Sheep state recognition device and method based on respiration monitoring

Abstract

The invention belongs to the technical field of sensors and detection, and provides a sheep state recognition device and method based on respiration monitoring, and the device comprises a stretchable strain sensor, a circuit module and other parts; the method comprises the steps of signal correction and filtering processing, waveform feature calculation and sheep state judgment. Through the stretchable strain sensor with better biocompatibility and electrical property, real-time monitoring and feedback on the sheep are realized, the manufacturing cost is lower, the fitness with the body shape of the sheep is better, excessive intervention on the sheep is reduced, and signal acquisition is more accurate; sheep state judgment is carried out through the logic evaluation truth table, and the accuracy is higher and more reliable.

Description

Technical Field

[0001] The present invention relates to the field of sensors and detection technology, and in particular to a sheep state recognition device and method based on respiratory monitoring. Background Art

[0002] At present, respiratory rate is commonly measured by observing the rise and fall of the animal's chest and abdominal wall in intensive farms. This method is simple but may be affected by the subjective judgment of the staff. Moreover, excessive human intervention in the normal activities of animals is required during the monitoring process, which leads to increased stress on animals and is not in line with animal welfare. Most importantly, traditional monitoring methods only provide static data and lack the ability to monitor and provide feedback on the changing vital signs of farm animals in real time. Therefore, the delay in obtaining information hinders the dynamic adjustment of care strategies and rapid response to emergencies. Summary of the invention

[0003] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a sheep status identification device and method based on respiratory monitoring, which realizes real-time monitoring and feedback of sheep through a stretchable strain sensor with better biocompatibility and electrical performance, reduces manufacturing costs and excessive intervention on sheep, and improves the accuracy of signal acquisition; sheep status is judged through a logical evaluation truth table to improve the accuracy and reliability of the identification result.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A sheep state recognition device based on respiratory monitoring, comprising: a stretchable strain sensor, a circuit module and other components; the stretchable strain sensor comprises: a flexible substrate layer, a liquid metal sensitive layer and a top packaging layer distributed in a sandwich structure; the circuit module comprises: a signal acquisition submodule, a signal processing submodule, a wireless signal transmission submodule and a display submodule connected in sequence; the other components comprise: a protective shell, an elastic bandage, a lithium battery and a wire;

[0006] The stretchable strain sensor is connected to the signal acquisition submodule;

[0007] The signal acquisition submodule is used to collect real-time resistance signals of changes in chest expansion and contraction during the breathing process of the target sheep; the signal processing submodule is used to preprocess the original resistance signal to obtain waveform characteristic data; the wireless signal transmission submodule is used to transmit the waveform characteristic data to the display submodule or the target device; the display submodule is used to display the waveform characteristic data; the elastic strap is used to fix the stretchable strain sensor.

[0008] Preferably, the preparation process of the flexible substrate layer comprises:

[0009] The polydimethylsiloxane solution and the curing agent were mixed in a ratio of 10:1, and stirred at room temperature for 20 minutes using a power-enhanced electric stirrer to obtain a solution A;

[0010] Dissolve dopamine in tris(hydroxymethylaminomethane) buffer, stir with a magnetic stirrer at room temperature for 20 minutes, and adjust the pH value of the solution to 8.5 to obtain solution B;

[0011] The solution A is mixed with the solution B, so that the solution B is evenly distributed in the network of the solution A, and a double network structure is generated by initiating a polymerization reaction of polydopamine with an initiator of potassium persulfate to obtain a mixed solution;

[0012] The mixed solution was applied on a glass sheet with a surface tension lower than 24 mN / m, and spin-coated at 1600 rpm for 10 minutes using an automatic spin-coating machine to obtain an original film with a thickness of 1 mm;

[0013] The original film was placed in a vacuum drying oven at 60° C. and dried for 20 minutes to obtain the flexible substrate layer.

[0014] Preferably, the preparation process of the liquid metal sensitive layer includes:

[0015] The dimensional parameters are obtained using a pre-designed strain signal response model;

[0016] Using a 3D printer to create a conductive channel mold on the flexible base layer according to the size parameters;

[0017] Injecting the gallium-indium alloy into the conductive channel mold, maintaining a preset temperature environment, and drying at room temperature for 30 minutes after the gallium-indium alloy is evenly distributed in the conductive channel mold to obtain a serpentine conductive path;

[0018] The two wires are respectively connected to the starting point and the end point of the serpentine conductive path to obtain the liquid metal sensitive layer.

[0019] Preferably, the process of constructing the strain signal response model includes:

[0020] Disassembling the serpentine conductive path into sensing units, and dividing the sensing units into horizontal sensing units, vertical sensing units and inflection point sensing units according to the relationship between positions and strains;

[0021] Determine the resistance value calculation formulas of the horizontal sensing unit, the vertical sensing unit and the inflection point sensing unit; the resistance value calculation formulas of the horizontal sensing unit, the vertical sensing unit and the inflection point sensing unit are respectively:

[0022] as well as

[0023] Wherein, R1 is the resistance value of the horizontal sensing unit; R2 is the resistance value of the vertical sensing unit; R3 is the resistance value of the inflection point sensing unit; ρ is the resistivity; w1 is the line segment width of the horizontal sensing unit; w2 is the line segment width of the vertical sensing unit; w3 is the line segment width of the inflection point sensing unit; r is the inflection point radius; d is the inflection point diameter; L1 is the line segment length of the horizontal sensing unit; L2 is the line segment length of the inflection point sensing unit; L3 is the line segment length of the vertical sensing unit; h1 is the line segment thickness of the horizontal sensing unit; h2 is the line segment thickness of the vertical sensing unit; h3 is the line segment thickness of the inflection point sensing unit;

[0024] The calculation formula for signal response change is determined according to the calculation formulas for the resistance values ​​of the horizontal sensing unit, the vertical sensing unit and the inflection point sensing unit; the calculation formula for signal response change is:

[0025]

[0026] Among them, ΔR 1,2,3 is the signal response change; set is the preset strain variable; ε is the strain variable integral variable; ν LM is the Poisson's ratio of the gallium-indium alloy material;

[0027] The calculation formula of the relative resistance change rate is determined according to the calculation formula of the signal response change; the calculation formula of the relative resistance change rate is:

[0028]

[0029] in, is the relative resistance change rate;

[0030] The strain signal response model is determined according to the calculation formula of the relative resistance change rate.

[0031] Preferably, a sheep status identification method based on respiratory monitoring comprises:

[0032] The real-time resistance signal of the target sheep is collected by using the stretchable strain sensor, and the real-time resistance signal is subjected to signal correction and filtering to obtain a breathing curve;

[0033] Calculating the waveform characteristics of the respiratory curve to obtain the waveform characteristic data;

[0034] The state of the target sheep is judged by using a logic evaluation truth table according to the waveform characteristic data to obtain a target recognition result; the target recognition result includes: a healthy state, a tired state, and a sick state.

[0035] Preferably, the real-time resistance signal of the target sheep is collected by using the stretchable strain sensor, and the real-time resistance signal is subjected to signal correction and filtering to obtain a breathing curve, including:

[0036] Measuring the initial resistance of the stretchable strain sensor under no-strain conditions to obtain a reference value;

[0037] Using the stretchable strain sensor to obtain the real-time resistance signal generated by chest expansion or contraction during the breathing process of the target sheep;

[0038] Calculating the deviation between the real-time resistance signal and the reference value to obtain an error term;

[0039] The average value of the error terms in adjacent items is used as the calibration point to obtain the chest circumference influencing factor;

[0040] The real-time resistance signal is corrected using the chest circumference influence factor to obtain a corrected resistance signal; the calculation formula of the corrected resistance signal is:

[0041]

[0042] Among them, R (i) is the resistance value after correction; R (j) To calibrate the resistance value; R (t) is the resistance value at the tth moment; R0 is the initial resistance value;

[0043] Performing multi-layer filtering on the correction resistance signal to obtain a filtered resistance signal;

[0044] Comparing the reference value with the filter resistance signal to obtain a real-time resistance change rate;

[0045] The breathing curve is obtained by forming a graph according to the real-time resistance change rate and the breathing time.

[0046] Preferably, performing waveform feature calculation on the respiratory curve to obtain the waveform feature data includes:

[0047] Determine the time difference between two adjacent trough values ​​of the respiratory curve as a respiratory cycle;

[0048] The breathing depth is calculated according to the breathing curve; the calculation formula of the breathing depth is:

[0049]

[0050] Among them, RD (t) is the breathing depth; R max is the maximum value of the breathing curve; R min is the minimum value of the breathing curve;

[0051] The respiratory frequency is calculated according to the respiratory curve; the calculation formula of the respiratory frequency is:

[0052]

[0053] Among them, RR (t) is the respiratory rate; t i is the moment of the peak in the breathing curve in the i-th breathing cycle;

[0054] The breathing depth and the breathing frequency are integrated to obtain the waveform characteristic data.

[0055] Preferably, judging the state of the target sheep according to the waveform characteristic data to obtain a target recognition result includes:

[0056] Extracting the number of occurrences of the normal breathing state, the deep breathing state and the shallow breathing state in the waveform characteristic data according to the preset normal breathing value; the range of the normal breathing state is: between 95% and 105% of the normal breathing value; the range of the deep breathing state is: greater than 105% of the normal breathing value; the range of the shallow breathing state is: less than 95% of the normal breathing value;

[0057] Inputting the occurrence times of the normal breathing state, the deep breathing state and the shallow breathing state into the logic evaluation truth table to obtain a logic result;

[0058] When the logic result is 0 / 0 / 0, determining that the target recognition result is the health state;

[0059] When the logic result is any one of 1 / 0 / 0, 1 / 0 / 1 or 1 / 1 / 0, determining that the target recognition result is the fatigue state;

[0060] When the logical result is 1 / 1 / 1, it is determined that the target recognition result is the disease state.

[0061] The present invention discloses the following technical effects:

[0062] The present invention provides a sheep state identification device and method based on respiratory monitoring. Through a stretchable strain sensor with better biocompatibility and electrical performance, the defect of excessive intervention on sheep in traditional technology is solved, and real-time monitoring and feedback of sheep are realized; sheep state judgment is performed through a logical evaluation truth table, which solves the defect of manual judgment in traditional technology and realizes the function of directly identifying the sheep state through the waveform characteristic data. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0064] Figure 1 An example diagram of a sheep state identification device based on respiratory monitoring provided by an embodiment of the present invention;

[0065] Figure 2 A module diagram of a sheep state identification device based on respiratory monitoring provided by an embodiment of the present invention;

[0066] Figure 3 A schematic diagram of the preparation process of a stretchable strain sensor provided in an embodiment of the present invention;

[0067] Figure 4 A schematic diagram of a breathing curve provided by an embodiment of the present invention;

[0068] Figure 5 A detailed flowchart provided for an embodiment of the present invention;

[0069] Description of reference numerals:

[0070] 1-stretchable strain sensor, 2-circuit module, 21-signal acquisition submodule, 22-signal processing submodule, 23-wireless signal transmission submodule, 24-display submodule, 221-signal filtering unit, 222-signal amplification unit, 223-digital-to-analog conversion unit, 3-other components. DETAILED DESCRIPTION

[0071] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0072] The purpose of the present invention is to provide a sheep status identification device and method based on respiratory monitoring, which realizes real-time monitoring and feedback of sheep through a stretchable strain sensor with better biocompatibility and electrical performance, reduces manufacturing costs and excessive intervention on sheep, and improves the accuracy of signal acquisition; and judges the sheep status through a logical evaluation truth table to improve the accuracy and reliability of the identification result.

[0073] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0074] Figure 1 This is an example diagram of a sheep status identification device based on respiratory monitoring provided by an embodiment of the present invention. Figure 2 A module diagram of a sheep state identification device based on respiratory monitoring provided by an embodiment of the present invention, such as Figure 1 and Figure 2 As shown, the present invention provides a sheep state recognition device based on respiratory monitoring, comprising: a stretchable strain sensor 1, a circuit module 2 and other components 3; the stretchable strain sensor 1 comprises: a flexible substrate layer, a liquid metal sensitive layer and a top packaging layer distributed in a sandwich structure; the circuit module 2 comprises: a signal acquisition submodule 21, a signal processing submodule 22, a wireless signal transmission submodule 23 and a display submodule 24 connected in sequence; the other components 3 comprise: a protective shell, an elastic bandage, a lithium battery and a wire;

[0075] The stretchable strain sensor 1 is connected to the signal acquisition submodule 21;

[0076] The signal acquisition submodule 21 is used to collect the real-time resistance signal of the chest expansion and contraction changes of the target sheep during breathing; the signal processing submodule 22 is used to pre-process the original resistance signal to obtain waveform characteristic data; the wireless signal transmission submodule 23 is used to transmit the waveform characteristic data to the display submodule 24 or the target device; the display submodule 24 is used to display the waveform characteristic data; the elastic strap is used to fix the stretchable strain sensor 1.

[0077] refer to Figure 3 , the preparation process of the flexible substrate layer includes:

[0078] The polydimethylsiloxane solution and the curing agent were mixed in a ratio of 10:1, and stirred at room temperature for 20 minutes using a power-enhanced electric stirrer to obtain a solution A;

[0079] Dissolve dopamine in tris(hydroxymethylaminomethane) buffer, stir with a magnetic stirrer at room temperature for 20 minutes, and adjust the pH value of the solution to 8.5 to obtain solution B;

[0080] Solution A is mixed with solution B, so that solution B is evenly distributed in the network of solution A, and a double network structure is generated by initiating a polymerization reaction of polydopamine with an initiator of potassium persulfate to obtain a mixed solution;

[0081] The mixed solution was coated on a glass sheet with a surface tension lower than 24 mN / m, and spin-coated at 1600 rpm for 10 minutes using an automatic spin-coating machine to obtain an original film with a thickness of 1 mm;

[0082] The original film was placed in a vacuum drying oven and dried at 60° C. for 20 minutes to obtain a flexible substrate layer.

[0083] refer to Figure 3 , the preparation process of the liquid metal sensitive layer includes:

[0084] The dimensional parameters are obtained using a pre-designed strain signal response model;

[0085] Using a 3D printer to create a conductive channel mold on the flexible base layer according to size parameters;

[0086] Injecting the gallium-indium alloy into the conductive channel mold, maintaining the preset temperature environment, and drying at room temperature for 30 minutes after the gallium-indium alloy is evenly distributed in the conductive channel mold to obtain a serpentine conductive path;

[0087] The two wires are respectively connected to the starting point and the end point of the serpentine conductive path to obtain a liquid metal sensitive layer.

[0088] Preferably, the process of constructing the strain signal response model includes:

[0089] The serpentine conductive path is disassembled into sensing units, and the sensing units are divided into horizontal sensing units, vertical sensing units and inflection point sensing units according to the relationship between position and strain;

[0090] Determine the resistance value calculation formulas of the horizontal sensing unit, the vertical sensing unit and the inflection point sensing unit; the resistance value calculation formulas of the horizontal sensing unit, the vertical sensing unit and the inflection point sensing unit are respectively:

[0091] as well as

[0092] Wherein, R1 is the resistance value of the horizontal sensing unit; R2 is the resistance value of the vertical sensing unit; R3 is the resistance value of the inflection point sensing unit; ρ is the resistivity; w1 is the line segment width of the horizontal sensing unit; w2 is the line segment width of the vertical sensing unit; w3 is the line segment width of the inflection point sensing unit; r is the inflection point radius; d is the inflection point diameter; L1 is the line segment length of the horizontal sensing unit; L2 is the line segment length of the inflection point sensing unit; L3 is the line segment length of the vertical sensing unit; h1 is the line segment thickness of the horizontal sensing unit; h2 is the line segment thickness of the vertical sensing unit; h3 is the line segment thickness of the inflection point sensing unit;

[0093] The calculation formula for the signal response change is determined according to the calculation formula for the resistance values ​​of the horizontal sensing unit, the vertical sensing unit and the inflection point sensing unit; the calculation formula for the signal response change is:

[0094]

[0095] Among them, ΔR 1,2,3 is the signal response change; ε set is the preset strain variable; ε is the strain variable integral variable; ν LM is the Poisson's ratio of the gallium-indium alloy material;

[0096] The calculation formula for the relative resistance change rate is determined based on the calculation formula for the signal response change; the calculation formula for the relative resistance change rate is:

[0097]

[0098] in, is the relative resistance change rate;

[0099] The strain signal response model is determined according to the calculation formula of the relative resistance change rate.

[0100] Preferably, a sheep status identification method based on respiratory monitoring comprises:

[0101] The real-time resistance signal of the target sheep is collected by using the stretchable strain sensor 1, and the real-time resistance signal is corrected and filtered to obtain a breathing curve;

[0102] Calculating waveform characteristics of the respiratory curve to obtain waveform characteristic data;

[0103] The state of the target sheep is judged by using a logic evaluation truth table according to the waveform feature data to obtain a target recognition result; the target recognition result includes: healthy state, tired state and sick state.

[0104] Preferably, the stretchable strain sensor 1 is used to collect the real-time resistance signal of the target sheep, and the real-time resistance signal is subjected to signal correction and filtering to obtain a breathing curve, including:

[0105] Measuring the initial resistance of the stretchable strain sensor 1 under a no-strain condition to obtain a reference value;

[0106] The stretchable strain sensor 1 is used to obtain a real-time resistance signal generated by chest expansion or contraction during the breathing process of the target sheep;

[0107] Calculate the deviation between the real-time resistance signal and the reference value to obtain an error term;

[0108] The average value of the adjacent error terms is used as the calibration point to obtain the chest circumference influencing factor;

[0109] The real-time resistance signal is corrected using the chest circumference influence factor to obtain a corrected resistance signal; the calculation formula for the corrected resistance signal is:

[0110]

[0111] Among them, R (i) is the resistance value after correction; R (j) To calibrate the resistance value; R (t) is the resistance value at the tth moment; R0 is the initial resistance value;

[0112] Performing multi-layer filtering on the correction resistance signal to obtain a filtered resistance signal;

[0113] Compare the reference value with the filtered resistance signal to obtain the real-time resistance change rate;

[0114] The breathing curve is obtained by composing a graph according to the real-time resistance change rate and breathing time.

[0115] Preferably, waveform feature calculation is performed on the respiratory curve to obtain waveform feature data, including:

[0116] The time difference between two adjacent trough values ​​of the respiratory curve is determined as a respiratory cycle;

[0117] Calculate the breathing depth based on the breathing curve; the calculation formula for breathing depth is:

[0118]

[0119] Among them, RD (t) is the breathing depth; R max is the maximum value of the breathing curve; R min is the minimum value of the breathing curve;

[0120] Calculate the respiratory rate based on the respiratory curve; the calculation formula for the respiratory rate is:

[0121]

[0122] Among them, RR (t) is the respiratory rate; t i is the moment of the peak of the respiratory curve in the i-th respiratory cycle;

[0123] Integrate breathing depth and breathing frequency to obtain waveform characteristic data.

[0124] Preferably, judging the state of the target sheep according to the waveform characteristic data to obtain the target recognition result includes:

[0125] The number of occurrences of normal breathing state, deep breathing state and shallow breathing state in the waveform feature data is extracted according to the preset normal breathing value; the range of normal breathing state is: between 95% and 105% of the normal breathing value; the range of deep breathing state is: greater than 105% of the normal breathing value; the range of shallow breathing state is: less than 95% of the normal breathing value; (reference Figure 4 )

[0126] Input the occurrence times of the normal breathing state, the deep breathing state and the shallow breathing state into the logic evaluation truth table to obtain the logic result;

[0127] When the logical result is 0 / 0 / 0, the target identification result is determined to be in a healthy state;

[0128] When the logic result is any one of 1 / 0 / 0, 1 / 0 / 1 or 1 / 1 / 0, the target identification result is determined to be in a fatigue state;

[0129] When the logical result is 1 / 1 / 1, it is determined that the target recognition result is a diseased state.

[0130] Preferably, the sheep status identification device based on respiratory monitoring provided in this embodiment is small in size, light in weight, easy to wear and has high test accuracy. The tested individuals, especially sheep in breeding environments, show high tolerance. The preparation method of the stretchable strain sensor 1 is: prepare a flexible substrate layer, prepare a liquid metal sensitive layer, and encapsulate the sensor in sequence; the circuit module 2 includes a signal acquisition submodule 21, a signal processing submodule 22, a wireless signal transmission submodule 23 and a display submodule 24, which is used to convert the signal output by the stretchable strain sensor 1 into visual data; other components 3 are necessary components for sheep to wear and the operation of the device, including a protective shell, an elastic strap, a lithium battery and a wire.

[0131] Preferably, the protective shell is used to protect the circuit module 2 to prevent physical damage and environmental factors. The elastic band is used to fix the stretchable strain sensor 1 so that it can change with the abdominal contraction and expansion caused by the sheep's breathing, and accurately monitor the sheep's respiratory state. The lithium battery provides power for the circuit module 2 and is connected to the power input terminal of the circuit module 2 through a wire. The wire is used to connect the sensor to the circuit system, and the circuit system to the lithium battery. The stretchable strain sensor 1 is electrically connected to the signal acquisition submodule 21 to transmit the deformation signal caused by the monitored sheep's breathing. The signal acquisition submodule 21 is electrically connected to the signal processing submodule 22 to transmit the converted digital signal. The signal processing submodule 22 is electrically connected to the wireless signal transmission submodule 23 to transmit the processed digital signal. The wireless signal transmission submodule 23 is wirelessly connected to the display submodule 24 to realize remote data transmission and display.

[0132] Specifically, the preparation process of the flexible substrate layer includes: mixing the polydimethylsiloxane solution and the curing agent in a ratio of 10:1, applying the mixture on an easily peelable glass sheet after sufficient stirring, and forming a 1mm thick film after spin coating. Then drying is performed to prepare the flexible substrate layer of the stretchable strain sensor 1; stirring includes using a power-enhanced electric stirrer at room temperature for 20 minutes; spin coating includes using an automatic spin coating machine at 1600 rpm for 10 minutes; and drying includes using a vacuum drying oven at 60°C for 20 minutes.

[0133] Preferably, a serpentine conductive path is prepared and external wires are led out, specifically comprising: using a high-precision 3D printer to create a mold of a conductive channel on a flexible substrate layer, then injecting liquid metal into the mold and drying it at room temperature for 30 minutes to obtain a serpentine conductive path; external wires are led out from the starting point and end point of the serpentine conductive path to obtain a liquid metal layer; the liquid metal is a gallium-indium alloy, and the gallium-indium alloy needs to be heated to an appropriate temperature to maintain its fluidity.

[0134] Specifically, the size parameters of the serpentine conductive path are estimated by the strain signal response model, and the process includes: first, the serpentine conductive path of the stretchable strain sensor 1 is disassembled into multiple sensing units, each sensing unit is regarded as an independent resistor element, and its resistance value is estimated. Next, the signal response change of each resistor element is determined according to the preset strain. Then, the overall resistance of the serpentine conductive path composed of each sensing unit is summarized, and the relative resistance change rate is calculated in combination with the signal response change, thereby constructing a strain signal response model. Finally, the model is input according to actual needs to obtain specific size parameters.

[0135] Furthermore, the plurality of sensing units are classified according to the position and strain relationship, and include a horizontal sensing unit, a vertical sensing unit and an inflection point sensing unit respectively; the resistance value calculation formulas of the horizontal sensing unit, the vertical sensing unit and the inflection point sensing unit are as follows:

[0136]

[0137] The calculation formula for determining the change in signal response of each resistor element according to the preset strain is as follows:

[0138]

[0139] The calculation formula for the relative resistance change rate is:

[0140]

[0141] Optionally, the sizes of the multiple sensing units are dimensionless and can be proportionally enlarged or reduced according to application conditions. However, due to process conditions, the sensing units cannot be infinitely reduced.

[0142] Preferably, the sensor is packaged using the same material and method as the flexible base layer. The same type of base layer and packaging layer can ensure that the stretchable strain sensor 1 produces consistent deformation when stretched during breathing, thereby ensuring and improving the linearity of the stretchable strain sensor 1.

[0143] Specifically, the circuit module 2 includes a signal acquisition submodule 21, a signal processing submodule 22, a wireless signal transmission submodule 23 and a display submodule 24, and is used to convert the signal output by the stretchable strain sensor 1 into visual data;

[0144] Optionally, other components 3 are necessary components for sheep to wear and for the device to operate, including a protective housing, an elastic strap, a lithium battery and a wire;

[0145] Specifically, the signal acquisition submodule 21 provides an interface connected to the stretchable strain sensor 1, which is used to receive the resistance signal of the stretchable strain sensor 1 as the chest expands / contracts during the sheep's breathing process; the signal processing submodule 22 is a component responsible for further processing the original resistance signal collected by the signal acquisition submodule 21, and enhances the signal through a preamplifier so that the resistance signal reaches an intensity that can be further processed. The wireless signal transmission submodule 23 is a component responsible for wirelessly sending the processed digital signal to the display submodule 24 or other receiving devices, and wireless transmission of the data after signal processing is achieved by adopting wireless communication technology. The display submodule 24 is a component for intuitively displaying the sheep's breathing data, providing an intuitive user interface to display key parameters such as the sheep's breathing rate and breathing depth.

[0146] Furthermore, the signal processing module further includes: a signal filtering unit 221 , a signal amplifying unit 222 and a digital-to-analog conversion unit 223 .

[0147] refer to Figure 5 This embodiment provides a method for identifying the state of sheep based on respiratory monitoring, using a stretchable strain sensor 1 as a measurement node. Its tensile deformation changes with the breathing process, but the electrical signal caused by the tensile deformation inevitably has motion artifacts or even drift. Signal processing methods such as filtering are used to optimize the real-time signal and extract the respiratory frequency and depth information of the individual being tested. These important vital signs provide an important basis for identifying the individual state of sheep. The following steps are used to identify the state of sheep:

[0148] S1, collecting real-time resistance signals and performing signal correction and filtering on the real-time resistance signals to obtain a breathing curve;

[0149] S2, calculating the waveform characteristics of the respiratory curve to obtain waveform characteristic data;

[0150] S3, judging the state of the tested individual according to the waveform characteristic data;

[0151] Further, in step S1, the real-time resistance signal is corrected and filtered to obtain a breathing curve, specifically including: measuring the initial resistance of the stretchable strain sensor 1 under strain-free conditions, recorded as a preset reference value; obtaining the real-time resistance signal generated by chest expansion / contraction during the sheep's breathing process in real time through a wearable respiratory monitoring device; obtaining an error term based on the preset reference value and the real-time resistance signal; using the average of two adjacent error terms as a calibration point to obtain a chest circumference influence factor for the tested individual; performing signal correction on the real-time resistance signal based on the chest circumference influence factor; using multi-layer filtering to obtain a filtered breathing signal; comparing the preset reference value with the filtered breathing signal to obtain a real-time resistance change rate; and obtaining a breathing curve based on the real-time resistance change rate and the breathing time, with the breathing time as the horizontal axis and the real-time resistance change rate as the vertical axis.

[0152] Specifically, the calculation formula for correcting the real-time resistance signal using the chest circumference influence factor is:

[0153]

[0154] refer to Figure 5The multi-layer filtering process adopts a processing method combining two types of filtering algorithms, wherein the two types of filtering algorithms are respectively recorded as the first filtering algorithm and the second filtering algorithm. Specifically, it includes: calculating the signal-to-noise ratio according to the respiratory signal obtained after processing by the first filtering algorithm, inputting the first signal-to-noise ratio into the first comparison discriminator to determine whether the data quality requirements are met; if the signal quality requirements are met, further extracting the signal waveform features; if the signal quality requirements are not met, using the second filtering algorithm; inputting the second signal-to-noise ratio into the second comparison discriminator to determine whether the data quality requirements are met; if the signal quality requirements are met, further extracting the signal waveform features; if the signal quality requirements are not met, continuing to execute the first filtering algorithm. This cycle is repeated until the signal waveform features are extracted after the signal quality requirements are met.

[0155] Specifically, the outputs of the first and second comparison discriminators are whether the data quality requirements are met;

[0156] Preferably, the calculation formula of the first filtering algorithm is:

[0157] J [ w(i) ] =E[e 2 (i)] = E{[d(i)-w H (i)x(i)] 2}

[0158] Where J[w(i)] is the signal error, w(i) is the filter proportional parameter vector, and w H (i) is the conjugate transpose of w(i), e 2 (i) is the error signal vector, d(i) is the expected output signal vector, x(i) is the input signal vector, and E{} is the mean square value operation.

[0159] Specifically, the second filtering algorithm is a superposition performed on the basis of the first filtering algorithm, and is an estimation method that converges to minimization through continuous feedback calls.

[0160] Preferably, in order to reduce the influence of noise and random fluctuations and improve the stability and reliability of the data, the signal-corrected data is smoothed before filtering;

[0161] Further, in step S2, the waveform characteristics of the breathing curve are calculated to obtain waveform characteristic data, specifically including that the breathing curve has a certain number of peaks and troughs within a certain time interval, two adjacent troughs represent a single breathing process, and the time interval T between two adjacent trough values ​​is a single breathing time; according to the single breathing process, the value of the peak in the single breathing process is extracted to obtain the breathing depth RD (t) ;According to the single breathing time, extract the number of peaks per minute to obtain the respiratory rate RR (t);

[0162] Specifically, the depth of breathing refers to the maximum extent to which the lungs are inflated or deflated during breathing. It is often used to describe the efficiency of breathing and the ventilation capacity of the lungs. The depth of breathing is calculated based on the respiratory waveform, and the calculation formula is as follows:

[0163]

[0164] Preferably, the respiratory rate refers to the number of breaths per minute. Under normal circumstances, the respiratory rate of a 3-month-old ewe is 12 to 20 times per minute. Monitoring the respiratory rate is of great significance for the determination of respiratory diseases. Low vitality can cause a low respiratory rate; cold and heat stress can speed up breathing; dysentery, pneumonia, respiratory center depression, etc. can also cause abnormal breathing.

[0165] The respiratory frequency is calculated based on the respiratory waveform. The calculation formula is as follows:

[0166]

[0167] By calculating the interval between adjacent peaks, the average time of one breath is obtained, and then the number of breaths within 60 seconds is calculated to obtain the respiratory frequency.

[0168] Further, in step S3, the state of the individual under test is judged according to the waveform characteristic data, specifically including judging the respiratory state of the sheep in real time based on the normal breathing value as the threshold value. When the peak value within a single breathing time is between 95% and 105% of the normal breathing value, it is judged as a normal breathing state. If it is higher than 105% of the normal breathing value, it is judged as a deep breathing state. If it is lower than 95% of the normal breathing value, it is judged as a shallow breathing state. The number of normal breathing states, deep breathing states and shallow breathing states per unit time are calculated to judge the state of the sheep.

[0169] Preferably, the status of the individual being tested includes three categories, namely healthy, tired and sick.

[0170] Furthermore, the number of normal breathing states, deep breathing states and shallow breathing states in a unit time is calculated, and the state of the sheep is judged specifically including: inputting the number of normal breathing states, deep breathing states and shallow breathing states into the logic evaluation truth table, wherein the logic judgment is "0" and "1". The logic result is set as: "0 / 0 / 0" indicates a healthy state; "1 / 0 / 0", "1 / 0 / 1" and "1 / 1 / 0" indicate a tired state; "1 / 1 / 1" indicates an ill state. Based on 1 minute as the unit time, the individual states are classified according to the critical values ​​of normal breathing (10 times), deep breathing (5 times) and shallow breathing (10 times), and the truth table example of the logic evaluation is shown in Table 1:

[0171] Table 1

[0172] Normal breathing rate Number of deep breaths Shallow breathing rate Physical condition ＞10 ＜5 ＜10 healthy ＞10 5＞ ＜10 exhausted ＞10 ＜5 ＞10 exhausted ＜10 5＞ ＞10 Illness

[0173] Preferably, the present embodiment provides a sheep state identification device based on respiratory monitoring for sheep, through which the information caused by breathing can be monitored in real time. In addition, the sensor in the device is a stretchable strain sensor 1, which has good flexibility and sensitivity, is resistant to bending, can be disassembled and reused at any time, has a simple preparation process, and has high sensitivity while being flexible. The breathing depth and breathing frequency are obtained according to the changes in the chest cavity during the breathing process. And when the strain sensor is made into a rectangle, it meets the wearing conditions of the current elastic bandage. It has a good fit with the sheep's body shape, does not produce a strong foreign body sensation, has high comfort, good biocompatibility, and more accurate monitoring data. The present embodiment also provides a sheep state identification method, using the above-mentioned wearable respiratory monitoring device for respiratory measurement, the monitoring method is simple and effective, with a flexible and stickable strain sensor as a measurement unit, and its resistance changes with the change of the breathing process. The respiratory signal is obtained by correcting and filtering the real-time resistance value, and the respiratory frequency and respiratory intensity information of the measured individual can be monitored in real time and obtained. Thanks to the proposed sheep state recognition method, there is no need to adjust different tested individuals, and accurate measurements can be performed adaptively for different users. After the tested person starts exercising, the real-time resistance value of the stretchable strain sensor 1 can be obtained, and the real-time resistance value is processed and compared with the preset reference resistance value to obtain the real-time resistance change rate. Then, with the breathing time as the horizontal axis and the resistance change rate as the vertical axis, the breathing curve of the stretchable strain sensor 1 is obtained, and multiple resistance change rate peak values ​​and trough values ​​are formed on the breathing curve, and a single breath is represented by two adjacent trough values. The time interval T between the two adjacent trough values ​​is the single breathing time, and the breathing frequency is 1 / T. The breathing state can also be judged in real time according to the preset normal breathing value.

[0174] The beneficial effects of the present invention are as follows:

[0175] Compared with traditional respiratory rate detection technology, the present invention avoids the subjectivity of human eye observation; and the design is more suitable for sheep to wear, reducing the discomfort and tension of the individual being measured during the measurement process: the respiratory rate of the individual being measured can be dynamically detected for a long time without restricting the activity of the individual being measured; in addition, the stretchable strain sensor improves the measurement accuracy, can monitor the signal in real time during exercise, is low-cost, easy to disassemble, safe and without hidden dangers, simple to operate and highly sensitive.

[0176] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0177] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A sheep status identification device based on respiratory monitoring, characterized in that: include: Stretchable strain sensors, circuit modules and other components; The stretchable strain sensor comprises: a flexible substrate layer, a liquid metal sensitive layer and a top packaging layer distributed in a sandwich structure; the circuit module comprises: a signal acquisition submodule, a signal processing submodule, a wireless signal transmission submodule and a display submodule connected in sequence; the other components comprise: a protective housing, an elastic band, a lithium battery and a wire; The stretchable strain sensor is connected to the signal acquisition submodule; The signal acquisition submodule is used to collect real-time resistance signals of chest expansion and contraction changes during the breathing process of the target sheep; the signal processing submodule is used to preprocess the original resistance signal to obtain waveform characteristic data; the wireless signal transmission submodule is used to transmit the waveform characteristic data to the display submodule or the target device; the display submodule is used to display the waveform characteristic data; the elastic strap is used to fix the stretchable strain sensor.

2. A sheep status identification device based on respiratory monitoring according to claim 1, characterized in that: The preparation process of the flexible substrate layer comprises: The polydimethylsiloxane solution and the curing agent were mixed in a ratio of 10:1, and stirred at room temperature for 20 minutes using a power-enhanced electric stirrer to obtain a solution A; Dissolve dopamine in tris(hydroxymethylaminomethane) buffer, stir with a magnetic stirrer at room temperature for 20 minutes, and adjust the pH value of the solution to 8.5 to obtain solution B; The solution A is mixed with the solution B, so that the solution B is evenly distributed in the network of the solution A, and a double network structure is generated by initiating a polymerization reaction of polydopamine with an initiator of potassium persulfate to obtain a mixed solution; The mixed solution was applied on a glass sheet with a surface tension lower than 24 mN / m, and spin-coated at 1600 rpm for 10 minutes using an automatic spin-coating machine to obtain an original film with a thickness of 1 mm; The original film was placed in a vacuum drying oven at 60° C. and dried for 20 minutes to obtain the flexible substrate layer.

3. The sheep status identification device based on respiratory monitoring according to claim 1, characterized in that: The preparation process of the liquid metal sensitive layer includes: The dimensional parameters are obtained using a pre-designed strain signal response model; Using a 3D printer to create a conductive channel mold on the flexible base layer according to the size parameters; Injecting the gallium-indium alloy into the conductive channel mold, maintaining a preset temperature environment, and drying at room temperature for 30 minutes after the gallium-indium alloy is evenly distributed in the conductive channel mold to obtain a serpentine conductive path; The two wires are respectively connected to the starting point and the end point of the serpentine conductive path to obtain the liquid metal sensitive layer.

4. The sheep state identification device based on respiratory monitoring according to claim 3 is characterized in that: The construction process of the strain signal response model includes: Disassembling the serpentine conductive path into sensing units, and dividing the sensing units into horizontal sensing units, vertical sensing units and inflection point sensing units according to the relationship between positions and strains; Determine the resistance value calculation formulas of the horizontal sensing unit, the vertical sensing unit and the inflection point sensing unit; the resistance value calculation formulas of the horizontal sensing unit, the vertical sensing unit and the inflection point sensing unit are respectively: as well as Wherein, R1 is the resistance value of the horizontal sensing unit; R2 is the resistance value of the vertical sensing unit; R3 is the resistance value of the inflection point sensing unit; ρ is the resistivity; w1 is the line segment width of the horizontal sensing unit; w2 is the line segment width of the vertical sensing unit; w3 is the line segment width of the inflection point sensing unit; r is the inflection point radius; d is the inflection point diameter; L1 is the line segment length of the horizontal sensing unit; L2 is the line segment length of the inflection point sensing unit; L3 is the line segment length of the vertical sensing unit; h1 is the line segment thickness of the horizontal sensing unit; h2 is the line segment thickness of the vertical sensing unit; h3 is the line segment thickness of the inflection point sensing unit; The calculation formula for signal response change is determined according to the calculation formulas for the resistance values ​​of the horizontal sensing unit, the vertical sensing unit and the inflection point sensing unit; the calculation formula for signal response change is: Wherein, ΔR1,2,3 are the signal response changes; ε set is the preset strain variable; ε is the strain variable integral variable; ν LM is the Poisson's ratio of the gallium-indium alloy material; The calculation formula of the relative resistance change rate is determined according to the calculation formula of the signal response change; the calculation formula of the relative resistance change rate is: in, is the relative resistance change rate; The strain signal response model is determined according to the calculation formula of the relative resistance change rate.

5. A sheep status identification method based on respiratory monitoring, characterized in that: The method for identifying sheep status based on respiratory monitoring as described in claim 1 comprises: The real-time resistance signal of the target sheep is collected by using the stretchable strain sensor, and the real-time resistance signal is subjected to signal correction and filtering to obtain a breathing curve; Calculating the waveform characteristics of the respiratory curve to obtain the waveform characteristic data; The state of the target sheep is judged by using a logic evaluation truth table according to the waveform characteristic data to obtain a target recognition result; the target recognition result includes: a healthy state, a tired state, and a sick state.

6. A sheep state identification method based on respiratory monitoring according to claim 5, characterized in that: The real-time resistance signal of the target sheep is collected by using the stretchable strain sensor, and the real-time resistance signal is corrected and filtered to obtain a breathing curve, including: Measuring the initial resistance of the stretchable strain sensor under no-strain conditions to obtain a reference value; Using the stretchable strain sensor to obtain the real-time resistance signal generated by chest expansion or contraction during the breathing process of the target sheep; Calculating the deviation between the real-time resistance signal and the reference value to obtain an error term; The average value of the error terms in adjacent items is used as the calibration point to obtain the chest circumference influencing factor; The real-time resistance signal is corrected using the chest circumference influence factor to obtain a corrected resistance signal; the calculation formula of the corrected resistance signal is: Among them, R (i) is the resistance value after correction; R (j) To calibrate the resistance value; R (t) is the resistance value at the tth moment; R0 is the initial resistance value; Performing multi-layer filtering on the correction resistance signal to obtain a filtered resistance signal; Comparing the reference value with the filter resistance signal to obtain a real-time resistance change rate; The breathing curve is obtained by forming a graph according to the real-time resistance change rate and the breathing time.

7. A sheep status identification method based on respiratory monitoring according to claim 6, characterized in that: Calculating the waveform characteristics of the breathing curve to obtain the waveform characteristic data includes: Determine the time difference between two adjacent trough values ​​of the respiratory curve as a respiratory cycle; The breathing depth is calculated according to the breathing curve; the calculation formula of the breathing depth is: Among them, RD (t) is the breathing depth; R max is the maximum value of the respiratory curve; R min is the minimum value of the breathing curve; The respiratory frequency is calculated according to the respiratory curve; the calculation formula of the respiratory frequency is: Among them, RR (t) is the respiratory rate; t i is the moment of the peak in the breathing curve in the i-th breathing cycle; The breathing depth and the breathing frequency are integrated to obtain the waveform characteristic data.

8. The sheep state identification method based on respiratory monitoring according to claim 6, characterized in that: The state of the target sheep is determined according to the waveform characteristic data to obtain a target recognition result, including: Extracting the number of occurrences of the normal breathing state, the deep breathing state and the shallow breathing state in the waveform characteristic data according to the preset normal breathing value; the range of the normal breathing state is: between 95% and 105% of the normal breathing value; the range of the deep breathing state is: greater than 105% of the normal breathing value; the range of the shallow breathing state is: less than 95% of the normal breathing value; Inputting the occurrence times of the normal breathing state, the deep breathing state and the shallow breathing state into the logic evaluation truth table to obtain a logic result; When the logic result is 0 / 0 / 0, determining that the target recognition result is the health state; When the logic result is any one of 1 / 0 / 0, 1 / 0 / 1 or 1 / 1 / 0, determining that the target recognition result is the fatigue state; When the logical result is 1 / 1 / 1, it is determined that the target recognition result is the disease state.

Images