker et al. | Sensing miscanthus swath volume for maximizing baler throughput rate | |
| Sun et al. | Design of feed rate monitoring system and estimation method for yield distribution information on combine harvester | |
| CN110414859B (en) | 5T evaluation method for rice storage operation based on Internet of things technology | |
| CN104142674B (en) | Multi-input fault diagnosis method for combined harvester | |
| CN116158215A (en) | Rotary tillage operation intelligent control system based on surface topography characteristic information | |
| CN203942804U (en) | A kind of survey produced device | |
| Singh et al. | Precision in grain yield monitoring technologies: a review | |
| US20220304232A1 (en) | Method and Apparatus for Measuring Crop Throughput | |
| CN115756037A (en) | Greenhouse vegetable management and control method and system | |
| AU2022319711A1 (en) | Measuring seed cotton properties using near infrared sensing | |
| Changhua et al. | Development of combine grain yield monitor system with self-feedback function | |
| Mostafa et al. | Automated pressure regulation for a silage bagging machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |