CN109479862B - Variable pesticide spraying system and control method thereof - Google Patents
Variable pesticide spraying system and control method thereof Download PDFInfo
- Publication number
- CN109479862B CN109479862B CN201910046698.3A CN201910046698A CN109479862B CN 109479862 B CN109479862 B CN 109479862B CN 201910046698 A CN201910046698 A CN 201910046698A CN 109479862 B CN109479862 B CN 109479862B
- Authority
- CN
- China
- Prior art keywords
- variable
- pesticide spraying
- pesticide
- spraying
- population
- 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
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01M—CATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- A01M7/00—Special adaptations or arrangements of liquid-spraying apparatus for purposes covered by this subclass
- A01M7/0089—Regulating or controlling systems
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01M—CATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- A01M7/00—Special adaptations or arrangements of liquid-spraying apparatus for purposes covered by this subclass
- A01M7/0025—Mechanical sprayers
- A01M7/0032—Pressure sprayers
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01M—CATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- A01M7/00—Special adaptations or arrangements of liquid-spraying apparatus for purposes covered by this subclass
- A01M7/0089—Regulating or controlling systems
- A01M7/0096—Testing of spray-patterns
-
- 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
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
Abstract
The invention discloses a variable spraying system and a control method thereof, belonging to the technical field of variable spraying, and providing a variable spraying intelligent system based on a truncated mean value adaptive genetic PID control algorithm. The invention can finish variable pesticide spraying operation from the acquisition of image information without manual judgment, saves manpower and material resources, has high system response speed, small pesticide spraying control error and high precision, reduces the waste of pesticide and the pollution to the ecological environment, and improves the safety of agricultural products.
Description
Technical Field
The invention belongs to the technical field of variable spraying, and particularly relates to a variable spraying system and a control method thereof.
Background
Under current agricultural production environment, the pesticide generally adopts the mode of spraying on extensive ground, according to the unanimous principle of weeds coverage degree, unified application of medicine is carried out to whole farmland, not only causes the waste of pesticide and manufacturing cost's increase like this, causes negative effects to ecological environment and agricultural product safety moreover, is unfavorable for the sustainable development of agricultural. The variable pesticide spraying technology is put into use according to the specific disease, pest and weed conditions of each region in the farmland, the application amount of the pesticide is finely and accurately adjusted, the utilization rate of the pesticide is exerted to the maximum extent, the farmland environment is prevented from being damaged, the safety of agricultural products is improved, and the sustainable development of modern agriculture is effectively promoted.
At present, the most commonly used variable spraying methods are mainly pressure type and Pulse Width Modulation (PWM). The pressure type realizes variable spraying control by adjusting the pressure in the spraying system. However, the method can seriously change the original atomization characteristic of the nozzle, and the flow regulation range is narrow, thereby influencing the actual effect of pesticide spraying. The Pulse Width Modulation (PWM) method is to realize variable spraying operation by changing the duty ratio of the pulse width signal of the driving coil. The PWM has very strict requirements on the frequency of the system, a high frequency will reduce the service life of the solenoid valve, and a low frequency will affect the spraying effect.
Disclosure of Invention
Aiming at the defects of the existing variable spraying control mode, the invention aims to provide a variable spraying intelligent system based on a truncated mean value self-adaptive genetic PID control algorithm, which can realize spraying according to needs, improve the utilization rate of pesticides, reduce the production cost and reduce the pollution to the ecological environment. The system can also realize the alarm of the blockage of the spray head, display the performance indexes of the pesticide box, such as the pesticide amount, the system pressure, the decision-making pesticide spraying amount, the actual pesticide spraying amount and the like in real time, and comprehensively monitor the pesticide spraying state.
The invention is realized by the following technical scheme:
a variable spraying system comprises a spraying device, an image acquisition and processing system, a variable spraying control system, a monitoring system and a power supply system 27;
the pesticide spraying device comprises a pesticide box 1, a pesticide spraying pipeline, a filter 3, a pesticide pump 4, a pressure gauge 5, a safety valve 6, a distributor 10, an electromagnetic valve 11 and a spray head 12; the pesticide box 1 is connected with a pesticide pump 4 through a filter 3, a water outlet of the pesticide pump 4 is divided into two paths, one path of water flows back to the pesticide box 1 through a pressure gauge 5 and a safety valve 6 in sequence, the other path of water flows into one end of a pesticide spraying pipeline, the other end of the pesticide spraying pipeline is connected with a distributor 10, a pressure sensor 7, an electric valve 8 and a flow sensor 9 are arranged on the pesticide spraying pipeline in sequence, the distributor 10 is connected with an electromagnetic valve 11 and a spray head 12, and the distributor 10, the electromagnetic valve 11 and the spray head 12 are located above a pesticide spraying test bed;
the image acquisition and processing system comprises a camera 14 arranged on the pesticide spraying device and used for acquiring farmland images according to a judgment formula
Solving decision-making pesticide spraying quantity Q; wherein A is the weed area corresponding to a single ridge, V is the running speed of agricultural machinery, W is the distance between the spray nozzles, C = R/A, and R is the required spraying amount per hectare; the decision-making pesticide spraying amount Q is transmitted to a variable pesticide spraying control system through a wireless communication module 25;
the variable spraying control system adopts an STM32 microcontroller 26, calculates the difference between the decision spraying amount sent by the image acquisition and processing system and the actual spraying amount returned by the monitoring system, calculates the control voltage value which should be output at present by adopting a truncation mean value self-adaptive genetic PID algorithm according to the spraying amount difference, controls the opening of the electromagnetic valve 11 and realizes variable spraying;
the monitoring system is used for detecting the working state of the variable spraying system in real time and displaying the working state on a display screen of the variable spraying control system; the monitoring system comprises a sensor, a signal acquisition circuit 23, a display module 28 and an alarm circuit 30; the sensor is used for monitoring the working state of the variable spraying system in real time, the signal acquisition circuit 23 acquires a sensor signal, and the working state of the system is displayed on the display module 28 after the sensor signal is processed by the variable spraying control system;
the power supply system 27 is used for providing power for the image acquisition and processing system and the variable spraying control system.
Furthermore, the pesticide spraying device also comprises a leakage-proof shed 15, a water pump 16, a reservoir 17, a supporting suspension beam 18 and a weed model 19; the weed model 19 is positioned on the pesticide spraying test bed, the water pump 16 is positioned below the pesticide spraying test bed, the leakage-proof shed 15 is positioned above the water pump 16 and used for protecting the water pump 16, the reservoir 17 is connected with one end of the water pump 16, and the other end of the water pump 16 is connected with the pesticide box; the supporting suspension beams 18 are fixed on two sides of the inner wall of the pesticide spraying device through screws and clamping shells, and play a supporting role.
Further, the number of the electromagnetic valves 11 and the number of the nozzles 12 are 3.
Further, the sensors of the monitoring system comprise a flow sensor 20, a pressure sensor 21 and a liquid level sensor 22, wherein the flow sensor 20 and the pressure sensor 21 are installed in the spraying pipeline and used for detecting the pressure of the spraying pipeline and the actual spraying amount; a liquid level sensor 22 mounted in the medicine box 1 for detecting the amount of medicine in the medicine box 1; the three sensors are all connected with the STM32 microcontroller 26, analog quantity is converted into digital signal quantity through the signal acquisition circuit 23 and is sent into the microcontroller 26, the microcontroller 26 calculates according to the data of each sensor to obtain actual physical quantity, the working state of the variable pesticide spraying system is displayed on the liquid crystal display screen, and timely alarming is carried out when a fault occurs.
Further, the power supply system 27 includes a 5v power module, a 5v to 3.3v voltage stabilization chip, a 12v power module, and a 24v power module, and provides power for the camera 14 and the variable spraying control system.
The invention also aims to provide a truncation mean value self-adaptive genetic PID control algorithm of the variable pesticide spraying intelligent system, which adjusts the pesticide spraying amount in real time according to the difference value between the decision pesticide spraying amount and the actual pesticide spraying amount.
Further, the control method of the variable spraying control system comprises the following specific steps:
step 1: setting PID parameter (Kp, ki, kd) population in a variable pesticide spraying control system, wherein Kp is a proportional coefficient, ki is an integral coefficient, kd is a differential coefficient, an initialization population is randomly generated, the population size M and the maximum evolution algebra T are set to be 100, and the value ranges of Kp, ki and Kd are respectively set to be 0,100, 0,20 and 0, 5;
step 2: inputting decision-making pesticide spraying amount in a variable pesticide spraying control system according to decision-making information of an image acquisition and processing system;
and step 3: constructing a fitness function F according to the decision-making pesticide spraying amount acquired in the step 2, the actual pesticide spraying amount measured by the flow sensor and the adjusting time, wherein,
e (t) is the error of spraying, u (t) is the output voltage value of the microcontroller, tu is the regulating time, the regulating time is the time from the beginning of the system operation to the actual spraying amount reaching the decision spraying amount, w 1 、w 2 、w 3 As a weight value, w 1 、w 2 、w 3 The values are respectively 0.999,0.001 and 2.0, and the individual fitness value of each PID parameter is calculated;
and 4, step 4: removing adverse characters and extreme variation individuals in the to-be-selected parameter individuals by using a truncation mean operator according to the following formula (1), performing roulette selection according to the fitness value, and selecting high-quality individuals to be inherited to the next generation;
in the formula, X (X) 1 ,x 2 ,……x n ) Is all individuals of the original population, X represents the set of individuals, i.e. the population, where X is 1 ,x 2 ,……x n Is the corresponding individual, f avg Is the average fitness of the original population individuals, X L Is a set of extreme variant individuals and adverse trait individuals in the original population, f' avg Is the average fitness of the population after the truncation operation, X new (x 1 ,x 2 ,……x m ) Is a population obtained after the comparison of the truncation mean values;
and 5: adopting a crossover operator in the formula (2) to carry out pairwise crossover on individuals in the PID parameter population obtained in the step (4) to obtain a new generation PID parameter population;
f max is the maximum fitness value in the population, f avg Is the mean fitness value for each generation population, f' is the large fitness value in the two individuals that intersect; pc is the crossover probability, P c1 ，P c2 Respectively the maximum value and the minimum value of the cross probability;
and 6: and (4) performing gene mutation on the new PID parameter population individuals generated in the step (5) by adopting the mutation probability determined by the formula (3), and generating a new parameter population again.
f max Is the maximum fitness value in the population, f avg Is the average fitness value of each generation of population, and f is the fitness value of the variant individual; pm is mutation probability, P m1 ，P m2 The maximum value and the minimum value of the variation probability are respectively;
and 7: judging whether the maximum value T of the evolution algebra is reached, if the judgment result is negative, returning to the step 3, and recalculating the fitness value; if the judgment result is yes, outputting PID optimization parameters and a system response curve, and finishing the pesticide spraying process.
Further, the population size M in step 1 is 10-50.
Compared with the prior art, the invention has the following advantages:
the system utilizes an image processing technology to collect farmland weed information, obtains decision-making pesticide spraying amount according to a rapid processing algorithm, adopts a tail-ending mean value self-adaptive genetic PID control algorithm by a variable pesticide spraying control system, adaptively searches for optimal control parameters, outputs control voltage according to the difference value of the actual pesticide spraying amount and the decision-making pesticide spraying amount, controls the opening size of an electric valve, realizes accurate and rapid control on the pesticide spraying amount, and can complete monitoring of system pressure, liquid level and flow information. The invention can complete variable pesticide spraying operation from the acquisition of image information without manual judgment, saves manpower and material resources, has high system response speed, small pesticide spraying control error and high precision, reduces the waste of pesticides and the pollution to the ecological environment, and improves the safety of agricultural products.
Drawings
FIG. 1 is a schematic structural diagram of a variable spray system of the present invention;
FIG. 2 is a flow chart of a control method of a variable spray system of the present invention;
FIG. 3 is a graph comparing the control effect of the present invention with that of the prior art;
FIG. 4 is a schematic circuit diagram of a variable spray control system of the present invention;
in the figure: the device comprises a medicine chest 1, a liquid level sensor 2, a filter 3, a medicine pump 4, a pressure gauge 5, a safety valve 6, a pressure sensor 7, an electric valve 8, a flow sensor 9, a distributor 10, an electromagnetic valve 11, an anti-dripping spray head 12, a water storage pipe 13, a camera 14, a leakage-proof shed 15, a water pump 16, a reservoir 17, a support suspension beam 18, a weed model 19, a flow sensor 20, a pressure sensor 21, a liquid level sensor 22, a signal acquisition circuit 23, a decision-making medicine spraying amount 24, a wireless communication module 25, an STM32 microcontroller 26, a power supply system 27, a display module 28, an electric valve control circuit 29 and an alarm circuit 30.
Detailed Description
Example 1:
as shown in fig. 1, a variable spraying system comprises a spraying device, an image acquisition and processing system, a variable spraying control system, a monitoring system and a power supply system 27;
the pesticide spraying device is arranged on a pesticide spraying test bed and comprises a pesticide box 1, a pesticide spraying pipeline, a filter 3, a pesticide pump 4, a pressure gauge 5, a safety valve 6, a distributor 10, an electromagnetic valve 11 and a spray head 12; the pesticide box 1 is arranged on a ceiling and connected with a pesticide spraying pipeline and a return pipeline, the pesticide box 1 is connected with a pesticide pump 4 through a filter 3, a water outlet of the pesticide pump 4 is divided into two paths, one path of the water flows back to the pesticide box 1 through a pressure gauge 5 and a safety valve 6 in sequence, the other path of the water is injected into one end of the pesticide spraying pipeline, the other end of the pesticide spraying pipeline is connected with a distributor 10, a pressure sensor 7, an electric valve 8 and a flow sensor 9 are sequentially arranged on the pesticide spraying pipeline, the distributor 10 is connected with an electromagnetic valve 11 and a spray head 12, a leakage-proof shed 15 is arranged below a pesticide spraying test bed to play a role in protecting a water pump 16, the water pump 16 is connected with a reservoir 17 and the pesticide box 1 through pipelines, the water pump is started when the pesticide box mixes the pesticide, a support cantilever beam 18 is fixed on two sides of the test bed through screws and clamping shells, and a weed model 19 is adhered to a spraying table top through viscose;
the image acquisition and processing system comprises a camera 14 arranged on the pesticide spraying device and used for acquiring farmland images according to a judgment formula
Solving decision-making pesticide spraying quantity Q; wherein A is the area of the weeds corresponding to a single ridge, V is the running speed of the agricultural machine, W is the distance between the spray heads, C is a constant, C = R/A, and R is the required spraying amount per hectare; the decision-making pesticide spraying amount Q is transmitted to a variable pesticide spraying control system through a wireless communication module 25;
the variable pesticide spraying control system adopts an STM32 microcontroller 26, calculates the difference between the decision pesticide spraying amount sent by the image acquisition and processing system and the actual pesticide spraying amount returned by the monitoring system, calculates the control voltage value which should be output currently by adopting a truncation mean value self-adaptive genetic PID algorithm according to the pesticide spraying amount difference, and controls the opening of the electromagnetic valve 11 to realize variable pesticide spraying;
the monitoring system is used for detecting the working state of the variable spraying system in real time and displaying the working state on a display screen of the variable spraying control system; the monitoring system comprises a sensor, a signal acquisition circuit 23, a display module 28 and an alarm circuit 30; the sensor is used for monitoring the working state of the variable spraying system in real time, the signal acquisition circuit 23 acquires a sensor signal, and the working state of the system is displayed on the display module 28 after the sensor signal is processed by the variable spraying control system;
the sensors of the monitoring system comprise a flow sensor 20, a pressure sensor 21 and a liquid level sensor 22, wherein the flow sensor 20 and the pressure sensor 21 are arranged in the pesticide spraying pipeline and are used for detecting the pressure of the pesticide spraying pipeline and the actual pesticide spraying amount; a liquid level sensor 22 is mounted in the medicine box 1 for detecting the amount of medicine in the medicine box 1; the three sensors are all connected with the STM32 microcontroller 26, analog quantity is converted into digital signal quantity through the signal acquisition circuit 23 and is sent into the microcontroller 26, the microcontroller 26 calculates according to the data of each sensor to obtain actual physical quantity, the working state of the variable pesticide spraying system is displayed on the liquid crystal display screen, and timely alarming is carried out when a fault occurs.
The power supply system 27 is used for providing power for the image acquisition and processing system and the variable spraying control system.
When in medicine spraying operation, firstly, the water pump extracts enough water from the reservoir 17 to the medicine chest 1, then the pesticide is put into the medicine chest 1 to be mixed into liquid medicine with a certain proportion, the liquid medicine output by the medicine pump 4 is divided into two paths, and one path flows back to the medicine chest 1 through the safety valve 6; the other path is injected into a main pesticide spraying pipeline. The main pesticide spraying pipeline is connected with a water outlet pipe, the water outlet pipe is connected with a pressure sensor 7, an electric valve 8, a flow sensor 9 and a distributor 10, and the distributor is connected with 3 anti-drip nozzles 12. When the system is operated, the medicine pump 4 starts to work, and the liquid medicine is sprayed to crops through the spray head to complete the basic medicine spraying function.
The MV-EM040M industrial camera produced by Shaanxi dimensional digital image technology Limited of the image acquisition and processing system adopts a C-port lens with variable focal length, so that the anti-shake effect is good, and the lens and a horizontal line form an angle of 45 degrees.
A flow sensor 20 in the monitoring system adopts a Hall water flow sensor, an output signal is an electric pulse signal, a power supply voltage is direct current 5V, a flow monitoring range is 0.15-1.5L/min, and an instantaneous flow calculation formula is F =76 × Q +/-5%; the device has the functions of converting the amount and the flowing condition of liquid medicine in the medicine spraying device into electric signals, sending the signals to an STM32 through an acquisition amplifying circuit and an A/D (analog to digital) conversion circuit, calculating the medicine spraying amount of a spray head according to the frequency of pulse signals, and displaying the medicine spraying amount on a display module; when the electric valve is opened, if the electric pulse signal disappears, the STM32 detects that the flow of the spray nozzle is zero, the spray nozzle is indicated to be blocked, the STM32 sends out a signal, and the alarm device is started to alarm the blockage. The pressure sensor 7 adopts a star analyzer CYYZ11 inlet diffused silicon water pressure sensor, the measuring range is 0-1.6MPa, the power supply voltage is 12-36 VDC, and the precision grade is 0.1% FS; the liquid level sensor 22 adopts a three-wire system voltage output liquid level sensor produced by a Sonye instrument, the voltage output range is 0-5V, the power supply voltage is 24V, and the accuracy is 0.1 level. The effect is that the pressure of medicine pipeline is mainly spouted in the monitoring, through the pressure conversion that will spout in the medicine pipeline for the signal of telecommunication to send the signal into STM32 through amplifier circuit and analog-to-digital conversion circuit, calculate through STM32 and handle and reachs the pressure of spouting the medicine pipeline, and show on display module, can the main pressure of spouting the medicine pipeline of real-time supervision, prevent that pressure is too big and the undersize, keep the stability of pressure. Liquid level sensor adopts the MIK-P260 liquid level sensor of the automatic company of american control, installs in the medicine chest, through the liquid level information conversion with in the medicine chest for the signal of telecommunication to send into STM32 with the signal amplification circuit through analog-to-digital conversion circuit, calculate the liquid level value that reachs the medical kit through STM32, and show on display module, can real-time detection medical kit liquid level, prevent that the medical kit from lacking the medicine or overflowing the condition and taking place.
In the variable pesticide spraying control system, a core controller adopts a 32-bit ARM core STM32F103 series processor, 3.3V power supply is adopted, and the working frequency of a chip is set to be 72MHz. The signal acquisition circuit selects an AD7606 multi-channel data acquisition module, a 16-bit 8-channel ADC and a sampling frequency of 200KHZ, and is used for acquiring the electric signals of the sensor and performing digital quantity conversion. The wireless communication module 25 is composed of a USR-C322WIFI communication module and a level conversion circuit. The display module 28 adopts a 12864 liquid crystal screen design and is used for displaying the spraying pressure value of the system, the current spraying amount and the liquid level information of the pesticide box. The electric valve control circuit 29 is composed of an electric valve and a voltage amplification module, wherein the electric valve is a PM-02 type micro electric valve, a 24V direct current power supply is used for supplying power, an input/output signal is an analog quantity of 0-10V, the opening degree of the valve is 0-100%, and the voltage amplification module is a linear magnetic isolation amplifier ISOEM-U5-P3-O5 which is used for linearly amplifying 0-3.3V to 0-10V so as to meet the input voltage requirement of the electric valve. The alarm circuit 30 is composed of a buzzer, an LED lamp and a triode circuit, and performs sound and light alarm when the pipeline is blocked. The power supply system 27 provides 5V and 3.3V dc power to the control system.
Example 2:
a variable spraying control method comprises the following specific steps:
step 1: setting PID parameter (Kp, ki, kd) population (PID parameter population means a group of individuals, each individual is a group of parameter values (Kp, ki, kd)), kp is a proportional coefficient, ki is an integral coefficient, kd is a differential coefficient, randomly generating an initialization population, setting the population scale M (the population scale is usually between 10 and 50, and the optimization effect of the actual problem is specifically determined) as 30, setting the maximum evolutionary algebra T as 100, and respectively setting the value ranges of Kp, ki and Kd as [0,100], [0,20], [0,5];
step 2: inputting decision-making pesticide spraying amount in a variable pesticide spraying control system according to decision-making information of an image acquisition and processing system;
and step 3: constructing a fitness function F according to the decision-making pesticide spraying amount in the step 2, the actual pesticide spraying amount error measured by the flow sensor and the adjusting time,
e (t) is a spraying error, u (t) is a controller output voltage value, tu is adjusting time (namely the time from the beginning of the system to the actual spraying amount reaching the decision-making spraying amount), w1, w2 and w3 are weights, the values of w1, w2 and w3 are respectively 0.999,0.001 and 2.0 (determined by experience and without an absolute calculation formula), and the individual fitness value of each PID parameter is calculated;
and 4, step 4: removing adverse characters and extreme variation individuals in the to-be-selected parameter individuals by using a truncation mean operator according to the formula (1), performing roulette selection according to the fitness value, and selecting high-quality individuals to be inherited to the next generation;
in the formula: x (X) 1 ,x 2 ,……x n ) Is all individuals of the original population (X represents the set of individuals, i.e. the population, where X is 1 ,x 2 ,……x n Is the corresponding individual), f avg The method is the average fitness of the original population individuals (the genetic algorithm is a process of obtaining an optimal value through multiple iterations, three basic operations of selection, intersection and variation are carried out on each generation, the population is updated after each basic operation is finished, and f in the formula avg Which can be interpreted as a new population obtained after the completion of the last basic operation), X) L Is a set of extreme variant individuals and adverse trait individuals in the original population, f' avg Is the average fitness of the population after the truncation operation, X new (x 1 ,x 2 ,……x m ) Is a population obtained after the comparison of the truncation mean values;
and 5: adopting a crossover operator in the formula (2) to carry out pairwise crossover on individuals in the PID parameter population obtained in the step (4) to obtain a new generation PID parameter population;
in the formula: f. of max Is the maximum fitness value in the population, f avg Is the mean fitness value of each generation population, f' is the greater fitness value of the two crossed individuals, and f is the fitness value of the variant individual. P is c1 ，P c2 The maximum value and the minimum value of the cross probability; pc is the crossover probability, i.e., the occurrence of the locus of any two individuals in the population, and the individuals are binary coded in the initialization of step 1, so that each individual can be regarded as a chromosome carrying the 01 code and can be compared with the biological chromosomes.
Step 6: and (4) performing gene mutation on the new PID parameter population individuals generated in the step (5) by adopting the mutation probability determined by the formula (3), and generating a new parameter population again.
In the formula: f. of max Is the maximum fitness value in the population, f avg Is the average fitness value of each generation of population, and f is the fitness value of the variant individual. P m1 ，P m2 The maximum value and the minimum value of the variation probability; pm is the mutation probability, and for a single individual, each individual can be regarded as a chromosome as explained above, and mutation means that the locus of a certain position is changed, and 0 becomes 1 or 1 becomes 0.
And 7: judging whether the maximum value T of the evolution algebra is reached, if not, returning to the step 3, and recalculating the fitness value; if the judgment result is yes, outputting PID optimization parameters and a system response curve, and finishing the pesticide spraying process.
Example 3:
a variable spraying control method comprises the following specific steps:
step 1: randomly generating PID parameters to initialize populations, wherein the population size M =30, namely [ Kp, ki, kd ] = [58.6804,15.8944,2.6784;86.8328,8.3480,3.1281;53.5777,3.7928,4.0420;7.3021,4.9462,1.3001;46.9795,10.1466,1.2366;44.6921,1.7009,3.9687;60.8798,6.2561,2.0723;7.5660,7.1945,0.1222;31.7595,10.7722,1.4272;80.2346,4.9658,2.5660;42.1408,13.1769,1.8719;57.5367,6.7253,4.6921;46.0997,8.6217,0.8798;24.1935,19.6481,0.2933;78.9150,8.9932,4.0371;27.5367,15.5425,3.8465;53.6657,1.3490,1.4027;51.9062,5.9238,2.4731;19.0909,10.8504,1.4956;29.0323,12.5318,4.6432;42.1408,17.1848,1.1193;28.9443,4.8876,3.5533;17.7713,8.3871,0.2688;64.6628,12.9423,4.0909;19.4428,10.4985,2.1261;66.7742,6.4712,0.1955;26.7449,4.4184,1.6471;89.0323,16.0899,3.6755;27.4487,13.9198,4.7898;77.9472,15.4839,2.7468], set the maximum evolution algebra T =100;
step 2: inputting decision-making spraying amount of 0.750L/min,0.850L/min,0.950L/min,1.050L/min,1.150L/min and 1.250L/min in a variable spraying control system according to decision-making information of an image acquisition and processing system;
and step 3: constructing fitness function
Calculating the fitness value of each PID parameter individual;
and 4, step 4: removing bad characters and extreme variation individuals in the to-be-selected parameter individuals by using a truncation mean operator according to the formula (1), performing roulette selection according to the fitness value, and selecting high-quality individuals to be inherited to the next generation, namely X new (x 1 ,x 2 ,……x m )；
And 5: adopting a crossover operator of the formula (2) to carry out pairwise crossover on individuals in the PID parameter population obtained in the step (4) to obtain a new generation PID parameter population, wherein the maximum crossover probability P c1 =0.9, minimum crossover probability P c2 ＝0.1；
And 6: generating the mutation probability determined by the formula (3) for the step 5The new PID parameter population individuals carry out gene mutation to generate a new parameter population again, wherein the maximum mutation probability P m1 =0.1, minimum mutation probability P m2 ＝0.01；
And 7: judging whether the maximum value of the evolution algebra is reached to 100, if the judgment result is negative, returning to execute the step 3, and recalculating the fitness value; if the judgment result is yes, outputting PID optimization parameters and a system response curve, and finishing the pesticide spraying process.
The test results obtained according to steps 1-7 and the control effect of the conventional method are shown in FIG. 3, and the data pairs are shown in Table 1.
TABLE 1 set amount of drug sprayed and actual amount of drug sprayed
Unit: l/min
Fig. 3 is a comparison graph of the control effects of the truncated mean value adaptive genetic PID algorithm, PWM, pressure type and conventional PID, which shows that the system response times of the truncated mean value adaptive genetic PID algorithm, PWM, pressure type and conventional PID control are 1.57s,2.03s,2.60 s and 3.30s, respectively, and the overshoot amounts are 1.25%, 2.01%, 2.21% and 4.63%, respectively.
As can be seen from the data analysis of Table 1, compared with the existing variable spraying technology including pulse width modulation, pressure type and conventional PID control, the system under the control of the self-adaptive genetic PID with the truncation mean value has the advantages of higher response speed, higher control precision and better stability, and can more quickly and accurately control the spraying amount.
Claims (6)
1. A variable spray control method is realized by applying a variable spray control system, and the variable spray control system comprises a spray device, an image acquisition and processing system, a variable spray control system, a monitoring system and a power supply system (27);
the pesticide spraying device comprises a pesticide box (1), a pesticide spraying pipeline, a filter (3), a pesticide pump (4), a pressure gauge (5), a safety valve (6), a distributor (10), an electromagnetic valve (11) and a spray head (12); the pesticide box (1) is connected with the pesticide pump (4) through the filter (3), a water outlet of the pesticide pump (4) is divided into two paths, one path of water flows back to the pesticide box (1) through the pressure gauge (5) and the safety valve (6) in sequence, the other path of water is injected into one end of the pesticide spraying pipeline, the other end of the pesticide spraying pipeline is connected with the distributor (10), the pesticide spraying pipeline is sequentially provided with the pressure sensor (7), the electric valve (8) and the flow sensor (9), the distributor (10) is connected with the electromagnetic valve (11) and the spray head (12), and the distributor (10), the electromagnetic valve (11) and the spray head (12) are positioned above the pesticide spraying test table;
the image acquisition and processing system comprises a camera (14) arranged on the pesticide spraying device and used for acquiring farmland images according to a judgment formula
Solving decision-making pesticide spraying quantity Q; wherein A is the area of the weeds corresponding to a single ridge, V is the running speed of the agricultural machine, W is the distance between the spray heads, C = R/A, and R is the spray amount required by each hectare; the decision-making pesticide spraying amount Q is transmitted to a variable pesticide spraying control system through a wireless communication module (25);
the variable pesticide spraying control system adopts an STM32 microcontroller (26), calculates the difference between the decision pesticide spraying amount sent by the image acquisition and processing system and the actual pesticide spraying amount returned by the monitoring system, calculates the current control voltage value to be output by adopting a truncated mean value self-adaptive genetic PID algorithm according to the pesticide spraying amount difference, and controls the opening of the electromagnetic valve (11) to realize variable pesticide spraying;
the monitoring system is used for detecting the working state of the variable spraying system in real time and displaying the working state on a display screen of the variable spraying control system; the monitoring system comprises a sensor, a signal acquisition circuit (23), a display module (28) and an alarm circuit (30); the sensor is used for monitoring the working state of the variable spraying system in real time, the signal acquisition circuit (23) acquires a sensor signal, and the working state of the system is displayed on the display module (28) after the sensor signal is processed by the variable spraying control system;
the power supply system (27) is used for providing power for the image acquisition and processing system and the variable pesticide spraying control system; the method is characterized by comprising the following specific steps:
step 1: setting PID parameter (Kp, ki, kd) population in a variable pesticide spraying control system, wherein Kp is a proportional coefficient, ki is an integral coefficient, kd is a differential coefficient, an initialization population is randomly generated, the population size M and the maximum evolution algebra T are set to be 100, and the value ranges of Kp, ki and Kd are respectively set to be 0,100, 0,20 and 0, 5;
step 2: inputting decision-making pesticide spraying amount in a variable pesticide spraying control system according to decision-making information of an image acquisition and processing system;
and step 3: constructing a fitness function F according to the decision-making pesticide spraying amount acquired in the step 2, the actual pesticide spraying amount measured by the flow sensor and the adjusting time, wherein,
e (t) is the error of spraying, u (t) is the output voltage value of the microcontroller, tu is the adjustment time, the adjustment time is the time from the beginning of the system operation to the actual spraying amount reaching the decision-making spraying amount, w 1 、w 2 、w 3 As a weight value, w 1 、w 2 、w 3 The values are respectively 0.999,0.001 and 2.0, and the individual fitness value of each PID parameter is calculated;
and 4, step 4: removing the bad characters and extreme variation individuals in the to-be-selected parameter individuals by using a truncation mean operator according to the following formula (1), performing roulette selection according to the fitness value, and selecting high-quality individuals to be inherited to the next generation;
in the formula, X (X) 1 ，x 2 ，......x n ) Is all individuals of the original population, X represents the set of individuals, i.e. the population, where X is 1 ，x 2 ，......x n Is the corresponding individual, f avg Is the average fitness of the original population individuals, X L Is a set of extreme variant individuals and adverse trait individuals in the original population, f' avg Is the average of the population after the truncation operationFitness, X new (x 1 ，x 2 ，......x m ) Is a population obtained after the comparison of the truncation mean values;
and 5: adopting a crossover operator in the formula (2) to carry out pairwise crossover on individuals in the PID parameter population obtained in the step (4) to obtain a new generation PID parameter population;
f max is the maximum fitness value in the population, f avg Is the mean fitness value for each generation population, f' is the large fitness value in the two individuals that intersect; pc is the crossover probability, P c1 ，P c2 Respectively the maximum value and the minimum value of the cross probability;
and 6: and (4) performing gene mutation on the new PID parameter population individuals generated in the step (5) by adopting the mutation probability determined by the formula (3), and generating a new parameter population again.
f max Is the maximum fitness value in the population, f avg F is the fitness value of the variant individual; pm is the mutation probability, P m1 ，P m2 The maximum value and the minimum value of the variation probability are respectively;
and 7: judging whether the maximum value T of the evolution algebra is reached, if the judgment result is negative, returning to the step 3, and recalculating the fitness value; if the judgment result is yes, outputting PID optimization parameters and a system response curve, and finishing the pesticide spraying process.
2. A variable spray control method as claimed in claim 1, wherein said spray means further comprises a leakage prevention shed (15), a water pump (16), a reservoir (17), a support suspension beam (18) and a weed model (19); the weed model (19) is positioned on the pesticide spraying test bed, the water pump (16) is positioned below the pesticide spraying test bed, the anti-leakage shed (15) is positioned above the water pump (16) and used for protecting the water pump (16), the reservoir (17) is connected with one end of the water pump (16), and the other end of the water pump (16) is connected with the pesticide box; the supporting suspension beam (18) is fixed on two sides of the inner wall of the pesticide spraying device through a screw and a clamping shell to play a supporting role.
3. The variable spray control method according to claim 1, wherein the number of the solenoid valves (11) and the spray heads (12) is 3.
4. The variable spray control method according to claim 1, wherein the sensors of the monitoring system comprise a flow sensor (20), a pressure sensor (21), and a liquid level sensor (22), and the flow sensor (20) and the pressure sensor (21) are installed in the spray pipeline for detecting the pressure of the spray pipeline and the actual spray amount; the liquid level sensor (22) is arranged in the medicine box (1) and is used for detecting the amount of the medicine in the medicine box (1); three sensors all link to each other with STM (32) microcontroller (26), convert the analog quantity into digital signal volume and send into microcontroller (26) through signal acquisition circuit (23), and microcontroller (26) calculate according to the data of each sensor and draw actual physical quantity, show the operating condition that variable spouts the medicine system on liquid crystal display to in time report to the police when breaking down.
5. The variable spraying control method as claimed in claim 1, wherein the power supply system (27) comprises a 5v power supply module, a 5v to 3.3v voltage stabilization chip, a 12v power supply module and a 24v power supply module, and supplies power to the camera (14) and the variable spraying control system.
6. A variable spray control method according to claim 1, wherein the population size M in step 1 is 10-50.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910046698.3A CN109479862B (en) | 2019-01-18 | 2019-01-18 | Variable pesticide spraying system and control method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910046698.3A CN109479862B (en) | 2019-01-18 | 2019-01-18 | Variable pesticide spraying system and control method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109479862A CN109479862A (en) | 2019-03-19 |
| CN109479862B true CN109479862B (en) | 2023-04-07 |
Family
ID=65714732
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910046698.3A Active CN109479862B (en) | 2019-01-18 | 2019-01-18 | Variable pesticide spraying system and control method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109479862B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112690265B (en) * | 2020-12-24 | 2022-09-06 | 吉林农业大学 | Timed pesticide spraying device and method for pest control |
| CN112913388A (en) * | 2021-01-11 | 2021-06-08 | 青岛农业大学 | Peanut seeder capable of spraying pesticide in precise quantity |
| CN112889786B (en) * | 2021-01-15 | 2023-04-28 | 吉林农业大学 | Pesticide spraying system for tracking field crop seedling area in real time and control method |
| DE102021101299A1 (en) * | 2021-01-22 | 2022-07-28 | Amazonen-Werke H. Dreyer SE & Co. KG | Method for controlling valves of a spray boom of an agricultural spreading machine |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5475614A (en) * | 1994-01-13 | 1995-12-12 | Micro-Trak Systems, Inc. | Method and apparatus for controlling a variable fluid delivery system |
| CN206576139U (en) * | 2017-03-23 | 2017-10-24 | 吉林农业大学 | Motor push rod type self-adjusting spraying machine |
| CN109122633A (en) * | 2018-06-25 | 2019-01-04 | 华南农业大学 | The accurate variable-rate spraying device of the plant protection drone of Decision of Neural Network and control method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8352400B2 (en) * | 1991-12-23 | 2013-01-08 | Hoffberg Steven M | Adaptive pattern recognition based controller apparatus and method and human-factored interface therefore |
| CN104082267B (en) * | 2014-07-08 | 2015-10-28 | 重庆科技学院 | Automatic variable spray medicine system and control method thereof |
| CN105165786B (en) * | 2015-08-12 | 2018-03-09 | 吉林大学 | Variable medicine spraying machine based on IMAQ |
-
2019
- 2019-01-18 CN CN201910046698.3A patent/CN109479862B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5475614A (en) * | 1994-01-13 | 1995-12-12 | Micro-Trak Systems, Inc. | Method and apparatus for controlling a variable fluid delivery system |
| CN206576139U (en) * | 2017-03-23 | 2017-10-24 | 吉林农业大学 | Motor push rod type self-adjusting spraying machine |
| CN109122633A (en) * | 2018-06-25 | 2019-01-04 | 华南农业大学 | The accurate variable-rate spraying device of the plant protection drone of Decision of Neural Network and control method |
Non-Patent Citations (1)
| Title |
|---|
| 张炳兰 ; .改进遗传算法PID在电液伺服系统的应用.自动化仪表.2017,(08),第32-36页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109479862A (en) | 2019-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109479862B (en) | Variable pesticide spraying system and control method thereof | |
| CN104082267B (en) | Automatic variable spray medicine system and control method thereof | |
| CN108450386B (en) | A kind of regulating and controlling water quality System and method for of industrialized aquiculture | |
| CN105165786B (en) | Variable medicine spraying machine based on IMAQ | |
| CN101716567B (en) | Intelligent variable medicine spraying machine based on prescription chart control | |
| CN102613161B (en) | Control system for boom sprayer | |
| CN106135174A (en) | A kind of Intelligent pesticide application equipment | |
| CN102141801A (en) | Rice field growth environment monitoring and irrigation control system | |
| CN107330804A (en) | A kind of wisdom water conservancy management and control cloud platform and method | |
| CN103960221A (en) | Real-time pesticide mixing variable spraying device and method | |
| CN108881423A (en) | A kind of wisdom agricultural planting system based on Internet of Things | |
| CN106338994B (en) | A kind of greenhouse logistics plant protection robot control system and method | |
| CN103657923A (en) | Spraying flow grouping regulation and control method for PWM intermittent spray type variable spraying system | |
| CN104121998A (en) | Temperature pre-warning method for use in daylight greenhouse environment intelligent monitoring system | |
| CN104678928A (en) | Universal variable operation controller and control method thereof | |
| CN104855248A (en) | Automatic irrigation system | |
| CN201565393U (en) | Intelligent variable insecticide spraying machine based on control of prescription map | |
| CN106773720A (en) | A kind of warmhouse booth environment automatic control system | |
| CN111937835A (en) | Orchard wind bag type targeting sprayer and method based on ultrasonic sensing | |
| Nan et al. | Research on profiling tracking control optimization of orchard sprayer based on the phenotypic characteristics of tree crown | |
| Xu et al. | Design and experiment of variable spraying system based on multiple combined nozzles | |
| CN113485470A (en) | Variable spray control method, device and system | |
| CN103404288B (en) | The method of Full-automatic greenhouse compounding and sprinkling and system thereof | |
| CN202757200U (en) | Flux balance adjusting system of heating system | |
| CN112488467A (en) | Water planting crop fertilizer injection unit based on multiscale habitat information |
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 |