CN116029604B - Cage-raised meat duck breeding environment regulation and control method based on health comfort level - Google Patents
Cage-raised meat duck breeding environment regulation and control method based on health comfort level Download PDFInfo
- Publication number
- CN116029604B CN116029604B CN202310076403.3A CN202310076403A CN116029604B CN 116029604 B CN116029604 B CN 116029604B CN 202310076403 A CN202310076403 A CN 202310076403A CN 116029604 B CN116029604 B CN 116029604B
- Authority
- CN
- China
- Prior art keywords
- information
- meat
- duck
- health
- index
- 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
- 235000013372 meat Nutrition 0.000 title claims abstract description 182
- 241000272525 Anas platyrhynchos Species 0.000 title claims abstract description 180
- 230000036541 health Effects 0.000 title claims abstract description 129
- 238000000034 method Methods 0.000 title claims abstract description 66
- 238000009395 breeding Methods 0.000 title claims abstract description 16
- 230000001488 breeding effect Effects 0.000 title claims abstract description 16
- 238000012544 monitoring process Methods 0.000 claims abstract description 63
- 241000272522 Anas Species 0.000 claims abstract description 52
- 238000012512 characterization method Methods 0.000 claims abstract description 51
- 230000007613 environmental effect Effects 0.000 claims abstract description 40
- 230000001276 controlling effect Effects 0.000 claims abstract description 10
- 230000001105 regulatory effect Effects 0.000 claims abstract description 10
- 238000013528 artificial neural network Methods 0.000 claims abstract description 7
- 230000005856 abnormality Effects 0.000 claims description 54
- 230000033001 locomotion Effects 0.000 claims description 35
- 238000004364 calculation method Methods 0.000 claims description 24
- 239000013598 vector Substances 0.000 claims description 22
- 230000036760 body temperature Effects 0.000 claims description 20
- 230000008859 change Effects 0.000 claims description 15
- 235000021050 feed intake Nutrition 0.000 claims description 12
- 238000004458 analytical method Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 10
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 238000009432 framing Methods 0.000 claims description 4
- 239000002131 composite material Substances 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000006399 behavior Effects 0.000 description 20
- 230000004927 fusion Effects 0.000 description 8
- 239000011159 matrix material Substances 0.000 description 7
- 238000011156 evaluation Methods 0.000 description 5
- XOFYZVNMUHMLCC-ZPOLXVRWSA-N prednisone Chemical compound O=C1C=C[C@]2(C)[C@H]3C(=O)C[C@](C)([C@@](CC4)(O)C(=O)CO)[C@@H]4[C@@H]3CCC2=C1 XOFYZVNMUHMLCC-ZPOLXVRWSA-N 0.000 description 5
- 238000010219 correlation analysis Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 3
- 208000012661 Dyskinesia Diseases 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 244000144977 poultry Species 0.000 description 2
- 230000000384 rearing effect Effects 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000012098 association analyses Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 244000144972 livestock Species 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000009374 poultry farming Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000002195 synergetic effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for regulating and controlling the breeding environment of cage-raised meat ducks based on health comfort, which belongs to the technical field of environment regulation and control and comprises the following steps: s1: acquiring monitoring information of the internal and external environments of a duck shed and health characterization information of meat ducks; s2: determining key indexes of health comfort level of the meat ducks according to the internal and external environment monitoring information of the duck shed and the health characterization information of the meat ducks; s3: determining the health comfort index of the meat ducks according to the health comfort key index of the meat ducks; s4: taking the internal and external environment monitoring information of the duck shed and the health comfort index of the meat duck as inputs of a neural network to obtain the internal environment information of the duck shed and the health comfort index predicted value of the meat duck at the next moment; s5: and carrying out environment regulation according to the environmental information in the duck shed and the meat duck health comfort index predicted value at the next moment. Compared with the prior art, the method takes the external environment information and meat duck information of the breeding device into consideration, so that the environment regulation and control reference information is more comprehensive and scientific, and the accuracy and reliability of the environment regulation and control are improved.
Description
Technical Field
The invention belongs to the technical field of environment regulation and control, and particularly relates to a cage-rearing meat duck cultivation environment regulation and control method based on health comfort.
Background
With the continuous perfection of the technology and equipment of the poultry farming industry, the raising mode of the cage-raised meat ducks is rapidly developed. Because the raising density of the meat ducks in the cage is high and the range of motion of the meat ducks is limited, the environmental conditions of the raising house become key constraint factors affecting the health and production performance of the meat ducks. Therefore, the control of the breeding environment of the meat ducks in the cage is of great significance.
At present, the poultry raising house environment regulation and control system is a regulation and control method based on an environment parameter monitoring result, and whether the environment regulation and control system accords with the house environment requirement or not is judged through monitoring data of an environment sensor in the raising house. Zhang Longai et al devised an air conditioner control method, control device and system for cultivation, which can realize automatic control of the air environment of a cultivation place, but can only regulate and control air and air quantity. Zhou Feng et al propose an intelligent multilayer three-dimensional meat duck breeding system based on big data, contain environmental monitoring and environmental regulation and control module, carry out environmental regulation and control according to detecting the environmental change, do not consider the regulation hysteresis quality of environmental regulation and control equipment, lead to environmental regulation and control's instantaneity and accuracy not enough. Li Baoming et al propose a temperature prediction control system and a control method thereof for livestock and poultry housing cultivation environment, which are based on a temperature prediction model to carry out fuzzy control on the temperature of the cultivation environment, so that the instantaneity and adaptability of temperature regulation of the cultivation housing are improved, but the temperature regulation model is not optimized for animal comfort, and the scientificity and adaptability are deficient. Therefore, a method for fusing multiple environmental parameters and monitoring information of meat ducks is needed, and the real-time performance, accuracy, scientificity and adaptability of the environment regulation of the breeding house are improved.
Disclosure of Invention
The invention aims to solve the problems of insufficient scientificity and adaptability and low real-time property and accuracy of the existing environment regulation and control, and provides a cage-raising meat duck breeding environment regulation and control method based on health comfort.
The technical scheme of the invention is as follows: the method for regulating and controlling the breeding environment of the cage-raised meat ducks based on the health comfort level comprises the following steps of:
s1: acquiring monitoring information of the internal and external environments of a duck shed and health characterization information of meat ducks;
s2: determining key indexes of health comfort level of the meat ducks according to the internal and external environment monitoring information of the duck shed and the health characterization information of the meat ducks;
s3: determining the health comfort index of the meat ducks according to the health comfort key index of the meat ducks;
s4: taking the internal and external environment monitoring information of the duck shed and the health comfort index of the meat duck as inputs of a neural network to obtain the internal environment information of the duck shed and the health comfort index predicted value of the meat duck at the next moment;
s5: and carrying out environment regulation according to the environmental information in the duck shed and the meat duck health comfort index predicted value at the next moment.
Further, in step S1, the meat duck health characterization information includes behavior abnormality awareness information, feed intake abnormality awareness information, motion abnormality awareness information, sound abnormality awareness information, and body temperature abnormality awareness information;
the specific method for acquiring the behavior abnormality awareness information and the feed intake abnormality awareness information comprises the following steps: acquiring duck shed videos, acquiring meat duck images from the duck shed videos by using a self-adaptive threshold method, and taking the meat duck images as input of a convolutional neural network to obtain behavior abnormality awareness information and feed intake abnormality awareness information;
the specific method for acquiring the voice abnormality awareness information comprises the following steps: collecting meat duck sound data, sequentially carrying out pre-emphasis, framing, windowing, fourier transformation and Mel frequency cepstrum coefficient solving on the meat duck sound data to obtain sound feature vectors as sound abnormality awareness information;
the specific method for acquiring the motion abnormality awareness information comprises the following steps: collecting meat duck motion data, extracting multidimensional features of the meat duck motion data by using a signal time domain analysis method, a signal frequency domain analysis method and a signal time-frequency domain analysis method to obtain motion feature vectors, and fusing the sound feature vectors and the motion feature vectors to obtain fused feature vectors serving as motion abnormality awareness information;
the specific method for acquiring the body temperature abnormality awareness information comprises the following steps: and acquiring meat duck body temperature data, and taking the body temperature data which is out of the set temperature threshold range as body temperature abnormality awareness information.
Further, step S2 comprises the sub-steps of:
s21: the method comprises the steps of carrying out time sequence pairing on internal and external environment monitoring information of a duck shed and meat duck health characterization information, adding time information, taking a set of the internal and external environment monitoring information of the duck shed as an environment monitoring transaction library, taking a set of the meat duck health characterization information as a meat duck health characterization transaction library, and taking a set of the environment monitoring transaction library, the meat duck health characterization transaction library and the time information as a total transaction library;
s22: calculating the transaction support and the transaction confidence according to the total transaction library;
s23: calculating the transaction promotion degree according to the transaction support degree and the transaction confidence degree;
s24: and taking the data with the transaction lifting degree larger than the set lifting degree threshold value as a meat duck health comfort key index.
Further, in step S22, the transaction Support degree Support (X i ,Y i ) The calculation formula of (2) is as follows:
wherein P (X) i ,Y i ) The representation comprises X i Inside and outside environment monitoring information and Y of duck shed in state i The proportion of the transactions of the meat duck health characterization information of the state to the total transaction library is number (X i ,Y i ) The representation comprises X i Inside and outside environment monitoring information and Y of duck shed in state i The transaction number of the meat duck health characterization information of the state, number (ALLSamples), represents the transaction number in the total transaction library;
in step S22, the transaction Confidence (X i →Y i ) The calculation formula of (2) is as follows:
wherein P (X) i 〡Y i ) X represents the monitoring information of the internal and external environments of the duck shed in the total transaction library i Inclusion in transactions of statesMeat duck health characterization information is Y i Ratio of states, P (X i ) X represents the monitoring information of the internal and external environments of the duck shed i Probability of state;
in step S23, the transaction Lift degree Lift (X i ,Y i ) The calculation formula of (2) is as follows:
wherein P (Y) i ) Representing that meat duck health characterization information is Y i Probability of state.
Further, step S3 comprises the sub-steps of:
s31: taking the internal and external environment monitoring information of the duck shed and the health characterization information of the meat ducks as element layers, taking key indexes of health comfort level of the meat ducks as index layers, and calculating subjective weights according to the element layers and the index layers;
s32: calculating objective weights according to the element layer and the index layer;
s33: according to the subjective weight and the objective weight, calculating the comprehensive subjective and objective weight;
s34: and calculating the health comfort index of the meat ducks according to the comprehensive subjective weight.
Further, in step S31, subjective weight μ 1 The calculation formula of (2) is as follows:
wherein U represents the number of element layers, C represents the number of index layers, and omega represents the weight of each element;
in step S32, normalizing the key indexes of the health and comfort level of the meat ducks, calculating the comprehensive data value of the ith index after normalization, calculating a variation coefficient according to the comprehensive data value of the ith index, and calculating objective weight according to the variation coefficient; wherein the integrated data value of the ith indexCoefficient of variation b i And objective weight mu 2 The calculation formulas of (a) are respectively as follows:
wherein t represents the number of key indexes, Z il First data representing the ith key index, x ij The parameter value of the ith index belonging to the jth class of key indexes in the t key indexes is represented,a composite data value representing the j-th indicator;
in step S33, the subjective and objective weights β are integrated i The calculation formula of (2) is as follows:
wherein alpha is i An experience factor representing the i-th index,subjective weight for the i-th index, < +.>Objective weight for the i-th index;
in step S34, the calculation formula of the health comfort index of the meat duck is:
wherein Z is i Indicating the i-th key index.
Further, in step S5, the specific method for performing the environmental control is as follows: calculating the deviation and the deviation change rate of the environmental information in the duck shed and the preset environmental information at the next moment, and performing environmental regulation by taking the meat duck health comfort index predicted value as a constraint condition when the deviation or the deviation change rate is larger than a set threshold value.
The beneficial effects of the invention are as follows:
(1) According to the invention, the environment monitoring information and the meat duck monitoring information are obtained in a synergistic manner, the meat duck monitoring information is subjected to health awareness processing to obtain the meat duck health characterization information, and the multisource information is fused efficiently by combining the correlation analysis method;
(2) The meat duck health comfort assessment process adopted by the invention is to upgrade the comprehensive environment assessment system of the existing breeding house, bring the meat duck health characterization information into the assessment index, and the fusion of subjective and objective weights enables the weights of the indexes to be more scientific and strict, thereby improving the accuracy and reliability of dynamic prediction of environmental information and health comfort;
(3) The meat duck breeding environment regulation and control method provided by the invention is formulated by taking the meat duck health comfort degree prediction index as a constraint on the basis of environment parameter prediction, and compared with a regulation and control mode based on an environment parameter detection result in the prior art, the meat duck breeding environment regulation and control method is more scientific and accurate and has higher instantaneity.
Drawings
FIG. 1 is a flow chart of a method for regulating and controlling the breeding environment of cage-raised meat ducks;
fig. 2 is a structural diagram of a system corresponding to the method for regulating and controlling the breeding environment of the cage-raised meat ducks.
Detailed Description
Embodiments of the present invention are further described below with reference to the accompanying drawings.
As shown in fig. 1, the invention provides a method for regulating and controlling the breeding environment of cage-raised meat ducks based on health comfort, which comprises the following steps:
s1: acquiring monitoring information of the internal and external environments of a duck shed and health characterization information of meat ducks;
s2: determining key indexes of health comfort level of the meat ducks according to the internal and external environment monitoring information of the duck shed and the health characterization information of the meat ducks;
s3: determining the health comfort index of the meat ducks according to the health comfort key index of the meat ducks;
s4: taking the internal and external environment monitoring information of the duck shed and the health comfort index of the meat duck as inputs of a neural network to obtain the internal environment information of the duck shed and the health comfort index predicted value of the meat duck at the next moment;
s5: and carrying out environment regulation according to the environmental information in the duck shed and the meat duck health comfort index predicted value at the next moment.
In step S1, health detection processing is performed on meat duck monitoring information according to a time sequence to obtain meat duck health characterization information (feed intake, movement track, behavior, etc.), and the environment information and the meat duck health characterization information are paired in a time sequence and added with time information.
In step S2, data fusion is carried out on the environment monitoring information and the meat duck health characterization information obtained in step S1 according to the time information, then a relation between fusion data is mined by using a correlation analysis algorithm, a strong correlation rule between multi-source information is found out, and key indexes are extracted to serve as important indexes established by a meat duck health comfort assessment system.
In the step S3, inputting the important indexes obtained in the step S2 into a meat duck health comfort level assessment system to obtain a current meat duck health comfort level index, performing time sequence pairing on environment information and the meat duck health comfort level, and adding time information.
In step S4, a neural network technology is adopted to establish a prediction model, and environmental monitoring information and a meat duck health comfort index time sequence are input into the neural network to obtain environmental information and a meat duck health comfort dynamic prediction value in the next time period.
In step S5, the environmental information predicted value obtained in step S4 is compared with the appropriate environmental information to obtain the deviation and the deviation change rate. Inputting the deviation of the predicted value of the environmental information and the deviation change rate into a fuzzy control model to obtain fuzzy control quantity of each environmental information of the farm, and making an environmental regulation and control device regulation and control rule by taking the predicted value of the health comfort level of the meat ducks as a constraint condition and combining a qualitative decision method to form a cage-rearing meat ducks rearing environmental regulation and control decision.
In the embodiment of the invention, in step S1, the meat duck health characterization information comprises behavior abnormality awareness information, feed intake abnormality awareness information, motion abnormality awareness information, sound abnormality awareness information and body temperature abnormality awareness information;
the specific method for acquiring the behavior abnormality awareness information and the feed intake abnormality awareness information comprises the following steps: acquiring duck shed videos, acquiring meat duck images from the duck shed videos by using a self-adaptive threshold method, and taking the meat duck images as input of a convolutional neural network to obtain behavior abnormality awareness information and feed intake abnormality awareness information;
the specific method for acquiring the voice abnormality awareness information comprises the following steps: collecting meat duck sound data, sequentially carrying out pre-emphasis, framing, windowing, fourier transformation and Mel frequency cepstrum coefficient solving on the meat duck sound data to obtain sound feature vectors as sound abnormality awareness information;
the specific method for acquiring the motion abnormality awareness information comprises the following steps: collecting meat duck motion data, extracting multidimensional features of the meat duck motion data by using a signal time domain analysis method, a signal frequency domain analysis method and a signal time-frequency domain analysis method to obtain motion feature vectors, and fusing the sound feature vectors and the motion feature vectors to obtain fused feature vectors serving as motion abnormality awareness information;
extracting multidimensional features by comprehensively using signal time domain, frequency domain and time frequency domain analysis methods, constructing a motion feature vector representing the behavior of the meat ducks, constructing a depth fusion network through stacking self-encoders to enable the sound features and the motion features to be alternately and optimally fused, obtaining sound-motion fusion feature vectors, and obtaining motion abnormality awareness information by using field self-adaptive theory;
the specific method for acquiring the body temperature abnormality awareness information comprises the following steps: and acquiring meat duck body temperature data, and taking the body temperature data which is out of the set temperature threshold range as body temperature abnormality awareness information.
And (3) segmenting the meat duck image by applying a self-adaptive threshold method to the accessed video stream data, so as to realize rapid and accurate extraction of the meat duck image region. On the basis of meat duck image segmentation, a Kalman filtering algorithm is combined to track meat duck targets. Firstly, a complete moving target area is found by using the proposed self-adaptive threshold meat duck image segmentation algorithm, each contour is traversed, the coordinates of a centroid point are calculated, and the dynamic meat duck target behavior tracking is realized by adopting an improved algorithm combining Kalman filtering and a foreground target segmentation algorithm to track in real time. And training and establishing a convolutional artificial neural network model aiming at the meat duck cage culture environment based on the deep learning convolutional neural network to obtain information such as meat duck behavior anomaly detection, feed intake anomaly detection, motion anomaly detection and the like.
Amplifying the accessed meat duck sound data, analyzing the collected meat duck sound characteristics, selecting a proper filtering mode, and determining the position of the meat duck which emits abnormal sound by combining frequency discrimination and a positioning algorithm. And designing acoustic features suitable for meat duck sound data based on the MFCC (Mel frequency cepstrum coefficient), and obtaining sound feature vectors representing meat duck behaviors through data pre-emphasis, framing, windowing, fast Fourier transform and MFCC solving. And analyzing the curve fluctuation of the meat duck movement behaviors in X, Y, Z triaxial directions by utilizing the meat duck movement behavior data, comprehensively utilizing signal time domain, frequency domain and time-frequency domain analysis methods to extract multidimensional features, and constructing movement feature vectors representing the meat duck behaviors. In order to fully exert the relevance and complementarity of the multi-source heterogeneous data, a depth fusion network is constructed through stacking self-encoders so as to lead the sound characteristics and the motion characteristics to be alternatively and optimally fused, and a sound-motion fusion characteristic vector is obtained. Aiming at the problem that the data set acquired by the sensor has unbalanced categories, namely the number of partial categories is far more than that of other categories, the adaptive theory of the application field is adopted, firstly, fusion feature vectors are normalized, a boundary loss function is defined, a spherical decision boundary center and a radius are adaptively determined by combining known behavior classification, and then, the decision boundary is determined to be used for meat duck behavior recognition, so that information such as behavior abnormality awareness, motion abnormality awareness, voice abnormality awareness and the like is obtained.
And the information such as behavior abnormality detection, feed intake abnormality detection, motion abnormality detection, sound abnormality detection, body temperature abnormality detection and the like is synthesized and used as the meat duck health characterization information.
In an embodiment of the present invention, step S2 comprises the sub-steps of:
s21: taking the collection of the internal and external environment monitoring information of the duck shed as an environment monitoring transaction library, and taking the collection of the meat duck health characterization information as a meat duck health characterization transaction library;
s22: calculating transaction support and transaction confidence according to the environment monitoring transaction library and the meat duck health characterization transaction library;
s23: calculating the transaction promotion degree according to the transaction support degree and the transaction confidence degree;
s24: and taking the data with the transaction lifting degree larger than the set lifting degree threshold value as a meat duck health comfort key index.
The Apriori-based multi-source information association analysis algorithm designed by the invention is applied in the step S2. Aiming at environment monitoring and meat duck health characterization multi-source information, the algorithm uses an iteration method of layer-by-layer searching to extract key indexes to establish a meat duck health comfort assessment system, and related concepts of the algorithm are as follows:
transaction: environmental monitoring data x with time information in the present invention i Meat duck health characterization data y i Respectively composed data set X i And Y i Referred to as transaction, transaction X i And Y i The data set is an environment monitoring transaction library X and a meat duck health characterization transaction library Y. The environmental monitoring information is divided into three ranges: normal, high and low, one of the transactions X i Can be expressed as:
[ … … with proper temperature, humidity and CO2 concentration]. Meat duck health characterization information is classified into three classes: normal, healthy and critical, one of the transactions Y i Can be expressed as: [ normal behavior, abnormal movement, normal sound and normal body temperature ]]。
k item sets: if transaction X i Contains k elements, then this transaction X is called i An event that is a set of k terms and that event X meets the minimum support threshold is referred to as a set of frequent k terms.
In the embodiment of the present invention, in step S22, the support degree refers to the probability that several transactions in the transaction library occur simultaneously; transaction Support Support (X) i ,Y i ) The calculation formula of (2) is as follows:
wherein P (X) i ,Y i ) Representing and containing monitoring information X of internal and external environments of duck shed i Meat-and-duck health characterization information Y i The transaction of (2) occupies the proportion of the transaction library, the number (X i ,Y i ) Representing and containing monitoring information X of internal and external environments of duck shed i Meat-and-duck health characterization information Y i Number (ALLSamples) represents the total number of transactions within the two transaction libraries;
in step S22, confidence pointers define association rules, P (X i 〡Y i ) Finger principal X i When occurring, transaction X i Push transaction Y i (“X i →Y i ") probability, transaction Confidence (X) i →Y i ) The calculation formula of (2) is as follows:
wherein P (X) i 〡Y i ) Indicating that the environment monitoring information is X i The health characterization information of the meat ducks is Y i Probability of P (X) i ) Indicating that the environment monitoring information is X i Probability;
in step S23, the degree of elevation means "X i →Y i Confidence of "Y i The ratio of the probability of occurrence; transaction promotion degree Lift (X) i ,Y i ) The calculation formula of (2) is as follows:
wherein P (Y) i ) Representing that meat duck health characterization information is Y i Probability.
Degree of elevation embodies X i And Y is equal to i When Lift (X) i ,Y i )>1, the greater the Lift value, the stronger the forward correlation, when Lift (X i ,Y i )<When 1, the smaller the Lift value, the weaker the negative correlation, lift (X) i ,Y i ) When=1, there is no correlation.
Calculating the support and the confidence of the environmental monitoring and meat duck health characterization transaction library, setting a minimum support threshold and a minimum confidence threshold, and searching and iterating layer by layer to find out a strong association rule (a rule meeting the minimum support threshold and the minimum confidence threshold at the same time) between multi-source information, for example: the key indexes (such as the temperature in the house, the humidity in the house and the CO2 concentration … …) are extracted from the key indexes as important indexes established by a meat duck health comfort assessment system, wherein the temperature in the house is suitable, the humidity in the house is suitable, the CO2 concentration is higher than … … ] → [ normal behavior, abnormal movement, normal sound and normal body temperature ].
In an embodiment of the present invention, step S3 comprises the sub-steps of:
s31: taking the internal and external environment monitoring information of the duck shed and the health characterization information of the meat ducks as element layers, taking key indexes of health comfort level of the meat ducks as index layers, and calculating subjective weights according to the element layers and the index layers;
s32: calculating objective weights according to the element layer and the index layer;
s33: according to the subjective weight and the objective weight, calculating the comprehensive subjective and objective weight;
s34: and calculating the health comfort index of the meat ducks according to the comprehensive subjective weight.
The invention designs a meat duck health comfort evaluation system. The evaluation system is applied in the step S3, and the whole evaluation system is divided into three layers, namely a target layer, an element layer and an index layer according to the result of the correlation analysis of the environment monitoring and the meat duck health characterization information in the step S2. The target layer isA meat duck health comfort index A; the element layer contains house internal environment monitoring information U 1 Environmental monitoring information U outside house 2 Meat-mixing duck health characterization information U 3 The method comprises the steps of carrying out a first treatment on the surface of the The index layer is a key index C, C= { C extracted after the correlation analysis of the environmental monitoring and meat duck monitoring information in the step S2 1 ,C 2 ,C 3 …C c And c is the number of key indexes. Meat duck health comfort assessment is:
subjective weight calculation: the element layer is used as a judgment basis to obtain a judgment matrix of the element layer, which is marked as PU 1 ，PU 2 ，PU 3 The method comprises the steps of carrying out a first treatment on the surface of the Taking the index layer as a judgment basis to obtain a judgment matrix of the index layer, and marking the judgment matrix as PC 1 ,PC 2 ,PC 3 ,…,PC c . The form of the judgment matrix P is as follows:
in the formula, for the judgment matrix PU, b of the element layer ij The importance degree of the element i relative to the element j is shown; judgment matrix PC, b for index layer ij The importance of index i with respect to index j is shown as follows:
the importance of the elements of each layer is compared with each other, and the weight of each element is calculated:
Pω＝λ max ω
wherein lambda is max Is the maximum feature root of R, ω= (ω) 1 ,ω 2 …,ω n ) T Is the weight vector of P. Calculating the element layer weight as omega by using the above formula U Index layer weight is omega C The weight of the index layer relative to the target layer is omega CA Comprehensive subjective weight isThe weight expressions are respectively as follows:
in the formula, u is the number of element layers, and c is the number of index layers. Will integrate subjective weightsRecorded as mu 1 。
Objective weight calculation: after all data of each key index are normalized, average statistics is carried out on each index, and the calculation process is as follows: constructing a decision matrix for meat duck health evaluation indexes, and marking as follows: x= (x ij ) m*t M is the number of key index types (the key index types of the invention comprise house internal environment information, house external environment information and meat duck health characterization information), and x is the number of the key index types ij And the parameter value of the ith index belonging to the jth (j is less than or equal to 1 and less than or equal to m) type key index in the t key indexes is represented. The larger the coefficient of variation value is, the larger the weight of the ith index between different evaluation orders can be represented.
Comprehensive subjective and objective weights: and correcting the subjective weight by using the variation coefficient.
Meat duck health comfort assessment: and (3) solving a comprehensive score F of the target layer by using a linear weighting method, namely, a meat duck health comfort index, wherein the higher the meat duck health comfort index is, the lower the possibility of abnormal health of the meat duck is.
In the embodiment of the present invention, in step S31, subjective weight is givenμ 1 The calculation formula of (2) is as follows:
wherein U represents the number of element layers, C represents the number of index layers, and omega represents the weight of each element;
in step S32, normalizing the key indexes of the health and comfort level of the meat ducks, calculating the comprehensive data value of the ith index after normalization, calculating a variation coefficient according to the comprehensive data value of the ith index, and calculating objective weight according to the variation coefficient; wherein the integrated data value of the ith indexCoefficient of variation b i And objective weight mu 2 The calculation formulas of (a) are respectively as follows:
wherein t represents the number of key indexes, Z il First data representing the ith key index, x ij The parameter value of the ith index belonging to the jth class of key indexes in the t key indexes is represented,a composite data value representing the j-th indicator;
in step S33, the subjective and objective weights β are integrated i The calculation formula of (2) is as follows:
wherein alpha is i An experience factor representing the i-th index,subjective weight for the i-th index, < +.>Objective weight for the i-th index;
in step S34, the calculation formula of the health comfort index of the meat duck is:
wherein Z is i Indicating the i-th key index.
In the embodiment of the present invention, in step S5, the specific method for performing environmental regulation is as follows: calculating the deviation and the deviation change rate of the environmental information in the duck shed and the preset environmental information at the next moment, and performing environmental regulation by taking the meat duck health comfort index predicted value as a constraint condition when the deviation or the deviation change rate is larger than a set threshold value.
And (3) comparing the environmental information predicted value obtained in the step (S4) with proper environmental information to obtain deviation E and deviation change rate delta E, and using the deviation E and the deviation change rate delta E together as input of an environmental regulation model. Taking the temperature monitoring index as an example, E=H-H 0 ，Wherein: h is a temperature predicted value; h 0 Is a suitable temperature in the house. Based on the fuzzy theory, the deviation E and the deviation change rate delta E are converted into a fuzzy control theory language value, and a specific conversion formula is as follows:
wherein X is a deviation value E and a change rate delta E, and a is the upper limit of a domain; b is the lower limit of the universe; y is the fuzzy language value corresponding to X.
The fuzzy theory field of the deviation of the temperature predicted value in the house is set as [ -3,3], which can be divided into 7 grades { -3, -2, -1,0,1,2,3}, and is a quantization process of the input value. Taking the fuzzy language value of the deviation: { nb=negative big, nm=negative, ns=negative small, z=0, pm=positive small, pm=median, pb=positive big }. Similarly, the rate of change of the deviation of the predicted value of the temperature in the house is divided into three grades of fuzzy language values (negative, medium and positive) according to the method; and (3) taking a fuzzy value of an output control quantity of the fuzzy control model: { nb=negative large, nm=negative, z=0, pm=positive, pb=positive }, the fuzzy output language value is converted into a fuzzy control amount of the temperature regulation apparatus. The specific rules of the environment regulation model are as follows: when the deviation E of the predicted value of the temperature in the house and the change rate delta E are large, the deviation should be eliminated as soon as possible; when they are small or zero, it is desirable to maintain the stability of the intra-house temperature predictions preferentially.
And finally, according to the fuzzy control quantity of the temperature regulation and control equipment output by the environment regulation and control model, the regulation and control rule of the temperature regulation and control equipment is formulated by taking the meat duck health comfort degree predicted value as a constraint condition qualitative analysis, so as to form a cage-rearing meat duck rearing environment regulation and control decision.
The cage-rearing meat duck cultivation environment regulation method is realized based on a cage-rearing meat duck cultivation environment regulation system and comprises an environment monitoring module, a meat duck monitoring module, an environment regulation module and a core processing module. The environment monitoring module acquires internal and external environment information (temperature, humidity, wind speed, illuminance, CO2 concentration, NH3 concentration, H2S concentration, PM2.5 concentration, PM10 concentration, external temperature, external humidity, external wind speed and external illuminance) of the duck house in real time by adopting an environment monitoring sensor, and acquires meat duck video, sound, movement and body temperature information in real time by adopting an RGB camera, a sound sensor, a movement sensor and a body temperature sensor, wherein the environment regulation module comprises an environment regulation model and environment regulation equipment, and the core processing module comprises a meat duck health comfort assessment system and a prediction model.
Those of ordinary skill in the art will recognize that the embodiments described herein are for the purpose of aiding the reader in understanding the principles of the present invention and should be understood that the scope of the invention is not limited to such specific statements and embodiments. Those of ordinary skill in the art can make various other specific modifications and combinations from the teachings of the present disclosure without departing from the spirit thereof, and such modifications and combinations remain within the scope of the present disclosure.
Claims (4)
1. The method for regulating and controlling the breeding environment of the cage-raised meat ducks based on the health comfort level is characterized by comprising the following steps of:
s1: acquiring monitoring information of the internal and external environments of a duck shed and health characterization information of meat ducks;
in the step S1, meat duck health characterization information comprises behavior abnormality awareness information, feed intake abnormality awareness information, motion abnormality awareness information, sound abnormality awareness information and body temperature abnormality awareness information;
the specific method for acquiring the behavior abnormality awareness information and the feed intake abnormality awareness information comprises the following steps: acquiring duck shed videos, acquiring meat duck images from the duck shed videos by using a self-adaptive threshold method, and taking the meat duck images as input of a convolutional neural network to obtain behavior abnormality awareness information and feed intake abnormality awareness information;
the specific method for acquiring the voice abnormality awareness information comprises the following steps: collecting meat duck sound data, sequentially carrying out pre-emphasis, framing, windowing, fourier transformation and Mel frequency cepstrum coefficient solving on the meat duck sound data to obtain sound feature vectors as sound abnormality awareness information;
the specific method for acquiring the motion abnormality awareness information comprises the following steps: collecting meat duck motion data, extracting multidimensional features of the meat duck motion data by using a signal time domain analysis method, a signal frequency domain analysis method and a signal time-frequency domain analysis method to obtain motion feature vectors, and fusing the sound feature vectors and the motion feature vectors to obtain fused feature vectors serving as motion abnormality awareness information;
the specific method for acquiring the body temperature abnormality awareness information comprises the following steps: acquiring meat duck body temperature data, and taking the body temperature data which is out of a set temperature threshold range as body temperature abnormality detection information;
s2: determining key indexes of health comfort level of the meat ducks according to the internal and external environment monitoring information of the duck shed and the health characterization information of the meat ducks;
s21: the method comprises the steps of carrying out time sequence pairing on internal and external environment monitoring information of a duck shed and meat duck health characterization information, adding time information, taking a set of the internal and external environment monitoring information of the duck shed as an environment monitoring transaction library, taking a set of the meat duck health characterization information as a meat duck health characterization transaction library, and taking a set of the environment monitoring transaction library, the meat duck health characterization transaction library and the time information as a total transaction library;
s22: calculating the transaction support and the transaction confidence according to the total transaction library;
s23: calculating the transaction promotion degree according to the transaction support degree and the transaction confidence degree;
s24: taking the data with the transaction lifting degree larger than the set lifting degree threshold value as a meat duck health comfort key index;
s3: determining the health comfort index of the meat ducks according to the health comfort key index of the meat ducks;
said step S3 comprises the sub-steps of:
s31: taking the internal and external environment monitoring information of the duck shed and the health characterization information of the meat ducks as element layers, taking key indexes of health comfort level of the meat ducks as index layers, and calculating subjective weights according to the element layers and the index layers;
s32: calculating objective weights according to the element layer and the index layer;
s33: according to the subjective weight and the objective weight, calculating the comprehensive subjective and objective weight;
s34: calculating the health comfort index of the meat ducks according to the comprehensive subjective weight;
s4: taking the internal and external environment monitoring information of the duck shed and the health comfort index of the meat duck as inputs of a neural network to obtain the internal environment information of the duck shed and the health comfort index predicted value of the meat duck at the next moment;
s5: and carrying out environment regulation according to the environmental information in the duck shed and the meat duck health comfort index predicted value at the next moment.
2. According to claimThe method for regulating and controlling the environment for raising meat ducks in cages based on health comfort level as set forth in claim 1, wherein in the step S22, transaction support is providedSupport(X i ,Y i ) The calculation formula of (2) is as follows:
in the method, in the process of the invention,P(X i ,Y i ) The representation comprisesX i Inside and outside environment monitoring information of duck house of stateY i The transactions of the meat duck health characterization information of the state account for the proportion of the total transaction library,number(X i ,Y i ) The representation comprisesX i Inside and outside environment monitoring information of duck house of stateY i The number of transactions of the meat duck health characterization information of the status,number(ALLSamples) Representing the number of transactions in the total transaction library;
in the step S22, transaction confidenceConfidence(X i →Y i ) The calculation formula of (2) is as follows:
in the method, in the process of the invention,P(X i Y i ) The monitoring information of the internal and external environments of the duck shed in the total transaction library is expressed asX i The status transaction comprises meat duck health characterization information as followsY i The ratio of the states is such that,P(X i ) The monitoring information of the internal and external environments of the duck shed is shown asX i Probability of state;
in the step S23, the transaction promotion degreeLift(X i ,Y i ) The calculation formula of (2) is as follows:
in the method, in the process of the invention,P(Y i ) The health characterization information of the meat ducks is expressed asY i Probability of state.
3. The method for regulating and controlling the environment of raising meat ducks in cages based on health comfort level as described in claim 1, wherein in the step S31, subjective weight is givenμ 1 The calculation formula of (2) is as follows:
in the method, in the process of the invention,Uthe number of the element layer elements is represented,Cthe number of the elements of the index layer is represented,ωrepresenting the weight of each element;
in the step S32, the key indexes of health and comfort level of the meat ducks are normalized, and the first normalization is calculatediThe integrated data value of each index according to the firstiCalculating a variation coefficient according to the comprehensive data value of each index, and calculating an objective weight according to the variation coefficient; wherein, the firstiComprehensive data value of individual indexCoefficient of variationb i Objective weightμ 2 The calculation formulas of (a) are respectively as follows:
in the method, in the process of the invention,tthe number of the key indexes is represented,Z il represent the firstiFirst of key indexlThe data of the plurality of data,x ij representation oftThe key index belongs tojClass key index ofiThe parameter values of the individual indicators are set,represent the firstjA composite data value for each index;
in the step S33, subjective and objective weights are integratedβ i The calculation formula of (2) is as follows:
in the method, in the process of the invention,α i represent the firstiThe empirical factor of the individual indicators is used,is the firstiSubjective weight of individual index,/->Is the firstiObjective weights of the individual indicators;
in the step S34, the calculation formula of the health comfort index of the meat duck is as follows:
in the method, in the process of the invention,Z i represent the firstiKey indexes.
4. The method for regulating and controlling the environment of cage-rearing meat ducks based on health comfort level according to claim 1, wherein in the step S5, the specific method for regulating and controlling the environment is as follows: calculating the deviation and the deviation change rate of the environmental information in the duck shed and the preset environmental information at the next moment, and performing environmental regulation by taking the meat duck health comfort index predicted value as a constraint condition when the deviation or the deviation change rate is larger than a set threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310076403.3A CN116029604B (en) | 2023-02-03 | 2023-02-03 | Cage-raised meat duck breeding environment regulation and control method based on health comfort level |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310076403.3A CN116029604B (en) | 2023-02-03 | 2023-02-03 | Cage-raised meat duck breeding environment regulation and control method based on health comfort level |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116029604A CN116029604A (en) | 2023-04-28 |
| CN116029604B true CN116029604B (en) | 2024-01-23 |
Family
ID=86081098
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310076403.3A Active CN116029604B (en) | 2023-02-03 | 2023-02-03 | Cage-raised meat duck breeding environment regulation and control method based on health comfort level |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116029604B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116720713B (en) * | 2023-08-07 | 2023-12-19 | 广州城建职业学院 | Poultry cultivation management method, system and storage medium for animal husbandry |
| CN118550347A (en) * | 2024-05-14 | 2024-08-27 | 南京农业大学 | Environment regulation and control system and control method for farm |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105929691A (en) * | 2016-05-13 | 2016-09-07 | 东华大学 | Internal mode compensation control method based on fuzzy control |
| CN107483240A (en) * | 2017-08-07 | 2017-12-15 | 国网安徽省电力公司淮北供电公司 | Power communication network service health degree analysis method based on Internet resources incidence relation |
| CN113009952A (en) * | 2021-02-26 | 2021-06-22 | 安徽强英鸭业集团有限公司 | Automatic supervision system of cowshed is raised in meat duck cage |
| CN113283604A (en) * | 2021-05-25 | 2021-08-20 | 广东工业大学 | Multi-element situation self-adaptive reasoning method and application |
| CN113297272A (en) * | 2021-05-30 | 2021-08-24 | 福建中锐网络股份有限公司 | Bridge monitoring data association rule mining and health early warning method and system |
| CN114444910A (en) * | 2022-01-14 | 2022-05-06 | 华北电力大学 | Electric power Internet of things-oriented edge network system health degree evaluation method |
| CN115119775A (en) * | 2022-07-19 | 2022-09-30 | 广西华兴食品集团有限公司 | Intelligent multilayer three-dimensional meat duck farming systems based on big data |
| CN115167587A (en) * | 2022-08-04 | 2022-10-11 | 山东交通学院 | Self-learning-based henhouse breeding environment health monitoring control system and method |
| CN115375127A (en) * | 2022-08-16 | 2022-11-22 | 上海方联技术服务有限公司 | Intelligent refined management and control system for engineering construction |
-
2023
- 2023-02-03 CN CN202310076403.3A patent/CN116029604B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105929691A (en) * | 2016-05-13 | 2016-09-07 | 东华大学 | Internal mode compensation control method based on fuzzy control |
| CN107483240A (en) * | 2017-08-07 | 2017-12-15 | 国网安徽省电力公司淮北供电公司 | Power communication network service health degree analysis method based on Internet resources incidence relation |
| CN113009952A (en) * | 2021-02-26 | 2021-06-22 | 安徽强英鸭业集团有限公司 | Automatic supervision system of cowshed is raised in meat duck cage |
| CN113283604A (en) * | 2021-05-25 | 2021-08-20 | 广东工业大学 | Multi-element situation self-adaptive reasoning method and application |
| CN113297272A (en) * | 2021-05-30 | 2021-08-24 | 福建中锐网络股份有限公司 | Bridge monitoring data association rule mining and health early warning method and system |
| CN114444910A (en) * | 2022-01-14 | 2022-05-06 | 华北电力大学 | Electric power Internet of things-oriented edge network system health degree evaluation method |
| CN115119775A (en) * | 2022-07-19 | 2022-09-30 | 广西华兴食品集团有限公司 | Intelligent multilayer three-dimensional meat duck farming systems based on big data |
| CN115167587A (en) * | 2022-08-04 | 2022-10-11 | 山东交通学院 | Self-learning-based henhouse breeding environment health monitoring control system and method |
| CN115375127A (en) * | 2022-08-16 | 2022-11-22 | 上海方联技术服务有限公司 | Intelligent refined management and control system for engineering construction |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116029604A (en) | 2023-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116029604B (en) | Cage-raised meat duck breeding environment regulation and control method based on health comfort level | |
| CN110263846B (en) | Fault diagnosis method based on fault data deep mining and learning | |
| CN111753918B (en) | Gender bias-removed image recognition model based on countermeasure learning and application | |
| Suryo et al. | Improved time series prediction using LSTM neural network for smart agriculture application | |
| CN110909672A (en) | Smoking action recognition method based on double-current convolutional neural network and SVM | |
| CN115099296A (en) | Sea wave height prediction method based on deep learning algorithm | |
| Jeantet et al. | Fully convolutional neural network: A solution to infer animal behaviours from multi-sensor data | |
| CN116720620A (en) | Grain storage ventilation temperature prediction method based on IPSO algorithm optimization CNN-BiGRU-Attention network model | |
| CN112560948A (en) | Eye fundus map classification method and imaging method under data deviation | |
| CN113627240B (en) | Unmanned aerial vehicle tree species identification method based on improved SSD learning model | |
| CN117891702A (en) | Method for determining performance index weight of embedded computing platform based on maximum entropy model | |
| CN116664545A (en) | Offshore benthos quantitative detection method and system based on deep learning | |
| CN115471491A (en) | Live pig health state judgment method based on multi-target tracking and related equipment | |
| CN111860441B (en) | Video target identification method based on unbiased depth migration learning | |
| CN113420705B (en) | Semi-supervised signal analysis method based on bidirectional embedding characteristics | |
| CN115310559A (en) | Target identification method based on Bayes-Transformer neural network model | |
| Li et al. | Early drought plant stress detection with bi-directional long-term memory networks | |
| CN115293249A (en) | Power system typical scene probability prediction method based on dynamic time sequence prediction | |
| CN114998731A (en) | Intelligent terminal navigation scene perception identification method | |
| CN114723998A (en) | Small sample image classification method and device based on large-boundary Bayes prototype learning | |
| CN114492657A (en) | Plant disease classification method and device, electronic equipment and storage medium | |
| Sudhir et al. | Plant Disease Severity Detection and Fertilizer Recommendation using Deep Learning Techniques | |
| CN118393898B (en) | Intelligent control method and system for biological isolation device | |
| CN117932313B (en) | Method and system for predicting technical actions of kayak based on artificial intelligence | |
| CN118426318B (en) | Internet of things service self-adaptive system based on process mining |
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 |