WO2021196494A1

WO2021196494A1 - Livestock and poultry breeding intelligent weighing method and system

Info

Publication number: WO2021196494A1
Application number: PCT/CN2020/108630
Authority: WO
Inventors: 杨卫; 蔡杰; 颜丙乾
Original assignee: 江苏深农智能科技有限公司
Priority date: 2020-03-29
Filing date: 2020-08-12
Publication date: 2021-10-07
Also published as: CN111428202B; CN111428202A

Abstract

A livestock and poultry breeding intelligent weighing method and system. The method comprises: S1, carry out measurement to obtain a weight data array an of livestock and poultry; S2, calculating whether an is less than 50% of an empirical value, and if so, taking an as a weighing system error b; S3, calculating a difference value between an-b and the empirical value; S4, determining whether the quantity of data is within a set value range, if so, executing S5, and otherwise, executing S7; S5, performing standardization processing on data of the weight data array; S6, if the difference value in S3 is within a confidence interval, storing the weight data into a weight record array if so, and otherwise, taking an which is smaller than a lower limit value of the confidence interval as a new system error value; S7, determining whether the difference value in S3 is within a set error range of the empirical value, if so, storing the weight data into the weight record array, and otherwise, taking an which is smaller than the empirical value as a new system error value; and S8, outputting a data mean value in the weight record array within a set time period as the measured weight. The method can automatically generate and adjust an empirical value and a confidence interval according to site conditions.

Description

Intelligent weighing method and system for livestock and poultry breeding

Technical field

The invention relates to the field of livestock and poultry breeding, and more specifically, to an intelligent weighing method and system for livestock and poultry breeding.

Background technique

In the process of livestock and poultry breeding, it is necessary to continue to pay attention to the changes in the body weight of livestock and poultry, so as to conduct comparative analysis and optimize breeding. In the breeding process, factors such as livestock excrement, feed, and weighing position of livestock and poultry will have a great impact on the measurement accuracy. The invention patent with the patent number 201810073977.4 discloses an automatic weighing method and system for poultry. It obtains the weighing data of poultry in real time, and uses the limiting filter method to determine whether each weight data is valid, so as to realize the automatic acquisition of weighing data. Filtering the weight data to make the weighing more accurate. At the same time, the valid weight data, the weighing serial number corresponding to the valid weight data, and the weighing time corresponding to the valid weight data are uploaded to the server, which is convenient for subsequent processing, statistical analysis, and decision-making early warning. . The problems with this method are: 1. Because the weight gain of the entire batch needs to be supported by data in livestock and poultry breeding, it is not an accurate measurement of the weight of a single animal, so multiple animals are randomly measured, and then the entire batch is obtained. The breeding weight has practical application value. 2. The weighing device cannot automatically identify the foreign matter on the weighing platform and its impact cannot be cleared through data processing. 3. The weighing device of this method judges that the threshold is manually input, and the threshold cannot be automatically generated and adjusted according to the on-site situation.

Summary of the invention

The purpose of the present invention is to provide an intelligent weighing method for livestock and poultry breeding to overcome the shortcomings of the prior art.

In order to achieve the above objectives, the technical solutions adopted by the present invention are as follows:

An intelligent weighing method for livestock and poultry breeding includes the following steps:

S1. Measure the weight of livestock and poultry to obtain the weight data array a n of livestock and poultry;

S2. Calculate whether a n is less than 50% of the empirical value, and if so, use it as the weighing system error b;

S3. Calculate the difference between a n-b and the experience value in the weight data array;

S4. Determine whether the number of data n in the weight data array is within the set value range, if yes, execute step S5, otherwise, execute step S7;

S5. Standardize the data in the weight data array;

S6. Determine whether the difference between a nb and the experience value in the weight data array in step S3 is within the confidence interval. If so, store the corresponding weight data in the weight record array, otherwise it will be less than a n of the lower limit of the confidence interval As the new system error value b;

S7. Determine whether the difference between a nb and the experience value in the weight data array in step S3 is within the set error range of the experience value, if so, store the corresponding weight data in the weight record array, otherwise it will be less than the experience value A n as the new system error value b;

S8. The average value of the data in the weight record array within the set time period is output as the measured weight.

Further, the specific method for determining the empirical value in step S3 is: determining the standard body weight growth curve of the specific livestock and poultry breed according to the specific livestock and poultry breed, and the standard body weight growth curve is the empirical value.

Further, in the first measurement, if the measured weight data is significantly lower than the weight in the standard weight growth curve at the age of the day, it is judged as the basic error data, and the most recent error value is used as the system error amount.

Further, the set value in the step S4 is 20.

Further, the method for determining the confidence interval in step S6 is:

If the number of data n in the weight data array a n is greater than the set value, the most recently measured weight of multiple livestock and poultry in the weight data array a n is estimated in intervals according to the following formula:

or

In the formula, x is the average value of multiple livestock and poultry weight values, σ is the standard deviation of the weight values, n is the number of weight data used, and α is the significance level in significant statistics, generally 0.1 or 0.05, Zα/ 2 is the value of α/2 used to determine the mathematical distribution.

Further, the setting error range of the empirical value in the step S7 is ±20% of the empirical value.

Further, the step S8 specifically includes:

S81. Calculate the average value or mathematical expectation of the data in the weight record array of the last day to obtain the average weight of the livestock and poultry in the last hour;

S82. Summarize the hourly data of the day, perform linear fitting on the twenty-four hour data, and obtain the data at twelve o'clock as the weight value of the day.

The present invention also provides a system for realizing the above intelligent weighing method for livestock and poultry breeding, including:

Weighing sensor, used to measure the weight of livestock and poultry, to obtain the data array a n of livestock and poultry weight;

The calculation module is used to calculate the difference between a n-b and the empirical value in the weight data array;

The first judgment module is used to judge whether the number of data n in the weight data array is within the set value range, and if so, standardize the data in the weight data array;

At the same time, it is judged that the difference between a nb and the empirical value in the weight data array is within the confidence interval, if so, the corresponding weight data is stored in the weight record array, otherwise, a n that is less than the lower limit of the confidence interval is regarded as the new system error Value b;

The second judgment module is used to judge whether the difference between a nb in the weight data array and the experience value is within the setting of the experience value when the first judgment module judges that the number of data n in the weight data array is less than the minimum value of the set range. Within a certain error range, if yes, store the corresponding weight data in the weight record array, otherwise, use a n when it is less than the empirical value as the new system error value b;

The output module is used to output the average value of the data in the weight record array within the set time period as the measured weight.

Further, the weighing sensor is installed under the drinking fountain for livestock and poultry, and a weighing platform is provided on the weighing sensor 100.

Further, the load cell includes a pin connector P9, a resistor R22, a self-recovery fuse F5, a diode D3, a diode VD2, and a diode VD5. The first pin of the pin connector is grounded, and the pin connector The second pin is connected to one end of resistor R22, one end of resettable fuse F5, and the cathode of diode VD2. The other end of resistor R22 is connected to a +3.3V power supply. The anode of diode VD2 is grounded. The other end of resettable fuse F5 is connected to the anode of diode D3 and the cathode of diode VD5. , The sampling port of the single-chip microcomputer is connected, the cathode of diode D3 is connected to the +3.3V power supply connection, and the anode of diode VD5 is grounded.

Compared with the prior art, the present invention has the advantages that: the present invention can not only accurately measure the weight of a single animal, but also can measure multiple animals at random to obtain the breeding weight of the entire batch, which has practical application value. The invention can automatically identify the foreign matter on the weighing platform, regard it as a system error, and eliminate its influence through data processing. The invention can automatically generate and adjust the empirical value and the confidence interval according to the on-site situation.

Description of the drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

Fig. 1 is a flowchart of the intelligent weighing method for livestock and poultry breeding according to the present invention.

Figure 2 is a frame diagram of the intelligent weighing system for livestock and poultry breeding according to the present invention.

Fig. 3 is a circuit diagram of the weighing sensor in the intelligent weighing system for livestock and poultry breeding according to the present invention.

Detailed ways

The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, so that the protection scope of the present invention can be defined more clearly.

Referring to Figure 1, the present invention provides an intelligent weighing method for livestock and poultry breeding, which includes the following steps:

Step S1: Use the weighing sensor to measure the weight of the livestock and poultry, and obtain the weight data array a n of the livestock and poultry, where n is the number of measurements.

Step S2: Calculate whether a n is less than 50% of the empirical value, and if so, use it as the weighing system error b.

Step S3: Calculate the difference between a n-b and the empirical value in the weight data array, where b is the error caused by impurities on the sensor, and use an-b as the weight data for data processing.

For example, after five consecutive measurements, calculating the difference between a n-b and the empirical value in the weight data array includes: calculating a 5-b, a 4-b, a 3-b, a 2-b;

Wherein, the specific method for determining the empirical value in step S3 is as follows: since each type of livestock and poultry has different reference values, this embodiment can determine and obtain the standard weight of the livestock and poultry breeds according to the specific breeds of livestock and poultry. Growth curve, the standard weight growth curve is an empirical value.

In the first measurement, if the measured weight data is significantly lower than the weight in the standard weight growth curve at the age of the day, it is judged as the basic error data, and the most recent error value is used as the system error amount. In this way, when the next measurement data minus the system error amount is within the allowable error range of the standard weight (usually ±20%), the difference of the measurement data minus the system error amount is judged to be the current measured livestock weight.

Step S4: It is judged whether the number of data n in the weight data array is within the set value range, if yes, step S5 is executed, otherwise, step S7 is executed.

In this embodiment, the above-mentioned setting range may be a numerical value, such as 20, that is to say, it is determined whether the number of data n in the weight data array is greater than 20.

Step S5: Perform standardization processing on the data of the weight data array, generally using z-score standardization processing.

Step S6: Determine whether the difference between a nb and the experience value in the weight data array in step S3 is within the confidence interval. If so, store the corresponding weight data in the weight record array, otherwise it will be less than a of the lower limit of the confidence interval. n is used as the new system error value b to update the weighing system error b in step S2.

Specifically, the method for determining the confidence interval is:

or

Step S7: Determine whether the difference between a nb and the experience value in the weight data array in step S3 is within the set error range of the experience value, if so, store the corresponding weight data in the weight record array, otherwise it will be less than the experience value The a n at the time is used as the new system error value b to update the weighing system error b in step S2.

In this embodiment, the setting error range of the empirical value is ±20% of the empirical value.

Step S8: The average value of the data in the weight record array within the set time period is output as the measured weight. Specifically include: S81. Average or mathematical expectation of the data in the weight record array of the last day to obtain the average weight of livestock and poultry in the last hour; S82. Summarize the hourly data of the day and combine the data for twenty-four hours Perform a linear fit, and obtain the data at twelve o'clock as the weight value of the day.

Referring to FIG. 2, the present invention also provides a system for implementing the aforementioned intelligent weighing method for livestock and poultry breeding, including: a weighing sensor 100 for measuring the weight of livestock and poultry to obtain the weight data array a n of livestock and poultry, which can be The weighing sensor is fixed under the drinking fountain, and a weighing platform is provided on the weighing sensor 100; the calculation module 200 is used to calculate the difference between a nb and the empirical value in the weight data array; the first judgment module 300 is used to Determine whether the number of data n in the weight data array is within the set value range, if so, standardize the data in the weight data array; at the same time, determine whether the difference between a nb and the experience value in the weight data array is within the confidence interval, If yes, store the corresponding weight data in the weight record array, otherwise, use a n that is less than the lower limit of the confidence interval as the new system error value b; the second judgment module 400 judges the weight data in the first judgment module When the number of data n is less than the minimum value of the set range, it is used to determine whether the difference between a nb and the experience value in the weight data array is within the set error range of the experience value, and if so, save the corresponding weight data into the weight record In the array, otherwise, a n when it is less than the empirical value is used as the new system error value b; the output module 500 is used to output the average value of the data in the weight record array within the set time period as the measured weight. In this embodiment, the calculation module 200, the first judgment module 300, the second judgment module 400, and the output module 500 are all set in a single-chip microcomputer.

3, the load cell 100 includes a pin connector P9, a resistor R22, a self-recovery fuse F5, a diode D3, a diode VD2, and a diode VD5. The first pin of the pin connector is grounded and inserted The second pin of the pin connector is connected to one end of resistor R22, one end of resettable fuse F5, and the cathode of diode VD2. The other end of resistor R22 is connected to a +3.3V power supply. The anode of diode VD2 is grounded. The other end of resettable fuse F5 is connected to the anode of diode D3. The cathode of diode VD5 is connected to the sampling port of the microcontroller, the cathode of diode D3 is connected to +3.3V power supply, and the anode of diode VD5 is grounded. The working principle of this circuit is: collect the output signal of the sensor, and transmit it to the computing chip, and can ensure that the sensor signal output does not exceed the design value.

Through the implementation of this embodiment, this embodiment can not only accurately measure the weight of a single animal, but also measure multiple animals at random to obtain the weight of the entire batch, which has practical application value. This embodiment can automatically identify foreign objects on the weighing platform and can clarify its impact through data processing. This embodiment can automatically generate and adjust the empirical value and the confidence interval according to the on-site situation.

Although the embodiments of the present invention are described in conjunction with the accompanying drawings, the patent owner can make various deformations or modifications within the scope of the appended claims, as long as they do not exceed the protection scope described in the claims of the present invention. Within the protection scope of the present invention.

Claims

An intelligent weighing method for livestock and poultry breeding, which is characterized in that it comprises the following steps:

S1. Measure the weight of livestock and poultry to obtain the weight data array a n of livestock and poultry;

S2. Calculate whether a n is less than 50% of the empirical value, and if so, use it as the weighing system error b;

S3. Calculate the difference between a n-b and the experience value in the weight data array;

S4. Determine whether the number of data n in the weight data array is within the set value range, if yes, execute step S5, otherwise, execute step S7;

S5. Standardize the data in the weight data array;

S6. Determine whether the difference between a nb and the experience value in the weight data array in step S3 is within the confidence interval. If so, store the corresponding weight data in the weight record array, otherwise it will be less than a n of the lower limit of the confidence interval As the new system error value b;

S7. Determine whether the difference between a nb and the experience value in the weight data array in step S3 is within the set error range of the experience value, if so, store the corresponding weight data in the weight record array, otherwise it will be less than the experience value A n as the new system error value b;

S8. The average value of the data in the weight record array within the set time period is output as the measured weight.
The intelligent weighing method for livestock and poultry breeding according to claim 1, characterized in that: the specific method for determining the empirical value in step S3 is: determining the standard weight of the livestock and poultry breeding species according to the specific livestock and poultry breeding species The growth curve, the standard weight growth curve is an empirical value.
The intelligent weighing method for livestock and poultry breeding according to claim 2, characterized in that: in the first measurement, if the measured weight data is significantly lower than the weight in the standard weight growth curve at the age of the day, it is judged as basic error data , And use the most recent error value as the system error amount.
The intelligent weighing method for livestock and poultry breeding according to claim 1, characterized in that: the set value in step S4 is 20.
The intelligent weighing method for livestock and poultry breeding according to claim 1, wherein the method for determining the confidence interval in step S6 is:

If the number of data n in the weight data array a n is greater than the set value, the most recently measured weight of multiple livestock and poultry in the weight data array a n is estimated in intervals according to the following formula:

or

In the formula, x is the average value of multiple livestock and poultry weight values, σ is the standard deviation of the weight values, n is the number of weight data used, and α is the significance level in significant statistics, taking 0.1 or 0.05, zα/2 In order to use the value of α/2 to determine the mathematical distribution.
The intelligent weighing method for livestock and poultry breeding according to claim 1, wherein the setting error range of the empirical value in the step S7 is ±20% of the empirical value.
The intelligent weighing method for livestock and poultry breeding according to claim 1, characterized in that: the step S8 specifically includes:

S81. Calculate the average value or mathematical expectation of the data in the weight record array of the last day to obtain the average weight of the livestock and poultry in the last hour;

S82. Summarize the hourly data of the day, perform linear fitting on the twenty-four hour data, and obtain the data at twelve o'clock as the weight value of the day.
A system for realizing the intelligent weighing method for livestock and poultry breeding according to any one of claims 1-7, characterized in that it comprises:

Weighing sensor, used to measure the weight of livestock and poultry, to obtain the data array a n of livestock and poultry weight;

The calculation module is used to calculate the difference between a n-b and the empirical value in the weight data array;

The first judgment module is used to judge whether the number of data n in the weight data array is within the set value range, and if so, standardize the data in the weight data array; at the same time, judge the difference between a nb in the weight data array and the empirical value The value is within the confidence interval. If so, store the corresponding weight data in the weight record array, otherwise, use a n that is less than the lower limit of the confidence interval as the new system error value b;

The second judgment module is used to judge whether the difference between a nb in the weight data array and the experience value is within the setting of the experience value when the first judgment module judges that the number of data n in the weight data array is less than the minimum value of the set range. Within a certain error range, if yes, store the corresponding weight data in the weight record array, otherwise, use a n when it is less than the empirical value as the new system error value b;

The output module is used to output the average value of the data in the weight record array within the set time period as the measured weight.
The system according to claim 8, wherein the weighing sensor is installed under the drinking fountain for livestock and poultry, and a weighing platform is provided on the weighing sensor.
The system according to claim 8, wherein the load cell includes a pin connector P9, a resistor R22, a self-recovery fuse F5, a diode D3, a diode VD2, and a diode VD5. One pin is grounded, the second pin of the pin connector is connected to one end of resistor R22, one end of resettable fuse F5, and the cathode of diode VD2. The other end of resistor R22 is connected to +3.3V power supply. The anode of diode VD2 is connected to ground. Resettable fuse F5 The other end is connected to the anode of diode D3, the cathode of diode VD5, and the sampling port of the single-chip microcomputer. The cathode of diode D3 is connected to a +3.3V power supply, and the anode of diode VD5 is grounded.