Pig house environment regulation and control method based on brain wave control

Abstract

The invention discloses a hog house environment regulation and control method based on brain wave control, and relates to the technical field of livestock breeding. The problems of stress response and limited accuracy are solved. The method specifically comprises the following steps: acquiring electroencephalogram signal data of a pig in real time by using an electroencephalogram acquisition device; electroencephalogram signal data are analyzed and processed through AI; an analysis result is input into a pig electroencephalogram characteristic database for query, and an execution mechanism is dynamically regulated and controlled; an environment sensor is used for collecting environment parameters in the hog house; the electroencephalogram collecting device is an implanted subcutaneous type / head-mounted type brain wave collector and comprises a flexible dry electrode array with more than 256 leads. The limitation that traditional environment monitoring only depends on physical parameters (such as temperature and humidity) is broken through, and compared with a traditional video monitoring mode, the temperature control accuracy can be better improved, the stress response can be achieved in milliseconds, the influence of individual differences is eliminated, the energy consumption of environment regulation and control equipment is reduced, and the production income of pigs is increased.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of livestock breeding, and in particular to a pig house environment regulation method based on brain wave control. BACKGROUND

[0002] The existing pig house environment regulation system generally monitors temperature, humidity, ammonia concentration and other parameters based on a sensor network, and automatically adjusts equipment (such as a fan, a water curtain) through a preset threshold to achieve the regulation of the environment. However, such a method cannot directly reflect the physiological comfort of pigs, for example, when pigs produce stress due to excessive light or noise interference, the traditional system may not respond because the environmental parameters are not out of the standard.

[0003] After searching, the patent with the Chinese patent application number CN202410397145.3 discloses an intelligent pig farm environment monitoring and control system and method, which includes real-time collection of environmental data in the pig farm to obtain first environmental data values and first image data; training an identification model, using the trained identification model to identify the first image data to obtain the breed and current growth stage of each pig in each pig house; obtaining the range interval of the suitable environmental data for each growth stage of different breeds of pigs to obtain the second environmental data interval suitable for the life of each pig house. The intelligent pig farm environment monitoring and control system and method in the above-mentioned patent have the following disadvantages: the environmental suitability is indirectly inferred by analyzing the behavior data of pigs, but the accuracy and real-time performance are limited by the complexity of the behavior characteristics, so there are problems of stress response and limited accuracy. The present application proposes to construct a "pig-environment" dynamic feedback closed loop by using brain wave signals as a direct physiological indicator combined with AI algorithms, which has a significant technical breakthrough. SUMMARY

[0004] The purpose of the present application is to solve the problems existing in the prior art, and to propose a pig house environment regulation method based on brain wave control.

[0005] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0006] The pig house environment regulation method based on brain wave control includes the following steps:

[0007] S1: using an electroencephalogram acquisition device to acquire real-time electroencephalogram data of pigs;

[0008] S2: AI analyzes and processes the electroencephalogram data;

[0009] S3: inputs the analysis results into a pig electroencephalogram feature database for query and dynamically regulates the execution mechanism;

[0010] The S1 further includes using an environmental sensor to collect environmental parameters in the pig house.

[0011] Preferably, the electroencephalogram acquisition device is a subcutaneous / head-mounted brain wave collector, which includes a flexible dry electrode array with 256 leads or more, an embedded preprocessing module and a low-power Bluetooth 5.3 transmission unit.

[0012] Preferably, the method for acquiring electroencephalogram signal data comprises the following contents:

[0013] A1: Select pigs of different growth stages and breeds uniformly distributed in the pig house as sentinel pigs;

[0014] A2: Install the electroencephalogram acquisition device on each sentinel pig in turn for real-time and stable acquisition of the delta wave, theta wave, alpha wave, beta wave and gamma wave frequency band signals of the pigs;

[0015] The proportion of the number of sentinel pigs is 1% to 3% of the total number of pigs.

[0016] Preferably, the analysis and processing step of the electroencephalogram signal data comprises the following contents:

[0017] B1: Noise filtering of electroencephalogram signal;

[0018] B2: Extraction of main features of electroencephalogram signal;

[0019] B3: Decoding and analyzing the mapping relationship between electroencephalogram signal and environment by algorithm.

[0020] Preferably, the noise filtering method comprises the following contents:

[0021] ① Use a band-pass filter to remove power frequency interference and electromyographic artifacts;

[0022] ② Independent component analysis separates eye movement and electrocardiogram noise;

[0023] ③ 3 standard deviation principle to remove amplitude abnormal data and normalize the signal.

[0024] Preferably, the main features include time domain features, frequency domain and time-frequency domain features, spatial domain features and nonlinear features;

[0025] The algorithm is a traditional machine learning, deep learning model or CNN-LSTM hybrid model.

[0026] Preferably, the pig electroencephalogram feature database includes the correlation parameters of environmental temperature and humidity and pig electroencephalogram signal, the correlation parameters of pig electroencephalogram signal and pig behavior and posture, and stress index parameters;

[0027] The pig behavior and posture includes heat stress posture, cold shivering frequency or huddling and crouching;

[0028] The corresponding relationship between the environmental temperature and humidity and the electroencephalogram signal, and the electroencephalogram signal and the pig behavior and posture is as follows:

[0029] 1) When the temperature and humidity interval is characterized by high temperature and high humidity, the physiological mechanism of the change in the electroencephalogram is that the gamma wave power of the frontal lobe is decreased by 38-45%, and the overall brain activity is inhibited, and the mouth breathing frequency of the pig is greater than 40 times / min.

[0030] 2) When the temperature and humidity interval is characterized by moderate temperature and moderate humidity, the physiological mechanism of the change in the electroencephalogram is that the beta wave coherence of the motor cortex is increased by 35%, the physiological state is balanced, and the pig actively explores the behavior by 35%.

[0031] 3) When the temperature and humidity interval is characterized by low temperature and low humidity, the physiological mechanism of the change in the electroencephalogram is that the theta wave oscillation frequency of the hippocampus is decreased by 2.1 Hz, the overall neural activity is inhibited, the metabolic rate is reduced, and the pig has cold shivering frequency greater than 5 times / min.

[0032] Preferably, the stress index parameter comprises temperature stress starting, ventilation stress starting and dehumidification stress starting.

[0033] The stress index parameter is controlled by using a composite stress index algorithm, and the specific algorithm is as follows: CSI = 0.3 x theta / alpha + 0.4 x beta / delta + 0.3 x (1-gamma norm) 2.5 CSI = 2.5 0.3 x theta / alpha + 0.4 x beta / delta + 0.3 x (1-gamma norm).

[0034] When CSI = 0.50-0.68, the dehumidification stress starting is triggered.

[0035] CSI > 0.68, the ventilation stress starting is triggered.

[0036] CSI > 0.82, the temperature stress starting, the ventilation stress starting and the dehumidification stress starting are triggered.

[0037] Preferably, the execution mechanism comprises ventilation equipment, heating equipment, cooling equipment and dehumidification equipment, and the starting of the execution mechanism further comprises priority starting sorting, and the specific priority starting sorting is as follows: temperature > humidity > ventilation > illumination.

[0038] Preferably, the content of the priority starting sorting further comprises:

[0039] C1: When multiple index conflicts occur, the overall comfort index of the pig group and the individual abnormal value ratio are used to automatically allocate the control priority.

[0040] C2: The fuzzy logic algorithm is used to realize the nonlinear control of the linkage and cooperative control of the execution mechanism.

[0041] The present application has the following beneficial effects:

[0042] 1.The present application provides a method for real-time regulation of pig house environment by collecting brain wave signals of pigs and combining artificial intelligence, aiming to directly reflect the subjective feelings of animals through physiological signals, breaking through the limitations of traditional environment monitoring which only relies on physical parameters (such as temperature, humidity), and compared with traditional video monitoring method, it can improve the accuracy of temperature control, achieve millisecond level stress response, eliminate the influence of individual differences, reduce the energy consumption of environment regulation equipment, and increase the production benefits of pigs.

[0043] 2.The present application uses an electroencephalogram acquisition device to obtain real-time δ / θ / α / β / γ band electroencephalogram signal data of pigs, analyzes and processes the electroencephalogram signal data through AI, transmits the analysis results to a pig electroencephalogram feature database for query, and outputs dynamic regulation of corresponding actuator devices according to the results, thereby achieving the purpose of regulating the environment in the pig house and reducing the incidence of pig diseases. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 The flowchart of the pig house environment regulation method based on brain wave control proposed by the present application is shown.

[0045] Figure 2 The figure of the influence of pigs on various parameters in the practical application of the pig house environment regulation method of the present application in test example 1 is shown. DETAILED DESCRIPTION

[0046] The technical solution of the present patent will be further described in detail in combination with specific embodiments.

[0047] The embodiments of the present patent will be described in detail below, and the examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are only used to explain the present patent, and cannot be understood as a limitation of the present patent.

[0048] Example 1

[0049] The pig house environment regulation method based on brain wave control, as shown in Figure 1 , includes the following steps:

[0050] S1: using an electroencephalogram acquisition device to obtain real-time electroencephalogram signal data of pigs;

[0051] Further, the electroencephalogram acquisition device is a subcutaneous / head-mounted brain wave collector, preferably, the head-mounted brain wave collector is selected in the present embodiment; it includes a flexible dry electrode array with 256 leads or more, an embedded pre-processing module and a low-power Bluetooth 5.3 transmission unit; it ensures spatial resolution and signal integrity (error ≤±0.5μV).

[0052] Preferably, the head-mounted brain wave collector is deployed with a LoRa wireless sensor network, and is fixed with a flexible electrode cap and a pig ear; the subcutaneous brain wave collector is packaged by biocompatible material.

[0053] Preferably, the hardware of the electroencephalogram acquisition device is a BioSemi ActiveTwo system, and the electrodes of the head-mounted brain wave collector are in a modified version of the 10-20 international system, which is suitable for the shape of pig skull.

[0054] Further, the method for acquiring electroencephalogram signal data comprises the following contents:

[0055] A1: Select pigs of different growth stages and breeds evenly distributed in the pig house as sentinel pigs;

[0056] Preferably, the proportion of the number of sentinel pigs is 1%-3% of the total number of pigs;

[0057] A2: Install the electroencephalogram acquisition device on each sentinel pig in turn for real-time and stable acquisition of the signals of δ wave, θ wave, α wave, β wave and γ wave of pigs;

[0058] Among them, δ wave (0-4 Hz), high inter-day duration of δ wave indicates metabolic inhibition or disease, and a 35%±7.2 increase in δ wave power indicates a lower environment;

[0059] Among them, θ wave (4-8 Hz), an increase of 42%±5.8 in θ wave energy indicates a high environmental temperature; and a θ wave / β wave power ratio >1.8 indicates chronic pain;

[0060] Among them, α wave (8-13 Hz), an α wave proportion of 30-40% indicates a good temperature and ventilation environment; and a 53%±9.1 decrease in α wave power indicates ammonia exceeding the standard (>20 ppm);

[0061] Among them, β wave (14-30 Hz), a β wave explosive growth (amplitude increase >80% within 200-300 ms) indicates noise >85 dB; and a β wave continuous proportion >25% under high-density breeding indicates group stress;

[0062] Among them, γ wave (>30 Hz), an abnormal decrease of 19%±3.8 in γ wave power during disease incubation period; and an increase of 22%±4.3 in γ wave synchronicity indicates positive emotions.

[0063] As a supplement, the S1 step further comprises using an environmental sensor to collect environmental parameters in the pig house, including temperature and humidity, NH3, CO2, etc.; and serving as a basis for correcting the accuracy of the pig electroencephalogram feature database.

[0064] S2: Analyze and process the electroencephalogram signal data by AI;

[0065] Further, the analysis and processing step of the electroencephalogram signal data comprises the following contents:

[0066] B1: Noise filtering of electroencephalogram signal;

[0067] Specifically, ① using a band-pass filter (such as 0.5-40 Hz) to remove power frequency interference (50 / 60 Hz) and electromyographic artifacts, preferably such as Butterworth / Chebyshev filter;

[0068] ② Independent component analysis (ICA) separates physiological noise such as eye movement and electrocardiogram;

[0069] ③ 3 standard deviation principle to remove amplitude abnormal data, and normalize the signal.

[0070] B2: Extracting main features of electroencephalogram signal;

[0071] Preferably, the main features include time domain features, frequency domain and time-frequency domain features, spatial domain features and nonlinear features, etc., and the main features are extracted by Fourier transform combined with wavelet transform, empirical mode decomposition, etc.

[0072] B3: Using algorithm to decode and analyze the mapping relationship between electroencephalogram signal and environment;

[0073] Preferably, the algorithm is traditional machine learning (such as support vector machine, linear discriminant analysis), deep learning model (such as convolutional neural network, recurrent neural network, hybrid architecture) or CNN-LSTM hybrid model (i.e. convolutional neural network (CNN) and long short-term memory network (LSTM));

[0074] S3: Input the analysis result into the pig electroencephalogram feature database query and dynamically control the actuator;

[0075] Further, the pig electroencephalogram feature database comprises the correlation parameters of environmental temperature and humidity and pig electroencephalogram signal, the correlation parameters of pig electroencephalogram signal and pig behavior posture, and stress index parameters, etc.

[0076] Preferably, the pig behavior posture includes heat stress posture such as mouth breathing, lying down, etc., cold frequency or huddling down, etc.

[0077] Further preferably, the corresponding relationship between the environmental temperature and humidity and the electroencephalogram signal, and the electroencephalogram signal and the pig behavior posture is as follows:

[0078] 1) When the temperature and humidity interval feature is >28℃ / RH>75%, i.e. high temperature and high humidity, the frontal lobe gamma wave power in the physiological mechanism of electroencephalogram change is ↓38-45%, the overall brain activity is inhibited, and the pig mouth breathing frequency is >40 times / min;

[0079] 2) When the temperature and humidity interval is 18-24℃ / RH60%, i.e. moderate temperature and humidity, the physiological mechanism of the change in the electroencephalogram is that the beta wave coherence of the motor cortex increases by 35%, the physiological state is balanced, and the pig actively explores the behavior, which increases by 35%;

[0080] 3) When the temperature and humidity interval is <15℃ / RH<50%, i.e. low temperature and low humidity, the physiological mechanism of the change in the electroencephalogram is that the theta wave oscillation frequency of the hippocampus decreases by 2.1 Hz, the overall neural activity is inhibited, the metabolic rate is reduced, and the pig has cold shivering with a frequency of >5 times / min.

[0081] Further preferably, the stress index parameter includes temperature stress start, ventilation stress start, and dehumidification stress start.

[0082] The stress index parameter is controlled by using a composite stress index algorithm (CSI), and the specific algorithm is as follows: CSI = 0.3 x θ / α + 0.4 x β / δ + 0.3 x (1-γnorm)2.5 CSI = 2.5 0.3 x θ / α + 0.4 x β / δ + 0.3 x (1-γnorm).

[0083] When CSI = 0.50-0.68, the dehumidification stress start is triggered.

[0084] When CSI > 0.68, the ventilation stress start is triggered.

[0085] When CSI > 0.82, the temperature stress start, ventilation stress start, and dehumidification stress start are triggered.

[0086] Further, the execution mechanism includes ventilation equipment, heating equipment, cooling equipment, and dehumidification equipment.

[0087] In addition, the start of the execution mechanism also includes priority start sequencing, and the specific priority start sequencing is as follows: temperature > humidity > ventilation > illumination.

[0088] In use, the brain electrical acquisition device is used to acquire the δ / θ / α / β / γ wave band electroencephalogram signal data of the pig in real time, the electroencephalogram signal data is analyzed by AI, the analysis result is transmitted to the pig electroencephalogram feature database for query, the corresponding execution mechanism device is output according to the result, so as to achieve the purpose of adjusting the environment in the pig house and reduce the incidence of pig diseases.

[0089] Embodiment 2:

[0090] The pig house environment regulation method based on brain wave control, as shown in Figure 1 In order to further avoid unnecessary energy consumption of environmental equipment start and stop, the following supplements are made on the basis of Embodiment 1: the content of the priority start sequencing also includes:

[0091] C1: When multiple indicators conflict, automatically assign control priority according to the overall comfort index of the pig group and the proportion of individual outliers;

[0092] For example: temperature and humidity conflict, when the temperature is suitable but the humidity is excessive, if 30% of the pigs show anxiety, the environmental dehumidification adjustment is started first.

[0093] C2: Based on fuzzy logic algorithm to realize nonlinear control of actuator linkage and collaborative control, avoid frequent start and stop.

[0094] The present application provides a method for real-time regulation of pig house environment by collecting the brain wave signals of pigs and combining artificial intelligence, which aims to directly reflect the subjective feelings of animals through physiological signals, break through the limitations of traditional environment monitoring which only relies on physical parameters (such as temperature, humidity), and improve the temperature control accuracy compared with the traditional video monitoring method. The stress response is realized in milliseconds, the influence of individual differences is eliminated, the energy consumption of the environment regulation equipment is reduced by 31%, and the production income of pigs is increased by 28%.

[0095] Test Example 1:

[0096] In a comparative test in a certain large-scale pig farm, compared with the traditional control method, the pig group daily weight gain of the method of the present application is increased by 25%, the theta wave energy is increased by 42% ± 5, the stress response rate is decreased by 68%, and the environmental regulation response delay is decreased by 95%.

[0097] The above is only the preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can make equivalent replacement or change according to the technical solution and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A method for controlling the environment of a pig house based on brain waves, characterized by, The method comprises the following steps: S1: using an electroencephalogram acquisition device to acquire real-time electroencephalogram signal data of pigs; S2: analyzing the electroencephalogram signal data by AI; S3: inputting the analysis results into a pig electroencephalogram feature database for query and dynamically controlling an execution mechanism. The S1 further comprises using an environmental sensor to collect environmental parameters in a pig house.

2. The pig house environment regulation method based on brain wave control according to claim 1, characterized in that, The electroencephalogram acquisition device is a subcutaneous / head-mounted brain wave collector, which comprises a flexible dry electrode array with 256 leads or more, an embedded preprocessing module and a low-power Bluetooth 5.3 transmission unit.

3. The pig house environment regulation method based on brain wave control according to claim 2, characterized in that, The method for acquiring the electroencephalogram signal data comprises the following contents: A1: selecting pigs of different growth stages and breeds evenly distributed in a pig house as sentinel pigs; A2: sequentially installing electroencephalogram acquisition devices on the sentinel pigs for real-time and stable acquisition of delta wave, theta wave, alpha wave, beta wave and gamma wave frequency band signals of the pigs; The proportion of the number of the sentinel pigs is 1%-3% of the total number of pigs.

4. The brainwave control-based pig house environment regulation method according to claim 1, characterized in that, The analysis and processing steps of the electroencephalogram signal data comprise the following contents: B1: noise filtering of the electroencephalogram signal; B2: extraction of main features of the electroencephalogram signal; B3: decoding and analyzing the mapping relationship between the electroencephalogram signal and the environment by using an algorithm.

5. The brainwave control-based pig house environment regulation method according to claim 4, characterized in that, The noise filtering method comprises the following contents: ① using a band-pass filter to remove power frequency interference and electromyographic artifacts; ② independent component analysis to separate eye movement and electrocardiogram noise; ③ 3 standard deviation principle to remove amplitude abnormal data and normalize the signal.

6. The brainwave control-based pig house environment regulation method according to claim 5, characterized in that, The main features include time domain features, frequency domain and time-frequency domain features, spatial domain features and nonlinear features; The algorithm is a traditional machine learning, deep learning model or a CNN-LSTM hybrid model.

7. The brainwave control-based pig house environment regulation method according to claim 1, characterized in that, The pig electroencephalogram feature database comprises associated parameters of environmental temperature and humidity and pig electroencephalogram signals, associated parameters of pig electroencephalogram signals and pig behavior and posture, and stress index parameters; The pig behavior and posture include heat stress posture, cold shivering frequency or huddling and lying down; The corresponding relationship between the environmental temperature and humidity and the electroencephalogram signal, and the electroencephalogram signal and the pig behavior and posture is as follows: 1) when the temperature and humidity interval feature is high temperature and high humidity, the gamma wave power in the frontal lobe of the electroencephalogram change physiological mechanism decreases by 38-45%, the overall electroencephalogram activity is inhibited, and the pig's panting frequency is >40 times / min; 2) when the temperature and humidity interval feature is moderate temperature and suitable humidity, the beta wave coherence in the motor cortex of the electroencephalogram change physiological mechanism increases by 35%, the physiological state is balanced, and the pig's active exploration behavior increases by 35%; 3) when the temperature and humidity interval feature is low temperature and low humidity, the theta wave oscillation frequency in the hippocampus of the electroencephalogram change physiological mechanism decreases by 2.1 Hz, the overall neural activity is inhibited, the metabolic rate is reduced, and the pig's cold shivering frequency is >5 times / min.

8. The pig house environment regulation method based on brain wave control according to claim 7, characterized in that, The stress index parameters include temperature stress start, ventilation stress start and dehumidification stress start; The stress index parameters are controlled by using a composite stress index algorithm, specifically as follows: CSI = 0.3 x θ / α + 0.4 x β / δ + 0.3 x (1-γnorm)2.5 CSI = 2.50.3 x θ / α + 0.4 x β / δ + 0.3 x (1-γnorm); When CSI = 0.50-0.68, the dehumidification stress start is triggered. CSI > 0.68, triggering ventilation stress start; CSI > 0.82, triggering temperature stress start, ventilation stress start and dehumidification stress start.

9. The brainwave control-based pig house environment regulation method according to claim 1, characterized in that, The execution mechanism includes ventilation equipment, heating equipment, cooling equipment and dehumidification equipment, and the start of the execution mechanism also includes priority start sequencing, specifically: temperature > humidity > ventilation > light.

10. The brainwave control-based pig house environment regulation method according to claim 9, characterized in that, The content of the priority start sequencing also includes: C1: When multiple indicators conflict, automatically allocate control priority according to the pig group overall comfort index and individual abnormal value proportion; C2: Realize nonlinear control of execution mechanism linkage and collaborative control based on fuzzy logic algorithm.

Images