CN118197635B - Beef cattle health detection data processing method - Google Patents
Beef cattle health detection data processing method Download PDFInfo
- Publication number
- CN118197635B CN118197635B CN202410606430.1A CN202410606430A CN118197635B CN 118197635 B CN118197635 B CN 118197635B CN 202410606430 A CN202410606430 A CN 202410606430A CN 118197635 B CN118197635 B CN 118197635B
- Authority
- CN
- China
- Prior art keywords
- body temperature
- predicted
- data
- beef cattle
- obtaining
- 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
- 235000015278 beef Nutrition 0.000 title claims abstract description 99
- 241000283690 Bos taurus Species 0.000 title claims abstract description 95
- 230000036541 health Effects 0.000 title claims abstract description 49
- 238000001514 detection method Methods 0.000 title claims abstract description 20
- 238000003672 processing method Methods 0.000 title claims abstract description 13
- 230000036760 body temperature Effects 0.000 claims abstract description 274
- 238000000034 method Methods 0.000 claims abstract description 36
- 238000011156 evaluation Methods 0.000 claims abstract description 31
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000007613 environmental effect Effects 0.000 claims description 8
- 238000005311 autocorrelation function Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 description 6
- 239000013598 vector Substances 0.000 description 6
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000009395 breeding Methods 0.000 description 4
- 230000001488 breeding effect Effects 0.000 description 4
- 201000010099 disease Diseases 0.000 description 4
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 208000035473 Communicable disease Diseases 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 230000001681 protective effect Effects 0.000 description 2
- 230000001932 seasonal effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 241000700605 Viruses Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000034994 death Effects 0.000 description 1
- 238000013399 early diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 244000144992 flock Species 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 230000012010 growth Effects 0.000 description 1
- 230000005802 health problem Effects 0.000 description 1
- 230000009545 invasion Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 230000033001 locomotion Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000005541 medical transmission Effects 0.000 description 1
- 230000004060 metabolic process Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000005180 public health Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
- A61B5/02055—Simultaneously evaluating both cardiovascular condition and temperature
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01K—ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
- A01K29/00—Other apparatus for animal husbandry
- A01K29/005—Monitoring or measuring activity, e.g. detecting heat or mating
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/70—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in livestock or poultry
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- Cardiology (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Environmental Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Physiology (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Animal Behavior & Ethology (AREA)
- Physics & Mathematics (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Primary Health Care (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Pulmonology (AREA)
- Biodiversity & Conservation Biology (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Epidemiology (AREA)
- Veterinary Medicine (AREA)
- Animal Husbandry (AREA)
Abstract
The invention relates to the technical field of data anomaly detection, in particular to a beef cattle health detection data processing method. The invention obtains an initial prediction order for executing a moving average prediction algorithm; forming a moving window, and obtaining predicted body temperature data at the next moment of the moving window according to the actual body temperature data and predicted influence weight at each moment in the moving window; obtaining a body temperature error evaluation value according to the difference between the predicted body temperature data and the corresponding actual body temperature data and the preset body temperature fluctuation parameter of the beef cattle; further judging whether the initial prediction order needs to be adjusted or not, and continuously adjusting until the initial prediction order does not need to be adjusted, so as to obtain an optimized prediction order; obtaining a predicted first body temperature; obtaining a predicted second body temperature of the beef cattle according to the predicted first body temperature, the distribution characteristics of the ambient temperature and the fluctuation characteristics of the heart rate data; and detecting the health of beef cattle. According to the method, the accuracy of predicting the body temperature data is improved by determining the proper prediction order, and the health of the beef cattle is judged in time.
Description
Technical Field
The invention relates to the technical field of data anomaly detection, in particular to a beef cattle health detection data processing method.
Background
With the expansion of market demands and breeding scales of beef cattle, the health condition and the growth environment of beef flocks are ensured, the risk of food pollution can be reduced, the production performance of beef cattle is improved, the economic cost is reduced, and meanwhile, the effective management of compliance with regulations and breeding plans is facilitated. However, when the beef cattle is infected with infectious diseases or external diseases, the whole breeding area is possibly infected when the threshold value is early-warned, the threshold value is not well controlled, serious economic loss is brought to the beef cattle industry, and serious hidden dangers are brought to the quality safety and public health safety of animal products, so that the detection of the health of the beef cattle is very important to discover and control the potential infectious diseases as soon as possible.
In the prior art, considering that the disease of beef cattle changes from body temperature, the beef cattle has obvious seasonal characteristics, the moving average prediction algorithm can be adopted to predict the health data, but under the condition of inaccurate prediction order, the beef cattle is easily affected by random noise in the data, so that the adaptability to the body temperature data is reduced, the body temperature cannot be accurately predicted, and the result of judging the health state of the beef cattle is poor.
Disclosure of Invention
In order to solve the technical problem that the result of beef cattle health detection is poor due to the fact that proper prediction orders cannot be determined, the invention aims to provide a beef cattle health detection data processing method, and the adopted technical scheme is as follows:
the invention provides a beef cattle health detection data processing method, which comprises the following steps:
acquiring health data of beef cattle at each moment in a history stage of the current moment; the health data comprises actual body temperature data and heart rate data of beef cattle;
performing trending polynomial fitting by taking the preset prediction order as the polynomial order for performing polynomial fitting on the actual body temperature data, and obtaining an initial prediction order for executing a moving average prediction algorithm according to the change trend of the body temperature data after trending polynomial fitting;
Forming a moving window by the initial prediction order actual body temperature data, and obtaining a reference judgment value in the moving window according to a preset prediction value corresponding to the actual body temperature data at each moment in the moving window; obtaining a difference value of each moment in the moving window according to the difference between the actual body temperature data and the reference judgment value of each moment in the moving window; obtaining a prediction influence weight of each moment in the moving window according to the change trend of the difference value of each moment in the moving window; obtaining predicted body temperature data at the next moment of the moving window according to the actual body temperature data at each moment in the moving window and the predicted influence weight;
Obtaining a body temperature error evaluation value according to the difference between the predicted body temperature data and the corresponding actual body temperature data and the preset body temperature fluctuation parameter of the beef cattle; judging whether the initial prediction order needs to be adjusted according to the body temperature error evaluation value, if so, performing trending polynomial fitting by taking the initial prediction order as the polynomial order for performing polynomial fitting on actual body temperature data to obtain a new initial prediction order until the initial prediction order does not need to be adjusted, and obtaining an optimized prediction order;
Obtaining a predicted first body temperature according to the optimized prediction order; acquiring the environmental temperature at the current moment, and acquiring a predicted second body temperature of the beef cattle according to the predicted first body temperature, the distribution characteristics of the environmental temperature and the fluctuation characteristics of heart rate data; and detecting the health of the beef cattle according to the predicted second body temperature.
Further, the method for obtaining the initial prediction order comprises the following steps:
Drawing an autocorrelation function diagram for body temperature data after trending polynomial fitting; the corresponding hysteresis period number when the autocorrelation function changes most negatively is used as the initial prediction order for executing the moving average prediction algorithm.
Further, the reference judgment value obtaining method includes:
And accumulating and averaging the preset predicted values at all the moments to obtain a reference judgment value.
Further, the method for obtaining the prediction influence weight comprises the following steps:
Obtaining the predicted impact weight according to an obtaining formula of the predicted impact weight, wherein the obtaining formula of the predicted impact weight is as follows:
; wherein, Representing the first in the moving windowPredicting influence weights at each moment; representing a differential mean; Representing the first in the moving window Differential values at each time instant; Representing the first in the moving window Differential values at each time instant; representing the magnitude of the initial prediction order.
Further, the method for acquiring the predicted body temperature data comprises the following steps:
And calculating the product of the actual body temperature data and the predicted influence weight at each moment in the moving window, and accumulating the weighted body temperature data in the moving window to obtain the predicted body temperature data at the next moment in the moving window as the weighted body temperature data at each moment.
Further, the method for acquiring the body temperature error evaluation value comprises the following steps:
calculating the difference between the predicted body temperature data and the corresponding actual body temperature data as a body temperature error;
And calculating the ratio between the body temperature error and the preset body temperature fluctuation parameter to obtain a body temperature error evaluation value.
Further, the determining whether the initial prediction order needs to be adjusted according to the body temperature error evaluation value includes:
If the body temperature error evaluation value is smaller than a preset first threshold value, the initial prediction order does not need to be continuously adjusted; if the body temperature error evaluation value is larger than a preset first threshold value, the initial prediction order is judged to need to be adjusted.
Further, the method for obtaining the predicted first body temperature comprises the following steps:
Acquiring predicted body temperature data of the next moment of the moving window corresponding to the optimized prediction order; and replacing the actual body temperature data in the corresponding moving window with the predicted body temperature data, and continuing moving average calculation until the actual body temperature data at all the historical moments participate in calculation, so as to obtain a final result mean value as a predicted first body temperature.
Further, the method for obtaining the predicted second body temperature comprises the following steps:
obtaining a predicted second body temperature according to an obtaining formula of the predicted second body temperature, wherein the obtaining formula of the predicted second body temperature is as follows:
; wherein, Indicating a predicted second body temperature; means for predicting a first body temperature; Representing the ambient temperature at the current time; representing the average temperature of the environment; Heart rate data representing beef cattle at the current moment; representing the normal average heart rate of beef cattle; maximum value representing beef cattle normal heart rate data; A minimum value representing normal heart rate data of the beef cattle; representing the normal heart rate fluctuation range of beef cattle; representing natural constants.
Further, the empirical value of the preset body temperature fluctuation parameter is 0.75.
The invention has the following beneficial effects:
In order to reduce the prediction error, the invention better knows the distribution and change of the data, forms a moving window by the actual body temperature data of the initial prediction order, obtains a reference judgment value of the actual body temperature data in the moving window, and reflects the centralized trend or distribution characteristics of the data; obtaining a differential value of actual body temperature data at each moment in a moving window, and reflecting the dynamic change condition of the data; obtaining the predicted influence weight of the actual body temperature data at each moment in the moving window, and reflecting the influence degree of the data at each moment on future prediction; according to the actual body temperature data and the predicted influence weight at each moment in the moving window, predicted body temperature data at the next moment in the moving window are obtained, and static and dynamic characteristics of the data are comprehensively considered, so that the prediction model can more accurately predict future data; obtaining a body temperature error evaluation value according to the difference between the predicted body temperature data and the corresponding actual body temperature data and the preset body temperature fluctuation parameter of the beef cattle, and evaluating the accuracy of the prediction model; judging whether the initial prediction order needs to be adjusted according to the body temperature error evaluation value, if so, performing trending polynomial fitting by taking the initial prediction order as the polynomial order for performing polynomial fitting on actual body temperature data to obtain a new initial prediction order, until the initial prediction order does not need to be adjusted, obtaining an optimized prediction order, continuously adjusting the initial prediction order, obtaining an optimal prediction order, and enabling a moving average prediction model to more accurately fit data; obtaining a predicted first body temperature according to the optimized prediction order; in order to avoid the influence caused by misjudgment as much as possible, comprehensively considering other various factors, comprehensively evaluating the health condition of the beef cattle, acquiring the environmental temperature at the current moment, and acquiring the predicted second body temperature of the beef cattle according to the predicted first body temperature, the distribution characteristics of the environmental temperature and the fluctuation characteristics of heart rate data; beef health is detected based on the difference between the predicted second body temperature and the corresponding actual body temperature data. According to the method, the accuracy of predicting the body temperature data is improved by determining the proper prediction order, and the health of the beef cattle is judged in time.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a beef cattle health detection data processing method according to an embodiment of the invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to specific implementation, structure, characteristics and effects of the beef cattle health detection data processing method according to the invention by combining the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the beef cattle health detection data processing method provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for processing beef cattle health detection data according to an embodiment of the invention is shown, where the method specifically includes:
step S1: acquiring health data of beef cattle at each moment in a history stage of the current moment; the health data includes actual body temperature data and heart rate data of the beef cattle.
In the embodiment of the invention, in order to avoid that protective measures cannot be timely taken before the beef cattle are infected, the health data of the beef cattle are acquired in real time, the related health data of the beef cattle are monitored by adopting the neck rings of the beef cattle, and the health data of the beef cattle at each moment in the history stage of the current moment are acquired, wherein the health data comprise the actual body temperature data and the heart rate data of the beef cattle. In order to facilitate the processing of the subsequent data, the difference between the units and the magnitude of the numerical values among the health data is avoided, and the subsequent calculation of the index does not consider the dimension.
It should be noted that, in one embodiment of the present invention, the historical stage of the health data is 24 hours, and the time interval between each acquisition is 5S-10S; in other embodiments of the present invention, the historical stage of the health data and the size of the time interval may be specifically set by the operator according to specific situations, which are not limited and described herein.
Step S2: and performing trending polynomial fitting by taking the preset prediction order as the polynomial order for performing polynomial fitting on the actual body temperature data, and obtaining the initial prediction order for executing the moving average prediction algorithm according to the change trend of the body temperature data after trending polynomial fitting.
The epidemic disease of beef cattle changes from the body temperature, and the beef cattle has obvious seasonal characteristics for the body temperature, namely, the beef cattle body temperature has irregular regular changes in different time periods due to environmental factors, metabolism and other physiological factors of the beef cattle, certain requirements are met on the stability of data in the prediction process, and under the condition of obvious seasonality, the prediction deviation of the data is larger, so that the original body temperature data needs to be processed linearly. And taking the preset prediction order as a polynomial order for performing polynomial fitting on the actual body temperature data, performing trending polynomial fitting, and obtaining an initial prediction order for executing a moving average prediction algorithm according to the change trend of the body temperature data after trending polynomial fitting.
In one embodiment of the invention, a method of polynomial fitting of actual body temperature data includes:
constructing a matrix with actual body temperature data acquired in a history stage, wherein the matrix is of the size of ; Each row of the matrix contains the power of the body temperature data at each moment, and each column contains the number of the body temperature data at all moments; for polynomial ordersThe first of the matrixThe individual rows will contain; The actual body temperature data at each moment is used to construct a target vector, which in one embodiment of the invention is expressed as:
；
wherein, A target vector representing actual body temperature data; Represent the first Actual body temperature data at each moment.
Setting a linear equation set and solving to obtain polynomial coefficients; in one embodiment of the invention, the formula of the system of linear equations is expressed as:
；
wherein, A target vector representing actual body temperature data; Representing a construction matrix; representing a transpose of the build matrix; Representing a polynomial coefficient vector.
Using polynomial coefficient vectorsTo construct a polynomial, the formula of which is expressed as:
；
wherein, Representing a polynomial function; Representing polynomial coefficient vectors Elements of (a) and (b); representing elements in a power sequence of actual body temperature data.
Evaluating and fitting the obtained polynomials to obtain fitted body temperature data; in one embodiment of the invention, the evaluation and fitting of the obtained polynomials can take the method of observing the variation of the difference between the actual values and the fitted values; if the difference gradually decreases along with the increase of the order of the polynomial, the higher order polynomial is indicated to be better fit to the data, and the positive integer 1 is sequentially increased; however, if the variance fluctuation is large, the distribution is uneven, that is, the variance is calculated for the obtained variance, if the variance is larger than the preset variance threshold, the uneven distribution may require selecting a lower polynomial order, and subtracting the positive integer 1 correspondingly; and repeating the steps until adjustment is not needed, obtaining an optimal trending polynomial, and obtaining body temperature data after fitting the trending polynomial as the trending polynomial, so that noise interference is reduced, and the data is more stable. It should be noted that, in one embodiment of the present invention, the preset prediction order is the prediction order when the moving average prediction is performed in the previous history stage; the preset variance threshold is 1; in other embodiments of the present invention, the magnitude of the preset variance threshold may be specifically set according to specific situations, which are not limited and described herein in detail. The specific trending polynomial fitting process is a technical means well known to those skilled in the art, and will not be described in detail herein.
Preferably, in one embodiment of the present invention, the method for acquiring the initial prediction order includes:
drawing an autocorrelation function diagram for the fitted body temperature data; the corresponding hysteresis period number when the negative variation is maximum in the autocorrelation function is used as the initial prediction order for executing the moving average prediction algorithm.
The hysteresis period number refers to a time interval between a time series and its own movement in time. The specific moving average prediction algorithm and the drawing of the autocorrelation function are technical means well known to those skilled in the art, and will not be described in detail herein.
Step S3: forming a moving window by using the actual body temperature data with the initial prediction order, and obtaining a reference judgment value in the moving window according to a preset prediction value corresponding to the actual body temperature data at each moment in the moving window; obtaining a difference value of each moment in the moving window according to the difference between the actual body temperature data and the reference judgment value of each moment in the moving window; obtaining a prediction influence weight of each moment in the moving window according to the change trend of the difference value of each moment in the moving window; and obtaining predicted body temperature data at the next moment of the moving window according to the actual body temperature data and the predicted influence weight at each moment in the moving window.
In a moving average prediction algorithm, dividing data into a plurality of shorter time periods by setting a moving window, and carrying out average calculation on the data in each time period so as to obtain a smooth moving average value; if the moving window is too large, it may cause the moving average line to be too smooth to capture subtle changes in the data; if the moving window is too small, it may result in the moving average line being too sensitive and susceptible to random noise in the data. Thus, a suitable moving window is determined according to the characteristics of the data and the prediction requirements, and a moving window is formed by the actual body temperature data with the initial prediction order.
Because the body temperature of beef cattle is not constant and the beef cattle is influenced by the neck ring equipment, partial body temperature data can possibly fluctuate, in order to obtain the overall trend and change of the data, the influence of abnormal data at a single moment on the overall prediction result is reduced by averaging the prediction values at a plurality of moments, and the stability of the data is improved; and obtaining a reference judgment value in the moving window according to a preset predicted value corresponding to the actual body temperature data at each moment in the moving window.
Preferably, in one embodiment of the present invention, the reference judgment value acquisition method includes:
and accumulating and averaging preset predicted values at all moments to obtain a reference judgment value. In one embodiment of the present invention, the reference judgment value is formulated as:
；
wherein, Representing a reference judgment value; Representing the first in the moving window Corresponding preset predicted values at each moment; representing the magnitude of the initial prediction order.
In the formula of the reference judgment value, the larger the preset prediction value is, the larger the reference judgment value is, the higher the fitting degree of the body temperature data is, and the prediction accuracy is correspondingly improved.
In other embodiments of the present invention, the preset predicted value is obtained when the moving average prediction algorithm is performed on the actual body temperature data of the previous stage.
In order to display the fluctuation amplitude of the body temperature data, the reliability of the data is better known, and the difference between the reference judgment value and the actual body temperature data is compared; if the difference between the actual body temperature data and the reference judgment value is smaller, the accuracy of the description data is higher; otherwise, if the difference is large, the accuracy of the description data is low, and the prediction model needs to be adjusted and optimized. The prediction model can be better improved by analyzing and processing the difference between the reference judgment value and the actual body temperature data, and the prediction precision of the model is improved; and obtaining a differential value of each moment in the moving window according to the difference between the actual body temperature data and the reference judgment value in each moment in the moving window, detecting abnormal body temperature data, and finding out the health problem of the beef cattle in time. In one embodiment of the invention, the differential value is formulated as:
；
wherein, Representing the first in the moving windowDifferential values at each time instant; Representing the first in the moving window Actual body temperature data at each time; representing the reference judgment value.
In the formula of the differential value,Represent the firstThe larger the difference between the actual body temperature data and the reference judgment value at each time point is, the more remarkable the change of the body temperature data is, and the larger fluctuation can exist between the actual temperature data and the predicted value.
If the data at a certain moment has a larger influence on future prediction, the weight of the predicted influence at the moment is relatively larger, so that more attention and consideration can be given to a prediction model, and a more accurate prediction result can be obtained. However, for data with larger fluctuation, the difference value is larger and possibly caused by noise or abnormal value, smaller weight needs to be given, the influence of the fluctuation on the whole prediction result is reduced, and the accuracy of the model is improved; and obtaining the predicted influence weight of each moment in the moving window according to the change trend of the differential value of each moment in the moving window.
Preferably, in one real-time example of the present invention, the method for obtaining the predicted impact weight includes:
Obtaining the predicted impact weight according to an obtaining formula of the predicted impact weight, wherein the obtaining formula of the predicted impact weight is as follows:
；
wherein, Representing the first in the moving windowPredicting influence weights at each moment; representing a differential mean; Representing the first in the moving window Differential values at each time instant; Representing the first in the moving window Differential values at each time instant; representing the magnitude of the initial prediction order.
In the acquisition formula of the prediction influence weight,The sum of absolute values of differences between the differential mean value and differential values at all moments in a moving window is represented, the larger the sum is, the larger the fluctuation amplitude of more data is, and smaller influence weight is required to be given to body temperature data with larger fluctuation amplitude, so that predicted data are more accurate; move in windowThe smaller the difference value of the actual body temperature data at each moment, the higher the prediction accuracy at the moment, the larger the influence on future prediction, and the larger the prediction influence weight.
It should be noted that, in other embodiments of the present invention, the positive-negative correlation may be constructed by other basic mathematical operations, and specific means are technical means well known to those skilled in the art, which are not described herein.
In order to obtain the predicted body temperature data with more accurate and more reference value, by comprehensively considering the reference judgment value, the differential value and the factors of the predicted influence weight, not only the static characteristics of the data are considered, but also the dynamic change condition of the data is fully considered, so that the real condition of the data can be more comprehensively reflected. And obtaining predicted body temperature data at the next moment of the moving window according to the actual body temperature data and the predicted influence weight at each moment in the moving window.
Preferably, in one embodiment of the present invention, the method for acquiring predicted body temperature data includes:
And calculating the product of the actual body temperature data and the predicted influence weight at each moment in the moving window, and accumulating the weighted body temperature data to obtain the predicted body temperature data at the next moment in the moving window as the weighted body temperature data at each moment. In one embodiment of the invention, the formula for predicting body temperature data is expressed as:
；
wherein, Predicted body temperature data representing a next time of the moving window; Representing the first in the moving window Actual body temperature data at each time; Representing the first in the moving window Predicting influence weights of the moments; representing the magnitude of the initial prediction order.
In the formula for predicting the body temperature data, the larger the actual body temperature data is, the larger the predicted influence weight is, the larger the influence on the predicted result is, and the larger the predicted body temperature data at the next moment is.
It should be noted that, in other embodiments of the present invention, the positive-negative correlation may be constructed by other basic mathematical operations, and specific means are technical means well known to those skilled in the art, which are not described herein.
Step S4: obtaining a body temperature error evaluation value according to the difference between the predicted body temperature data and the corresponding actual body temperature data and the preset body temperature fluctuation parameter of the beef cattle; and judging whether the initial prediction order needs to be adjusted according to the body temperature error evaluation value, if so, performing trending polynomial fitting by taking the initial prediction order as the polynomial order for performing polynomial fitting on the actual body temperature data to obtain a new initial prediction order until the initial prediction order does not need to be adjusted, and obtaining the optimized prediction order.
By selecting a proper moving window, random fluctuation can be smoothed, random noise in data is eliminated, and the method is more suitable for the prediction problem of different scenes. In order to check the size of the selected moving window, namely the accuracy of the initial prediction order, the prediction in the actual body temperature data sequence is ensured to be accurate enough, so that the aim of checking the adaptability of the initial prediction order is fulfilled, the difference trend between the body temperature data obtained by moving average prediction and the actual body temperature data is analyzed, the larger the difference trend is, the more the normal body temperature fluctuation of beef cattle is exceeded, the worse the accuracy of the initial prediction order for prediction is, and the more likely the corresponding adjustment is needed; and obtaining a body temperature error evaluation value according to the difference between the predicted body temperature data and the corresponding actual body temperature data and the preset body temperature fluctuation parameters of the beef cattle.
Preferably, in one embodiment of the present invention, the method for acquiring the body temperature error evaluation value includes:
calculating the difference between the predicted body temperature data and the corresponding actual body temperature data as a body temperature error; and calculating the ratio between the body temperature error and the preset body temperature fluctuation parameter to obtain a body temperature error evaluation value. In one embodiment of the present invention, the formula for the body temperature error assessment value is expressed as:
；
wherein, A body temperature error evaluation value; predicted body temperature data representing a next time of the moving window; actual body temperature data representing the next moment of the moving window; is a preset body temperature fluctuation parameter of beef cattle.
In the formula of the body temperature error evaluation value,The larger the ratio of the difference between the predicted value of the window and the real temperature value at the next moment to the preset body temperature fluctuation parameter is, the larger the difference between the predicted value and the real body temperature data is, the larger the body temperature error estimated value is, which means that the worse the adaptability of the predicted order to the body temperature data of the section is, and the predicted order should be selected again.
It should be noted that, in one embodiment of the present invention, the preset body temperature fluctuation parameter of the beef cattle may take a tested value of 0.75 ℃; the body temperature fluctuation parameter can be calculated through a big data neural network, and the specific big data neural network is a technical means well known to those skilled in the art, and is not described herein.
The body temperature error evaluation value can indicate the adaptability of the initial prediction technology to the actual body temperature data in the moving window, if the body temperature error evaluation value is larger, the adaptability is worse, the accurate predicted body temperature data cannot be obtained, and the adjustment is needed again so as to achieve a better prediction effect; in order to better adapt to the characteristics of the data, the trending polynomial fitting is required to be continuously carried out, the prediction error is smoothed, the optimal prediction order is obtained through adjustment, whether the initial prediction order needs to be adjusted is judged according to the body temperature error evaluation value, if so, the initial prediction order is used as the polynomial order for carrying out the polynomial fitting on the actual body temperature data to carry out the trending polynomial fitting, a new initial prediction order is obtained, and until the initial prediction order does not need to be adjusted, the optimal prediction order is obtained.
Preferably, in one embodiment of the present invention, determining whether the initial prediction order needs to be adjusted according to the body temperature error evaluation value includes:
If the body temperature error evaluation value is smaller than a preset first threshold value, the initial prediction order does not need to be continuously adjusted; if the body temperature error evaluation value is larger than a preset first threshold value, the initial prediction order is judged to need to be adjusted. It should be noted that, in one embodiment of the present invention, the magnitude of the preset first threshold is 1; in other embodiments of the present invention, the preset first threshold may be specifically set according to specific situations, which are not limited and described herein in detail.
Step S5: obtaining a predicted first body temperature according to the optimized prediction order; acquiring the ambient temperature at the current moment, and acquiring a predicted second body temperature of the beef cattle according to the predicted first body temperature, the distribution characteristics of the ambient temperature and the fluctuation characteristics of heart rate data; beef health is detected based on the predicted second body temperature.
The size of the prediction order directly influences the prediction performance of the model, and the proper prediction order can enable the model to be better suitable for the change rule of data, so that the prediction accuracy is improved. By optimizing the prediction order, the prediction model can be dynamically adjusted according to the change condition of the data, and the most suitable model parameters are found, so that the prediction result is more accurate and reliable. And obtaining a predicted first body temperature according to the optimized prediction order.
Preferably, in one embodiment of the present invention, the method for obtaining the predicted first body temperature includes:
Acquiring predicted body temperature data of the next moment of the moving window corresponding to the optimized prediction order; and replacing the actual body temperature data in the corresponding moving window with the predicted body temperature data, and continuing moving average calculation until the actual body temperature data at all the historical moments participate in calculation, so as to obtain a final result mean value as a predicted first body temperature.
Because the temperature prediction of the beef cattle is not comprehensive and accurate only by temperature change, the situation of large error can exist to cause misjudgment of the current health state of the beef cattle, thereby wasting monitoring resources; in order to reduce the influence of erroneous judgment as much as possible, the influence of the first body temperature combined with other physiological and environmental factors of the beef cattle needs to be predicted to obtain an accurate prediction result. And obtaining the predicted second body temperature of the beef cattle according to the predicted first body temperature, the distribution characteristics of the ambient temperature and the fluctuation characteristics of the heart rate data.
Preferably, in one embodiment of the present invention, the method for obtaining the predicted second body temperature includes:
obtaining a predicted second body temperature according to an obtaining formula of the predicted second body temperature, wherein the obtaining formula of the predicted second body temperature is as follows:
；
wherein, Indicating a predicted second body temperature; means for predicting a first body temperature; Representing the ambient temperature at the current time; representing the average temperature of the environment; Heart rate data representing beef cattle at the current moment; representing the normal average heart rate of beef cattle; maximum value representing beef cattle normal heart rate data; A minimum value representing normal heart rate data of the beef cattle; representing the normal heart rate fluctuation range of beef cattle; representing natural constants.
In the obtaining formula for predicting the second body temperature, the higher the predicted first body temperature of the beef cattle per se is, the higher the influence on the subsequent body temperature is, and the higher the predicted second body temperature is; by exponential function based on natural constantA negative correlation mapping is performed and the correlation value is calculated,Representing a ratio between a difference between heart rate data at a current time and a normal average heart rate and a normal heart rate fluctuation range; the larger the ratio is, the more the heart rate data at the current moment deviates from the normal heart rate data, the worse the health condition of beef cattle is, and the greater the influence on the body temperature is generated; Representing the difference between the ambient temperature at the current time and the ambient average temperature, for the ambient temperature at the current time to be lower than the normal ambient average temperature, a corresponding decrease in the predicted first temperature is required, For-1, the ratio of the heart rate ratio is larger, the second body temperature is predicted to be reduced more, and conversely, the second body temperature is predicted to be reduced more; for the higher the ambient temperature at the current moment than the normal ambient temperature, the corresponding increase of the predicted first body temperature is required,For 1, the higher the heart rate ratio, the greater the predicted second body temperature rise, and conversely, the less the predicted second body temperature rise, corresponding to the ratio of the term.
It should be noted that, in other embodiments of the present invention, the positive-negative correlation may be constructed by other basic mathematical operations, and specific means are technical means well known to those skilled in the art, which are not described herein.
According to the degree of the rise of the body temperature and the difference of the accompanying symptoms, the type and the severity of the diseases of the beef cattle can be judged, so that the change trend of the second body temperature is analyzed and predicted, and the health state of the beef cattle is judged; beef health is detected based on the predicted second body temperature.
In one embodiment of the invention, the difference between the predicted second body temperature and the corresponding actual body temperature data is calculated, the difference result is compared with a preset second threshold value, if the difference result is smaller than the preset experience threshold value, the body temperature is judged to fluctuate within a normal range, and the beef cattle is in a healthy state without taking protective measures; if the difference result is larger than a preset second threshold value, abnormal fluctuation of the body temperature is judged, and the beef cattle needs to be immediately checked for health due to the fact that virus invasion starts to generate lesions, so that the beef cattle can be timely treated, early diagnosis and early treatment are achieved, epidemic disease transmission and death rate are reduced, and breeding loss is reduced. It should be noted that, in one embodiment of the present invention, the preset second threshold value is 0.5 ℃; in other embodiments of the present invention, the magnitude of the preset second threshold may be specifically set according to specific situations, which are not limited and described herein in detail.
In summary, the present invention obtains an initial prediction order; obtaining predicted body temperature data at the next moment of the moving window according to the actual body temperature data and predicted influence weight at each moment in the moving window; obtaining a body temperature error evaluation value according to the difference between the predicted body temperature data and the corresponding actual body temperature data and the preset body temperature fluctuation parameter of the beef cattle; judging whether the initial prediction order needs to be adjusted according to the body temperature error evaluation value, if so, performing trending polynomial fitting by taking the initial prediction order as the polynomial order for performing polynomial fitting on the actual body temperature data to obtain a new initial prediction order until the initial prediction order does not need to be adjusted, and obtaining an optimized prediction order; further obtaining a predicted first body temperature; obtaining a predicted second body temperature of the beef cattle according to the predicted first body temperature, the distribution characteristics of the ambient temperature and the fluctuation characteristics of the heart rate data; and detecting the health of beef cattle. According to the method, the accuracy of predicting the body temperature data is improved by determining the proper prediction order, and the health of the beef cattle is judged in time.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (7)
1. A beef cattle health detection data processing method, characterized in that the method comprises the following steps:
acquiring health data of beef cattle at each moment in a history stage of the current moment; the health data comprises actual body temperature data and heart rate data of beef cattle;
performing trending polynomial fitting by taking the preset prediction order as the polynomial order for performing polynomial fitting on the actual body temperature data, and obtaining an initial prediction order for executing a moving average prediction algorithm according to the change trend of the body temperature data after trending polynomial fitting;
Forming a moving window by the initial prediction order actual body temperature data, and obtaining a reference judgment value in the moving window according to a preset prediction value corresponding to the actual body temperature data at each moment in the moving window; obtaining a difference value of each moment in the moving window according to the difference between the actual body temperature data and the reference judgment value of each moment in the moving window; obtaining a prediction influence weight of each moment in the moving window according to the change trend of the difference value of each moment in the moving window; obtaining predicted body temperature data at the next moment of the moving window according to the actual body temperature data at each moment in the moving window and the predicted influence weight;
Obtaining a body temperature error evaluation value according to the difference between the predicted body temperature data and the corresponding actual body temperature data and the preset body temperature fluctuation parameter of the beef cattle; judging whether the initial prediction order needs to be adjusted according to the body temperature error evaluation value, if so, performing trending polynomial fitting by taking the initial prediction order as the polynomial order for performing polynomial fitting on actual body temperature data to obtain a new initial prediction order until the initial prediction order does not need to be adjusted, and obtaining an optimized prediction order;
Obtaining a predicted first body temperature according to the optimized prediction order; acquiring the environmental temperature at the current moment, and acquiring a predicted second body temperature of the beef cattle according to the predicted first body temperature, the distribution characteristics of the environmental temperature and the fluctuation characteristics of heart rate data; detecting beef cattle health according to the predicted second body temperature;
The reference judgment value acquisition method comprises the following steps:
accumulating and averaging the preset predicted values at all moments to obtain a reference judgment value;
The method for acquiring the predicted impact weight comprises the following steps:
Obtaining the predicted impact weight according to an obtaining formula of the predicted impact weight, wherein the obtaining formula of the predicted impact weight is as follows:
; wherein, Representing the first in the moving windowPredicting influence weights at each moment; representing a differential mean; Representing the first in the moving window Differential values at each time instant; Representing the first in the moving window Differential values at each time instant; representing the magnitude of the initial prediction order;
The method for obtaining the predicted second body temperature comprises the following steps:
obtaining a predicted second body temperature according to an obtaining formula of the predicted second body temperature, wherein the obtaining formula of the predicted second body temperature is as follows:
; wherein, Indicating a predicted second body temperature; means for predicting a first body temperature; Representing the ambient temperature at the current time; representing the average temperature of the environment; Heart rate data representing beef cattle at the current moment; representing the normal average heart rate of beef cattle; maximum value representing beef cattle normal heart rate data; A minimum value representing normal heart rate data of the beef cattle; representing the normal heart rate fluctuation range of beef cattle; representing natural constants.
2. The beef cattle health detection data processing method according to claim 1, wherein the initial prediction order acquisition method comprises the following steps:
Drawing an autocorrelation function diagram for body temperature data after trending polynomial fitting; the corresponding hysteresis period number when the autocorrelation function changes most negatively is used as the initial prediction order for executing the moving average prediction algorithm.
3. A beef cattle health detection data processing method according to claim 1, the method for acquiring the predicted body temperature data is characterized by comprising the following steps:
And calculating the product of the actual body temperature data and the predicted influence weight at each moment in the moving window, and accumulating the weighted body temperature data in the moving window to obtain the predicted body temperature data at the next moment in the moving window as the weighted body temperature data at each moment.
4. The method for processing beef cattle health detection data according to claim 1, wherein the method for acquiring the estimated body temperature error value comprises the steps of:
calculating the difference between the predicted body temperature data and the corresponding actual body temperature data as a body temperature error;
And calculating the ratio between the body temperature error and the preset body temperature fluctuation parameter to obtain a body temperature error evaluation value.
5. The method of claim 1, wherein determining whether the initial predicted order needs to be adjusted based on the body temperature error assessment value comprises:
If the body temperature error evaluation value is smaller than a preset first threshold value, the initial prediction order does not need to be continuously adjusted; if the body temperature error evaluation value is larger than a preset first threshold value, the initial prediction order is judged to need to be adjusted.
6. The beef cattle health detection data processing method according to claim 1, wherein the method for obtaining the predicted first body temperature comprises the following steps:
Acquiring predicted body temperature data of the next moment of the moving window corresponding to the optimized prediction order; and replacing the actual body temperature data in the corresponding moving window with the predicted body temperature data, and continuing moving average calculation until the actual body temperature data at all the historical moments participate in calculation, so as to obtain a final result mean value as a predicted first body temperature.
7. The method for processing beef cattle health detection data according to claim 4, wherein the empirical value of the preset body temperature fluctuation parameter is 0.75.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410606430.1A CN118197635B (en) | 2024-05-16 | 2024-05-16 | Beef cattle health detection data processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410606430.1A CN118197635B (en) | 2024-05-16 | 2024-05-16 | Beef cattle health detection data processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118197635A CN118197635A (en) | 2024-06-14 |
| CN118197635B true CN118197635B (en) | 2024-07-09 |
Family
ID=91403953
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410606430.1A Active CN118197635B (en) | 2024-05-16 | 2024-05-16 | Beef cattle health detection data processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118197635B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112504491A (en) * | 2020-12-11 | 2021-03-16 | 无锡博智芯科技有限公司 | Body temperature measuring method based on wearable equipment |
| CN114417242A (en) * | 2021-12-20 | 2022-04-29 | 淮阴工学院 | Big data detection system for livestock and poultry activity information |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160331244A1 (en) * | 2015-05-12 | 2016-11-17 | Razzberry Inc. | Core body temperature system |
| CN115191956B (en) * | 2021-04-09 | 2024-08-23 | 广东小天才科技有限公司 | Core body temperature detection method, wearable device and computer readable storage medium |
| CN117786325B (en) * | 2024-02-27 | 2024-04-30 | 山东晨晖电子科技有限公司 | Ambient temperature wisdom early warning system of heavy-calibre thing networking water gauge |
-
2024
- 2024-05-16 CN CN202410606430.1A patent/CN118197635B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112504491A (en) * | 2020-12-11 | 2021-03-16 | 无锡博智芯科技有限公司 | Body temperature measuring method based on wearable equipment |
| CN114417242A (en) * | 2021-12-20 | 2022-04-29 | 淮阴工学院 | Big data detection system for livestock and poultry activity information |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118197635A (en) | 2024-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108681633B (en) | Condensate pump fault early warning method based on state parameters | |
| CN116757534B (en) | Intelligent refrigerator reliability analysis method based on neural training network | |
| TW201915727A (en) | Fault detection and classification method of multi-sensors | |
| Mo et al. | Multi-head CNN-LSTM with prediction error analysis for remaining useful life prediction | |
| CN116400126B (en) | Low-voltage power box with data processing system | |
| CN113255848A (en) | Water turbine cavitation sound signal identification method based on big data learning | |
| CN108956111B (en) | Abnormal state detection method and detection system for mechanical part | |
| CN117196353B (en) | Environmental pollution assessment and monitoring method and system based on big data | |
| CN117238507A (en) | Intensive care monitoring system based on Internet of things | |
| CN117708748B (en) | Operation monitoring system and method for centrifugal fan | |
| CN116595327B (en) | Sluice deformation monitoring data preprocessing system and method | |
| CN112651119A (en) | Multi-performance-parameter accelerated degradation test evaluation method for space harmonic reducer | |
| CN118376284B (en) | Multi-mode monitoring data optimized transmission method for Internet of things | |
| CN115204227A (en) | Uncertainty quantitative calibration method in equipment fault diagnosis based on deep learning | |
| CN118228130B (en) | Monitoring method, system and storage medium based on equipment health state | |
| CN112836180A (en) | Sensing data determination method and device, electronic equipment and readable storage medium | |
| CN118197635B (en) | Beef cattle health detection data processing method | |
| CN107437112B (en) | A kind of mixing RVM model prediction methods based on the multiple dimensioned kernel function of improvement | |
| CN118014115A (en) | Production line loss monitoring method based on long-short-term memory network | |
| JP6969588B2 (en) | Anomaly detectors, anomaly detection methods, and computer programs | |
| CN115795999B (en) | Early warning method for abnormal performance of long-term service pumped storage unit | |
| CN113868957B (en) | Residual life prediction and uncertainty quantitative calibration method under Bayes deep learning | |
| CN113569324A (en) | Slope deformation monitoring abnormal data analysis and optimization method | |
| CN113822580A (en) | Equipment working condition evaluation method and related equipment | |
| CN118312727B (en) | Cooling system temperature monitoring method for intelligent carbon dioxide incubator |
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 |