CN110501983B - Expert control system and control method based on batch-type coating machine - Google Patents
Expert control system and control method based on batch-type coating machine Download PDFInfo
- Publication number
- CN110501983B CN110501983B CN201910698404.5A CN201910698404A CN110501983B CN 110501983 B CN110501983 B CN 110501983B CN 201910698404 A CN201910698404 A CN 201910698404A CN 110501983 B CN110501983 B CN 110501983B
- Authority
- CN
- China
- Prior art keywords
- coating
- expert
- module
- machine
- parameter
- 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
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM]
- G05B19/41875—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM] characterised by quality surveillance of production
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32368—Quality control
Abstract
The invention discloses an expert control system based on a batch-type coating machine, which comprises a coating machine, an image detection system and an expert system, wherein the expert system comprises a control cabinet and a display screen, a man-machine interaction interface, a knowledge base module, an inference machine module, a feedback self-learning module and a control center are arranged in the control cabinet, the control center is an industrial personal computer, the industrial personal computer is respectively and electrically connected with the man-machine interaction interface, the feedback self-learning module and a PLC (programmable logic controller) control module of the coating machine in a bidirectional mode, the man-machine interaction interface is electrically connected with the knowledge base module in an output mode, the feedback self-learning module is electrically connected with the inference machine module in an output mode, and the signal input end of the feedback self-learning module is further in signal connection with the signal output end of the image detection. The invention realizes automatic optimization and combination of processing parameters aiming at different seeds, improves the efficiency, saves the dosage of the coating agent and has obvious effect on the premise of meeting the coating quality and the qualification rate.
Description
Technical Field
The invention belongs to the technical field of seed coating, and particularly relates to an expert control system and an expert control method based on a batch-type coating machine.
Background
The seed coating technology can effectively prevent and control plant diseases and insect pests, protect seed growth, improve seed vigor, reduce environmental pollution, reduce production cost and promote crop yield increase and harvest, has huge economic and ecological benefits, is an important way for realizing seed processing mechanization and improved seed standardization, and is widely applied worldwide. The seed coating machine has been in the history of hundreds of years, and is developed rapidly in China in recent years. The design of the traditional seed coating machine is more focused on a mechanical part, an electric control system adopts a simple device, accurate control is difficult to realize, all parts can not work in a coordinated and ordered way, manual participation is needed, processing parameters depend on the experience of an operator, and the improvement of seed coating quality is severely restricted. Although the domestic seed coating equipment is improved by products of several generations on the basis of the structure principle, a coating machine control system is developed by taking a PLC (programmable logic controller) or a chip development board as a control core to achieve the aim of accurate control, the control system still has a great improvement space.
In recent years, expert systems have been increasingly used, and have become the most active and most important field in artificial intelligence technology. At present, the coating machine expert system is rarely researched, and a set of expert system is constructed by taking a singlechip as a core in Yanwanxia and the like to achieve the aim of accurate coating, but the knowledge acquisition and reasoning rules need to be further improved. In addition, the design of the traditional seed coating machine is more focused on a mechanical part, an electric control system adopts a simple device, manual participation is needed, accurate control is difficult to realize, all parts can not work in a coordinated and ordered way, processing parameters depend on the experience of an operator, the intelligent degree is low, and the improvement of the seed coating quality is severely restricted. Therefore, an expert system of the coating machine with on-line feedback and intelligent optimization strategies needs to be designed to realize the intellectualization and the precision of the control.
Disclosure of Invention
The invention aims to construct an expert control system based on a batch-type coating machine, and simultaneously provides a corresponding control method to realize intellectualization and precision of control of the coating machine.
The invention is realized by the following technical scheme:
an expert control system based on a batch-type coating machine comprises the coating machine and an image detection system, wherein the coating machine is the batch-type coating machine and consists of a dosing system, a feeding system, a coating system and a PLC control module, and the PLC control module is respectively in signal connection with the dosing system, the feeding system and the coating system and is used for controlling the coating machine to operate; the image detection system is arranged on one side of the coating system, the signal input end of the image detection system is in signal connection with the signal output end of the PLC control module and is used for detecting the seed coating quality and feeding back the coating qualification rate in real time, the image detection system further comprises an expert system, the expert system is arranged on one side of the image detection system and comprises a control cabinet and a display screen, the display screen is installed on the front side of the control cabinet, a man-machine interaction interface, a knowledge base module, an inference engine module, a feedback self-learning module and a control center are arranged in the control cabinet, the control center is an industrial personal computer, the industrial personal computer is respectively and electrically connected with the man-machine interaction interface, the feedback self-learning module and the PLC control module of the coating machine in a bidirectional mode, the man-machine interaction interface is electrically connected with the knowledge base module, the feedback self-learning module is electrically connected with the inference engine module in an output mode, the, and the coating qualification rate is used for receiving the feedback of the image detection system.
The technical scheme for further solving the problem is that the medicine adding system is arranged on one side of the coating system and comprises a medicine storage barrel, a medicine conveying pump and a medicine conveying pipe; the feeding system is arranged at the upper part of the coating system and comprises a feeding bin and a storage bin; the coating system comprises a throwing disc, a drum, a discharge port, a throwing disc motor and a drum motor; one end of the medicine conveying pipe is positioned in the medicine storage barrel, the other end of the medicine conveying pipe is positioned above the throwing disc, a medicine conveying pump is arranged on the medicine conveying pipe, and the medicine conveying pipe conveys the seed coating agent in the medicine storage barrel to the throwing disc through the medicine conveying pump; the feeding bin is positioned at the upper part of the drum, a storage bin is arranged right above the feeding bin, and seeds are placed into the drum through the storage bin; the throwing disc is positioned at the upper part of the middle part in the drum, the lower part of the throwing disc is provided with a throwing disc motor, and the throwing disc is electrically connected with the throwing disc motor and is driven to rotate by the control of the throwing disc motor; a drum motor is arranged below the disc throwing motor, and the drum is electrically connected with the drum motor and is driven to rotate by the drum motor; the discharge port is positioned on the side wall of the drum, and an image detection system is arranged at the end part of the discharge port; and the signal output end of the PLC control module is respectively connected with the signal input ends of the medicine conveying pump, the disc throwing motor and the drum motor.
The technical scheme for further solving the problem is that a weighing sensor is arranged in the storage bin, and the signal input end of the weighing sensor is connected with the signal output end of the PLC control module and used for controlling the feeding weight of each batch.
The invention further solves the technical scheme that the throwing disc and the rotary drum are hollow round tables with trapezoidal sections.
The invention also discloses a control method of the expert control system based on the batch-type coating machine, which comprises the following specific steps:
s1: starting and initializing an expert system and an image monitoring system;
s2: the expert system recommends each processing parameter of the coating machine through a neural network model according to the coating quality target value and the seed characteristic set by the user, and the processing parameters are confirmed or modified by the user;
s3: the coating machine receives the parameters confirmed or modified by the user and performs coating operation;
s4: the image detection system detects the coating quality and the qualification rate of the product according to the sampling period set by the expert system and feeds the coating quality and the qualification rate back to the expert system in real time for reasoning optimization;
s5: and (4) the expert system receives the result fed back in the step S4 to further train the neural network model on line, when the accuracy of the predicted value of the neural network model is within reasonable errors within the specified times, a coating optimization strategy based on the artificial bee colony algorithm is started, otherwise, the expert system returns to the step S2 to re-determine all processing parameters of the coating machine, and the optimization is stopped until the actual processing result is within the reasonable field of the expected target.
Further, in step S2, fuzzy recommendation is performed on each processing parameter of the coating machine by the feedback self-learning module and the inference machine module of the expert system, and the specific steps are as follows:
s21: meeting yield for desired targetAnd (3) meeting the requirement, designing a multi-objective value function, wherein the multi-objective value function is specifically as follows: min S ═ α | z1|+β|M|,z1＝yd-yp1(ii) a Wherein S is an evaluation function; z is a radical of1Deviation of expected and actual yield; y isp1The seed percent of pass for actual obtaining; y isdThe expected coating pass rate is obtained; m is the drug seed ratio; alpha and beta are weight coefficients aiming at the deviation of the qualified rate and the actual drug-to-drug ratio respectively;
s22: carrying out unified normalization processing on the time-varying parameters to obtain optimally adjusted time-varying parameters, wherein a specific adjustment formula is as follows:
wherein the content of the first and second substances,
is the original parameter value, xiIs a normalized parameter; pLiAnd PUiRespectively a lower bound and an upper bound of the corresponding parameter;
s23: inputting the time-varying parameters subjected to normalization processing in the step S22 as a neural network model, outputting the seed qualification rate obtained actually, and setting the prediction output of the neural network;
s24: the artificial bee colony algorithm is adopted, so that the parameter adjustment is changed to a good direction, and the situation that the parameter is trapped in local optimization is avoided;
s25: and returning to the step S21 to calculate an evaluation function value according to the result fed back by each cycle and the expected coating qualification rate obtained in the step S23, and selecting an adjusting parameter according to the evaluation function value.
Further, in step S22, the time-varying parameter includes any one of the drug ratio, the main motor frequency, the batch duration, the drug supply duration, the pump delay drug supply, the discharge duration, the pump rated power and the coating quality.
Further, in step S25, the system adopts an avoidance maneuver, and sets a predetermined number of times for parameter variation to limit parameter optimization to fall into a dead loop.
The invention has the beneficial effects that:
the invention designs an intelligent control system which takes an expert system as a core and comprises a batch type seed coating machine and a coating quality image detection system. With the seed coating quality and the qualified rate as targets, the online optimization of the processing parameters of the coating machine is realized by constructing a coating parameter self-learning and intelligent mixed parameter optimization strategy system based on a neural network according to an experience knowledge base and a real-time feedback value, the automatic optimization and combination of the processing parameters aiming at different seeds are realized, the efficiency is improved, the coating agent dosage is saved, and the effect is obvious on the premise of meeting the coating quality and the qualified rate.
Drawings
FIG. 1 is a schematic view of the structure of the present invention.
FIG. 2 is a block diagram of the connection structure of the expert system of the present invention.
FIG. 3 is a flow chart of expert system control of the present invention.
FIG. 4 is a flow chart of the expert system parameter optimization principle of the present invention.
Fig. 5 is a schematic diagram of the neural network of the present invention.
In the figure, the number is 1, a control cabinet, 2, a display screen, 3, an image detection system, 4, a discharge port, 5, a rotary drum, 6, a throwing disc, 7, a feeding bin, 8, a storage bin, 9, a medicine conveying pipe, 10, a medicine conveying pump, 11, a medicine storage drum, 12, a rotary drum motor, 13, a throwing disc motor, 14, a PLC control module, 101, a man-machine interaction interface, 102, a knowledge base module, 103, an inference machine module, 104, a feedback self-learning module and 105, and an industrial personal computer.
Detailed Description
The invention will be further described with reference to the accompanying drawings and examples.
Referring to fig. 1-2, the expert control system based on the batch-type coating machine comprises a coating machine and an image detection system 3, wherein the coating machine is a batch-type coating machine and comprises a dosing system, a feeding system, a coating system and a PLC control module 14, and the PLC control module 14 is in signal connection with the dosing system, the feeding system and the coating system respectively and is used for controlling the coating machine to operate; the PLC control module 14 mainly includes PLC components and touch screens, and preferably, the PLC control module selected in this embodiment is siemens S7-200 type, so that the PLC control module controls the working conditions in the coating machine, and the PLC control module collects and monitors data, and the like, which are well known to those skilled in the art; the image detection system 3 is arranged at one side of the coating system, the signal input end of the image detection system 3 is in signal connection with the signal output end of the PLC control module 14, the image detection system is used for detecting the coating quality of seeds and feeding back the coating qualification rate and the coating quality in real time, the specific structure of the image detection system is as that of the seed coating qualification rate detection system disclosed in the patent document CN201820897736.7, and the image detection system specifically comprises a feeding and discharging system, an image acquisition system, an image processing control system and the like, the feeding and discharging system comprises a vibrating disc, a feeding guide groove, a discharging guide rail, a discharging electromagnetic valve, an air compressor and the like, the vibrating disc is provided with a hopper for containing the coated seeds, the coated seeds are discharged in sequence under the vibration of the vibrating disc and enter a detection box through the feeding guide groove, when the detection is finished, the discharging electromagnetic valve is opened, and the high-pressure air of the air compressor blows, the seed enters the seed collecting box through the discharging guide rail; the image acquisition system consists of 4 industrial cameras and 4 light sources, the cameras and the light sources are arranged on a circumference with the detection box as the center of a circle and form an angle of 90 degrees with each other, when seeds enter the detection box, the 4 cameras take pictures simultaneously, and images are transmitted to the image processing control system; the image processing control system receives the image from the industrial camera, processes and calculates the image to obtain the coating quality of each seed and the coating qualification rate of a batch of seeds, and feeds the coating quality and the coating qualification rate back to the expert system; the expert control system based on the batch-type coating machine further comprises an expert system, wherein the expert system is arranged on one side of the image detection system and comprises a control cabinet 1 and a display screen 2, the display screen 2 is installed on the front side of the control cabinet 1, and a human-computer interaction interface 101, a knowledge base module 102, an inference machine module 103, a feedback self-learning module 104 and a control center are arranged in the control cabinet 1; the human-computer interaction interface can be used for a user to perform relevant operations such as parameter setting, data query and control operation through the interface; the knowledge base module can access and manage expert knowledge and experience for the inference engine to use, and has the functions of storage, editing, retrieval, modification, expansion and the like; the inference engine module carries out inference by decomposing the design requirements input by the user and continuously matching with rule conditions; the feedback self-learning module is used for making a proper adjustment rule and storing an optimal solution in real time; the control center is an industrial personal computer 105, the industrial personal computer 105 is respectively and electrically connected with a human-computer interaction interface 101, a feedback self-learning module 104 and a PLC control module 14 of the coating machine in a two-way mode, the human-computer interaction interface 101 is electrically connected with a knowledge base module 102 in an output mode, the feedback self-learning module 104 is electrically connected with an inference machine module 103 in an output mode, and the signal input end of the feedback self-learning module 104 is further in signal connection with the signal output end of the image detection system 3 and used for receiving the coating qualification rate fed back by the image detection system.
In the embodiment, the expert system works on the principle that the expert system controls the coating machine and the image detection system to operate, establishes an expert system knowledge base and an inference optimization rule, and realizes automatic optimization of processing parameters with result feedback. On the premise of ensuring the coating quality and the qualified rate, the processing parameters are optimized, the drug-to-seed ratio and the batch processing time are reduced, the dosage is saved, and the efficiency is improved. The system intelligently recommends processing parameters of the coating machine through a neural network model according to expected values of coating quality and qualification rate in combination with seed types and characteristic parameters, a user enters an automatic processing program after confirming or modifying, coating is carried out according to specified parameters, a feedback result is obtained through an image detection system, and the neural network model is further trained on line through the feedback result. And when the accuracy of the neural network model predicted value is within reasonable errors within the specified times, starting a coating optimization strategy based on an artificial bee colony algorithm, otherwise, adopting a rule optimization strategy, entering the next processing cycle, and stopping optimization until the actual processing result is within the reasonable field of the expected target. The neural network model is trained all the time in the operation process. The operation parameters and results of each time are recorded in the knowledge base, and the more accurate the model and the recommended processing parameters are along with the increase of the data volume of the knowledge base.
Referring to fig. 1, in the present embodiment, the dosing system is disposed at one side of the coating system, and includes a drug storage barrel 11, a drug delivery pump 10 and a drug delivery pipe 9, the drug storage barrel 11 is made of corrosion-resistant material and stores the seed coating agent, and the drug delivery pipe 9 is a corrosion-resistant plastic pipe; the feeding system is arranged at the upper part of the coating system and comprises a feeding bin 7 and a storage bin 8, seeds needing to be coated are stored in the internal space of the storage bin 8, and automatic feeding can be realized when the seeds are few; the coating system comprises a throwing disc 6, a rotary drum 5, a discharge hole 4, a throwing disc motor 13 and a rotary drum motor 12; one end of the medicine conveying pipe 9 is positioned in the medicine storage barrel 11, the other end of the medicine conveying pipe 9 is positioned above the throwing disc 6, a medicine conveying pump 10 is arranged on the medicine conveying pipe 9, and the medicine conveying pipe 9 conveys the seed coating agent in the medicine storage barrel 11 to the throwing disc 6 through the medicine conveying pump 10; the feeding bin 7 is positioned at the upper part of the drum 5, a bin 8 is arranged right above the feeding bin 7, and seeds are placed into the drum 5 through the bin 8; the throwing disc 6 is positioned at the upper part of the middle part in the rotary drum 5, the lower part of the throwing disc 6 is provided with a throwing disc motor 13, the throwing disc 6 is a hollow round table with a trapezoidal section and is connected with the throwing disc motor 13, the throwing disc 6 can rotate at high speed under the driving of the motor, and the medicament forms mist after entering and is mixed with seeds; a drum motor 12 is arranged below the disc throwing motor 13, the drum 5 is also a hollow round table with a trapezoidal section and is connected with the drum motor 12, the drum 5 can rotate at high speed under the driving of the motor, and seeds rotate together under the action of centrifugal force after entering and are mixed with the medicament to finish coating; the discharge port 4 is positioned on the side wall of the drum 5, when the coating is finished, the discharge port is opened, the coated seeds leave the drum and are conveyed away, the image detection system 3 is arranged at the end part of the discharge port 4, and the coated seeds enter a feeding disc of the image detection system to carry out coating quality detection; and the signal output end of the PLC control module 14 is respectively connected with the signal input ends of the medicine conveying pump 10, the disc throwing motor 13 and the drum motor 12.
In this embodiment, be equipped with weighing sensor in the feed bin 8, preferably, weighing sensor's model is Shenzhen south China Hengtai industry Limited's 108BA-150Kg type, weighing sensor's signal input part be connected with PLC control module 14's signal output part, when reaching the formula weight that adds of every batch, stop the blowing, when every batch processing begins, the storehouse door is opened in the feeding bin, puts down the seed and gets into the barrel.
The invention also discloses a control method of the expert control system based on the batch-type coating machine, which comprises the following specific steps of:
s1: starting and initializing an expert system and an image monitoring system;
s2: the expert system recommends each processing parameter of the coating machine through a neural network model according to the coating quality target value and the seed characteristic set by the user, and the processing parameters are confirmed or modified by the user;
s3: the coating machine receives the parameters confirmed or modified by the user and performs coating operation;
s4: the image detection system detects the coating quality and the qualification rate of the product according to the sampling period set by the expert system and feeds the coating quality and the qualification rate back to the expert system in real time for reasoning optimization;
s5: and (4) the expert system receives the result fed back in the step S4 to further train the neural network model on line, when the accuracy of the predicted value of the neural network model is within reasonable errors within the specified times, a coating optimization strategy based on the artificial bee colony algorithm is started, otherwise, the expert system returns to the step S2 to re-determine all processing parameters of the coating machine, and the optimization is stopped until the actual processing result is within the reasonable field of the expected target.
Referring to fig. 4, in the present embodiment, in the step S2, the fuzzy recommendation of the coating machine processing parameters is performed by the feedback self-learning module and the inference engine module of the expert system, and the specific steps are as follows:
s21: aiming at the requirement that the expected target meets the qualification rate, designing a multi-objective cost function, wherein the multi-objective cost function specifically comprises the following steps: min S ═ α | z1|+β|M|,z1＝yd-yp1(ii) a Wherein S is an evaluation function; z is a radical of1Deviation of expected and actual yield; y isp1The seed percent of pass for actual obtaining; y isdThe expected coating pass rate is obtained; m is the drug seed ratio; alpha and beta are weight coefficients aiming at the deviation of the qualified rate and the actual drug-to-seed ratio respectively, and the coefficients can be designed according to requirements;
s22: carrying out unified normalization processing on the time-varying parameters to obtain time-varying parameters which are optimized and adjusted, specifically adjustedThe whole formula is as follows:
wherein the content of the first and second substances,
is the original parameter value, xiIs a normalized parameter; pLiAnd PUiRespectively a lower bound and an upper bound of the corresponding parameter; the time-varying parameters comprise any one of the parameters of the drug species ratio, the frequency of a main motor, the batch time, the drug supply time, the delayed drug supply of a pump, the discharge time, the rated power of the pump and the coating quality;
s23: inputting the time-varying parameters subjected to the normalization processing in the step S22 as a neural network model, setting the predicted output of the neural network by taking the coating qualification rate as output, wherein the coating qualification rate is the actually obtained seed qualification rate in the step S21, and the predicted output of the neural network is the expected coating qualification rate in the step S21; the neural network model used in this embodiment is a BP neural network, see fig. 5, where the output of the hidden layer node is
The weighting from the input layer node to the hidden layer node is represented as Wih(ii) a The weighting from the hidden layer node to the output layer node is represented as Whj(ii) a Subscripts i, h and j respectively represent a certain input node, a hidden layer node and an output node, and superscript k represents a serial number of a training pair;
s24: when the algorithm enters a certain field, in order to further effectively improve parameter optimization, an artificial bee colony algorithm is adopted, so that parameter adjustment is changed to a good direction, and local optimization is avoided;
s25: and returning to the step S21 to calculate an evaluation function value according to the result fed back by each cycle and the expected coating qualification rate obtained in the step S23, and selecting an adjusting parameter according to the evaluation function value. In addition, since the parameter adjustment may generate illegal solutions, the system adopts an avoidance strategy, and meanwhile, the system is provided with a specified number of times aiming at parameter variation, so that the parameter optimization is limited to fall into a dead loop.
The specific operation process of the expert control system comprises the following steps:
after the system is started, a disc throwing motor and a drum motor are opened, feed inlets below a storage bin are opened in sequence until the batch weight is reached, the feed inlets are closed, seeds to be coated are placed in the storage bin, coating quality and qualified rate target values are set, related parameters of the seeds are determined, and when the seeds which need to be coated do not exist in the system, new varieties and characteristics can be added; after setting, the system gives out optimized coating parameters and a predicted coating result, recommended parameter values can be modified, meanwhile, related parameters such as image detection, coating operation and the like are set in the system, after the parameters are determined, the system enters an automatic coating state and operates according to the specified parameters, a drug delivery pump sprays coating agents to a throwing disc after seeds operate for a set time, the coating agents are atomized, the coating agents are mixed with the seeds for a set time and then are discharged from a discharge port, a part of the samples are taken to the image detection system, the image detection system feeds back coating quality and qualification rate in real time, a parameter optimization module writes the result into a knowledge base through self-learning and optimization, and adjusts processing parameters in real time to enter next batch processing until the set processing amount is reached.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various changes and modifications can be made without departing from the inventive concept of the present invention, and these changes and modifications are all within the scope of the present invention.
Claims (6)
1. An expert control system based on a batch-type coating machine comprises a coating machine and an image detection system (3), wherein the coating machine is a batch-type coating machine and consists of a dosing system, a feeding system, a coating system and a PLC (programmable logic controller) control module (14), and the PLC control module (14) is in signal connection with the dosing system, the feeding system and the coating system respectively and is used for controlling the coating machine to operate; one side of capsule system is located in image detection system (3), and the signal input part of image detection system (3) and the signal output part signal connection of PLC control module (14) for detect seed coating quality and feed back the coating qualification rate in real time, its characterized in that: the intelligent coating machine is characterized by further comprising an expert system, the expert system is arranged on one side of the image detection system and comprises a control cabinet (1) and a display screen (2), the display screen (2) is installed on the front face of the control cabinet (1), a man-machine interaction interface (101), a knowledge base module (102), an inference machine module (103), a feedback self-learning module (104) and a control center are arranged in the control cabinet (1), the control center is an industrial personal computer (105), the industrial personal computer (105) is respectively and electrically connected with the man-machine interaction interface (101), the feedback self-learning module (104) and a PLC control module (14) of the coating machine in a two-way mode, the man-machine interaction interface (101) is electrically connected with the knowledge base module (102) in an output mode, the feedback self-learning module (104) is electrically connected with the inference machine module (103) in an output mode, and a signal input end of the feedback self-learning module (104, the coating qualification rate is used for receiving the feedback of the image detection system;
the control method of the expert control system comprises the following specific steps:
s1: starting and initializing an expert system and an image monitoring system;
s2: the expert system recommends each processing parameter of the coating machine through a neural network model according to the coating quality target value and the seed characteristic set by the user, and the processing parameters are confirmed or modified by the user;
s3: the coating machine receives the parameters confirmed or modified by the user and performs coating operation;
s4: the image detection system detects the coating quality and the qualification rate of the product according to the sampling period set by the expert system and feeds the coating quality and the qualification rate back to the expert system in real time for reasoning optimization;
s5: the expert system receives the result fed back in the step S4 to further train a neural network model on line, when the accuracy of the neural network model prediction value is within reasonable errors within the specified times, a coating optimization strategy based on an artificial bee colony algorithm is started, otherwise, the expert system returns to the step S2 to re-determine each processing parameter of the coating machine, and the optimization is stopped until the actual processing result is within the reasonable field of the expected target;
in the step S2, fuzzy recommendation is performed on each processing parameter of the coating machine by a feedback self-learning module and an inference engine module of an expert system, and the specific steps are as follows:
s21: aiming at the requirement that the expected target meets the qualification rate, designing a multi-objective cost function, wherein the multi-objective cost function specifically comprises the following steps:
minS＝α|z1|+β|M|,z1＝yd-yp1wherein S is an evaluation function; z is a radical of1Deviation of expected and actual yield; y isp1The seed percent of pass for actual obtaining; y isdThe expected coating pass rate is obtained; m is the drug seed ratio; alpha and beta are weight coefficients aiming at the deviation of the qualified rate and the actual drug-to-drug ratio respectively;
s22: carrying out unified normalization processing on the time-varying parameters to obtain optimally adjusted time-varying parameters, wherein a specific adjustment formula is as follows:
s23: inputting the time-varying parameters subjected to normalization processing in the step S22 as a neural network model, outputting the seed qualification rate obtained actually, and setting the prediction output of the neural network;
s24: the artificial bee colony algorithm is adopted, so that the parameter adjustment is changed to a good direction, and the situation that the parameter is trapped in local optimization is avoided;
s25: and returning to the step S21 to calculate an evaluation function value according to the result fed back by each cycle and the expected coating qualification rate obtained in the step S23, and selecting an adjusting parameter according to the evaluation function value.
2. The expert control system based on the batch coater according to claim 1, wherein: the medicine feeding system is arranged on one side of the coating system and comprises a medicine storage barrel (11), a medicine conveying pump (10) and a medicine conveying pipe (9); the feeding system is arranged at the upper part of the coating system and comprises a feeding bin (7) and a storage bin (8); the coating system comprises a throwing disc (6), a rotary drum (5), a discharge hole (4), a throwing disc motor (13) and a rotary drum motor (12); one end of the medicine conveying pipe (9) is positioned in the medicine storage barrel (11), the other end of the medicine conveying pipe is positioned above the throwing disc (6), a medicine conveying pump (10) is arranged on the medicine conveying pipe (9), and the medicine conveying pipe (9) conveys the seed coating agent in the medicine storage barrel (11) to the throwing disc (6) through the medicine conveying pump (10); the feeding bin (7) is positioned at the upper part of the drum (5), a bin (8) is arranged right above the feeding bin (7), and seeds are placed into the drum (5) through the bin (8) by the feeding bin (7); the throwing disc (6) is positioned on the upper part of the middle part in the drum (5), a throwing disc motor (13) is arranged at the lower part of the throwing disc (6), the throwing disc (6) is electrically connected with the throwing disc motor (13), and the throwing disc motor (13) is controlled to drive the throwing disc motor to rotate; a drum motor (12) is arranged below the disc throwing motor (13), the drum (5) is electrically connected with the drum motor (12) and is driven to rotate by the control of the drum motor (12); the discharge port (4) is positioned on the side wall of the drum (5), and an image detection system (3) is arranged at the end part of the discharge port (4); and the signal output end of the PLC control module (14) is respectively connected with the signal input ends of the medicine conveying pump (10), the disc throwing motor (13) and the drum motor (12).
3. The expert control system based on the batch coater according to claim 2, wherein: and a weighing sensor is arranged in the stock bin (8), and the signal input end of the weighing sensor is connected with the signal output end of the PLC control module (14) and used for controlling the charging weight of each batch.
4. The expert control system based on the batch coater according to claim 2, wherein: the throwing disc (6) and the rotary drum (5) are both hollow round tables with trapezoidal sections.
5. The expert control system based on the batch coater according to claim 1, wherein: in step S22, the time-varying parameter includes any one of the drug ratio, the frequency of the main motor, the time of the batch, the time of the drug supply, the time of the delayed drug supply of the pump, the time of the discharge, the rated power of the pump, and the quality of the coating.
6. The expert control system based on the batch coater according to claim 1, wherein: in step S25, the system adopts an avoidance policy, sets a predetermined number of times for parameter variation, and restricts parameter optimization from falling into a dead loop.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910698404.5A CN110501983B (en) | 2019-07-31 | 2019-07-31 | Expert control system and control method based on batch-type coating machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910698404.5A CN110501983B (en) | 2019-07-31 | 2019-07-31 | Expert control system and control method based on batch-type coating machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110501983A CN110501983A (en) | 2019-11-26 |
| CN110501983B true CN110501983B (en) | 2021-03-26 |
Family
ID=68587054
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910698404.5A Active CN110501983B (en) | 2019-07-31 | 2019-07-31 | Expert control system and control method based on batch-type coating machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110501983B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20240012400A1 (en) * | 2020-08-28 | 2024-01-11 | Siemens Corporation | Failure prediction in surface treatment processes using artificial intelligence |
| CN113547162A (en) * | 2021-07-30 | 2021-10-26 | 武汉科技大学 | Circle shear parameter integrated control system and method |
Citations (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2324754Y (en) * | 1998-04-03 | 1999-06-23 | 王洪斌 | Spraying type seed dressing machine |
| CN2377808Y (en) * | 1999-07-23 | 2000-05-17 | 徐至仁 | Seed coating machine |
| CN101689044A (en) * | 2007-04-12 | 2010-03-31 | 西门子公司 | Method for computer-supported control and/or regulation of a technical system |
| CN101782771A (en) * | 2010-03-17 | 2010-07-21 | 东华大学 | Spinning process intelligent optimized design method based on immune neural network |
| CN201569892U (en) * | 2009-09-22 | 2010-09-01 | 赵武云 | Automatic control device of seed coater based on singlechip |
| CN102540879A (en) * | 2011-12-28 | 2012-07-04 | 北京工业大学 | Multi-target evaluation optimization method based on group decision making retrieval strategy |
| CN103444307A (en) * | 2012-05-28 | 2013-12-18 | 甘肃农业大学 | Electronic control system based on expert system of intelligent seed coating machine |
| CN103808509A (en) * | 2014-02-19 | 2014-05-21 | 华北电力大学（保定） | Fan gear box fault diagnosis method based on artificial intelligence algorithm |
| CN103843493A (en) * | 2013-12-03 | 2014-06-11 | 石河子市华农种子机械制造有限公司 | Large numerical control seed coating system |
| CN104299052A (en) * | 2014-10-12 | 2015-01-21 | 刘岩 | Building energy consumption prediction method based on artificial bee colony algorithm and neural network |
| CN104763999A (en) * | 2015-03-04 | 2015-07-08 | 内蒙古瑞特优化科技股份有限公司 | Power plant pulverized coal boiler combustion performance online optimizing method and system |
| CN104855002A (en) * | 2015-06-09 | 2015-08-26 | 湖南力本种业科技有限公司 | Batch continuous coating control method and batch continuous coating machine with seed coating premixing function |
| CN205596545U (en) * | 2016-04-26 | 2016-09-28 | 农业部南京农业机械化研究所 | Mechanism is broken up to peanut capsule blanking |
| CN107154042A (en) * | 2017-05-18 | 2017-09-12 | 湖南大学 | A kind of seed-coating machine visible detection method and device |
| CN107321276A (en) * | 2017-06-16 | 2017-11-07 | 天津科技大学 | A kind of controllable heat fusing fluid bed microcapsules preparation system of product quality and preparation method |
| CN107506565A (en) * | 2017-10-16 | 2017-12-22 | 安徽工业大学 | Method for building up based on neutral net alloy cast steel roll Expert System for Materials Design |
| CN108244105A (en) * | 2018-01-17 | 2018-07-06 | 安徽省华禾种业有限公司 | A kind of vegetable seeds coating agent and preparation method thereof |
| CN108464032A (en) * | 2015-10-16 | 2018-08-28 | 罗马大学 | The routing policy of node in underwater network and the method for re-transmission policy and its realization device are managed in a manner of adaptive and engagement |
| CN108491968A (en) * | 2018-03-17 | 2018-09-04 | 北京工业大学 | Based on agricultural product quality and safety emergency resources scheduling model computational methods |
| CN108593663A (en) * | 2018-06-11 | 2018-09-28 | 农业部南京农业机械化研究所 | A kind of seed pelleting qualification rate detecting system and method |
| CN108614415A (en) * | 2018-03-27 | 2018-10-02 | 苏州联点数据技术有限公司 | A kind of intelligence control system and control method of plant growth factor |
| CN109634098A (en) * | 2018-12-25 | 2019-04-16 | 江苏大学 | A kind of fattening house environment conditioning system and method |
| CN109781099A (en) * | 2019-03-08 | 2019-05-21 | 兰州交通大学 | A kind of navigation methods and systems of adaptive UKF algorithm |
| CN209085657U (en) * | 2017-08-02 | 2019-07-09 | 强力物联网投资组合2016有限公司 | For data gathering system related or industrial environment with chemical production technology |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6338013B1 (en) * | 1999-03-19 | 2002-01-08 | Bryan John Ruffner | Multifunctional mobile appliance |
| US6582516B1 (en) * | 1999-11-29 | 2003-06-24 | Gustafson, Llc | Continuous batch seed coating system |
| CN103484485B (en) * | 2003-07-02 | 2017-08-15 | Bp法人北美有限公司 | Dextranase, encode they nucleic acid and preparation and use their method |
| US8357503B2 (en) * | 2008-08-29 | 2013-01-22 | Bunge Oils, Inc. | Hydrolases, nucleic acids encoding them and methods for making and using them |
| CN107090445A (en) * | 2011-11-25 | 2017-08-25 | 诺维信公司 | The polynucleotides of polypeptide and coding said polypeptide with lysozyme activity |
| US9392739B2 (en) * | 2014-05-20 | 2016-07-19 | Usc, Llc | Batch seed coating apparatus |
| WO2016025334A1 (en) * | 2014-08-10 | 2016-02-18 | Carnegie Institution Of Washington | Photosynthetic acclimation and increased biomass production in fluctuating light environments |
| WO2017084560A1 (en) * | 2015-11-16 | 2017-05-26 | Novozymes A/S | Cellulase variants and polynucleotides encoding same |
| US10225974B2 (en) * | 2016-08-10 | 2019-03-12 | Usc, Llc | Batch seed coater with lift assembly |
| CN106529672A (en) * | 2016-11-09 | 2017-03-22 | 上海电机学院 | Selective neural network integration algorithm based on artificial bee colony algorithm |
-
2019
- 2019-07-31 CN CN201910698404.5A patent/CN110501983B/en active Active
Patent Citations (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2324754Y (en) * | 1998-04-03 | 1999-06-23 | 王洪斌 | Spraying type seed dressing machine |
| CN2377808Y (en) * | 1999-07-23 | 2000-05-17 | 徐至仁 | Seed coating machine |
| CN101689044A (en) * | 2007-04-12 | 2010-03-31 | 西门子公司 | Method for computer-supported control and/or regulation of a technical system |
| CN201569892U (en) * | 2009-09-22 | 2010-09-01 | 赵武云 | Automatic control device of seed coater based on singlechip |
| CN101782771A (en) * | 2010-03-17 | 2010-07-21 | 东华大学 | Spinning process intelligent optimized design method based on immune neural network |
| CN102540879A (en) * | 2011-12-28 | 2012-07-04 | 北京工业大学 | Multi-target evaluation optimization method based on group decision making retrieval strategy |
| CN103444307A (en) * | 2012-05-28 | 2013-12-18 | 甘肃农业大学 | Electronic control system based on expert system of intelligent seed coating machine |
| CN103843493B (en) * | 2013-12-03 | 2016-04-20 | 石河子市华农种子机械制造有限公司 | Large-sized numerical control seed pelleting system |
| CN103843493A (en) * | 2013-12-03 | 2014-06-11 | 石河子市华农种子机械制造有限公司 | Large numerical control seed coating system |
| CN103808509A (en) * | 2014-02-19 | 2014-05-21 | 华北电力大学（保定） | Fan gear box fault diagnosis method based on artificial intelligence algorithm |
| CN104299052A (en) * | 2014-10-12 | 2015-01-21 | 刘岩 | Building energy consumption prediction method based on artificial bee colony algorithm and neural network |
| CN104763999A (en) * | 2015-03-04 | 2015-07-08 | 内蒙古瑞特优化科技股份有限公司 | Power plant pulverized coal boiler combustion performance online optimizing method and system |
| CN104855002A (en) * | 2015-06-09 | 2015-08-26 | 湖南力本种业科技有限公司 | Batch continuous coating control method and batch continuous coating machine with seed coating premixing function |
| CN108464032A (en) * | 2015-10-16 | 2018-08-28 | 罗马大学 | The routing policy of node in underwater network and the method for re-transmission policy and its realization device are managed in a manner of adaptive and engagement |
| CN205596545U (en) * | 2016-04-26 | 2016-09-28 | 农业部南京农业机械化研究所 | Mechanism is broken up to peanut capsule blanking |
| CN107154042A (en) * | 2017-05-18 | 2017-09-12 | 湖南大学 | A kind of seed-coating machine visible detection method and device |
| CN107321276A (en) * | 2017-06-16 | 2017-11-07 | 天津科技大学 | A kind of controllable heat fusing fluid bed microcapsules preparation system of product quality and preparation method |
| CN209085657U (en) * | 2017-08-02 | 2019-07-09 | 强力物联网投资组合2016有限公司 | For data gathering system related or industrial environment with chemical production technology |
| CN107506565A (en) * | 2017-10-16 | 2017-12-22 | 安徽工业大学 | Method for building up based on neutral net alloy cast steel roll Expert System for Materials Design |
| CN108244105A (en) * | 2018-01-17 | 2018-07-06 | 安徽省华禾种业有限公司 | A kind of vegetable seeds coating agent and preparation method thereof |
| CN108491968A (en) * | 2018-03-17 | 2018-09-04 | 北京工业大学 | Based on agricultural product quality and safety emergency resources scheduling model computational methods |
| CN108614415A (en) * | 2018-03-27 | 2018-10-02 | 苏州联点数据技术有限公司 | A kind of intelligence control system and control method of plant growth factor |
| CN108593663A (en) * | 2018-06-11 | 2018-09-28 | 农业部南京农业机械化研究所 | A kind of seed pelleting qualification rate detecting system and method |
| CN109634098A (en) * | 2018-12-25 | 2019-04-16 | 江苏大学 | A kind of fattening house environment conditioning system and method |
| CN109781099A (en) * | 2019-03-08 | 2019-05-21 | 兰州交通大学 | A kind of navigation methods and systems of adaptive UKF algorithm |
Non-Patent Citations (1)
| Title |
|---|
| "基于专家系统的智能化种子包衣机控制系统研制";杨婉霞;《中国农机化学报》;20140131;第216-219页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110501983A (en) | 2019-11-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110501983B (en) | Expert control system and control method based on batch-type coating machine | |
| CN101907867B (en) | Self-tuning expert control method of burning trough point parameter based on working condition recognition | |
| CN106406083A (en) | Fuzzy control coal dressing method | |
| CN107673083B (en) | Screw material blanking device and its controller based on variable Rate study | |
| CN112327960B (en) | Intelligent control system for loosening and dampening equipment | |
| CN103173584A (en) | Blast furnace burden-distribution control system with self-learning control function | |
| CN100546791C (en) | Asphalt mixture stirring equipment cold aggregate feeding control method | |
| CN106418656A (en) | Method and device for controlling moisture in production of tobacco shred | |
| CN106771031A (en) | A kind of intelligent feeding equipment based on the monitoring of real-time cultivation water | |
| CN106155024B (en) | A kind of Anthracite coal Intelligent Control System of Gas-collector Pressure | |
| CN108934416A (en) | One kind being based on BP neural network combined harvester multi-parameter operation-control system and method | |
| CN109636152B (en) | Homogenization processing regulation and control method and system for chemical components of product | |
| CN1525153A (en) | Flexible measuring method for overflow particle size specification of ball mill grinding system | |
| KR102540899B1 (en) | Intelligent automatic feeding system for fish farms using artificial intelligence | |
| CN206269905U (en) | A kind of automatic weighing-batching system | |
| CN211672404U (en) | Device for adjusting air separation rate of tobacco flake air separator | |
| CN106937607A (en) | A kind of livestock-raising manually intelligent automatic charging device | |
| CN104068297A (en) | Full-automatic numerical control cold noodle machine | |
| CN208757390U (en) | A kind of solid material foods machine | |
| CN110412960A (en) | Into squeezing and wisdom production management method and system are squeezed based on the sugar refinery of cloud computing | |
| CN206363654U (en) | A kind of logistics laboratory | |
| CN206808582U (en) | A kind of intelligent tea production line | |
| CN108829049A (en) | A kind of method of liquid crystal glass base manufacture real time monitoring rate of feeding and inventory | |
| Vo | Development of insect production automation: Automated processes for the production of Black Soldier Fly (Hermetia illucens) | |
| TW201909093A (en) | Intelligent scheduling system and method thereof comprising a real-time data, a historical database, an analysis module, an automatic watering module, and a parameter feedback database |
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 |