Breeding environment temperature control system for pig farm

Abstract

The invention relates to the technical field of data acquisition control, in particular to a breeding environment temperature control system for a pig farm, and the system comprises a temperature acquisition module which collects the temperature data of each monitoring point in the pig farm in real time, and converts the temperature data into a temperature value sequence. According to the method, temperature data of a plurality of monitoring points in a pig farm are collected in real time, a discrete temperature data sequence is converted into basic environment information such as average temperature and temperature difference, and trend and anomaly analysis is carried out on the basis, so that the opportunity and demand of temperature adjustment are determined, the future temperature change trend of the environment is predicted, and a corresponding regulation and control plan is generated; on the basis of prediction, real-time biological feedback information, such as heart rate and body temperature, of pigs is fused, dynamic evaluation is carried out on the comfort level of individual animals, and an environment temperature regulation and control strategy is further optimized by taking the comfort level of the animals as a benchmark, so that the regulation of the environment temperature is more accurately matched with the growth requirements of the pigs, and personalized and intelligent control of the breeding environment is realized.

Description

Technical Field

[0001] The invention relates to the technical field of data acquisition control, and in particular to a breeding environment temperature control system for a pig farm. Background Art

[0002] The field of data acquisition and control technology is a comprehensive field that integrates sensor technology, information processing technology and automatic control technology. It mainly involves real-time monitoring of the on-site environment, equipment status and operating parameters, data acquisition, data processing and analysis, and automatic control based on the analysis results. This field is widely used in many industries such as agricultural breeding, industrial automation, intelligent buildings, and environmental monitoring.

[0003] In the existing technology, there is a lack of active analysis and prediction of data trends and abnormal fluctuations, and the trend of environmental changes cannot be effectively identified, resulting in hysteresis in environmental control. At the same time, the physiological feedback information of individual animals is not collected and analyzed in a timely manner, and the differences between individual animals and changes in their dynamic comfort needs are ignored, making it difficult for the temperature control strategy to truly meet the physiological needs of pigs. Therefore, improvements are needed. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a breeding environment temperature control system for a pig farm.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: A pig farm breeding environment temperature control system comprises:

[0006] The temperature acquisition module collects the temperature data of each monitoring point in the pig farm in real time, converts the temperature data into a temperature value sequence, and generates a temperature original data sequence; based on the temperature original data sequence, the average temperature value and temperature difference are calculated to obtain the basic temperature information of the environment;

[0007] A temperature analysis module receives the basic ambient temperature information, analyzes temperature trends and abnormal points, determines whether the temperature needs to be adjusted through trend analysis, and generates a temperature adjustment signal; predicts future temperature changes based on the temperature adjustment signal and formulates a prediction adjustment plan;

[0008] The biofeedback integration module receives the heart rate and body temperature of the pig, evaluates the comfort of the pig in combination with the prediction adjustment plan, adjusts the temperature setting value, and generates an adjusted temperature setting; based on the adjusted temperature setting, analyzes the adaptability of the pig to the current environment, optimizes the temperature control strategy, and obtains the temperature control strategy;

[0009] The execution control module receives the temperature control strategy, controls the heating or cooling equipment to adjust the temperature according to the control strategy, and generates equipment operation records.

[0010] Preferably, the steps of acquiring the original temperature data sequence are:

[0011] Use multiple temperature sensors located inside the pig farm to monitor the ambient temperature of each monitoring point in real time, and record the temperature data of each monitoring point in real time. The data of each sensor includes a timestamp and a temperature value, and obtain temperature data including a timestamp and a temperature value;

[0012] Based on the temperature data including the timestamp and the temperature value, all the temperature data are sorted by the timestamp and uniformly converted into a temperature value sequence to obtain a temperature original data sequence.

[0013] Preferably, the steps of obtaining the ambient basic temperature information are:

[0014] Based on the original temperature data sequence, the temperature values ​​recorded at each monitoring point are summed up, and the number of temperature records is counted to obtain the total temperature and the number of monitoring points;

[0015] Using the total temperature and the number of monitoring points, the average temperature value of the pig farm environment is calculated by dividing the total temperature by the number of monitoring points to generate an average temperature value;

[0016] Based on the temperature original data sequence, the highest temperature value and the lowest temperature value in the sequence are identified, the difference between the highest temperature value and the lowest temperature value is calculated to obtain the temperature difference, and the ambient basic temperature information is obtained in combination with the average temperature value.

[0017] Preferably, the step of obtaining the temperature adjustment signal is:

[0018] Parsing the temperature data sequence in the basic ambient temperature information, and dividing the time series data into fixed time windows to obtain a temperature time series data set;

[0019] Based on the temperature time series data set, the temperature trend change value is calculated, and the calculation formula is:

[0020]

[0021] Among them, M k is the temperature trend change value at the kth time point, P i is the temperature value at the i-th time point, value, P i-1 is the temperature value at the previous time point, t i is the timestamp of the i-th time point, t i-1 is the timestamp of the previous time point, m is the total number of observation points in the time window, b represents the impact scale of the data in the time window on the current calculation point, and t k is the timestamp of the kth time point;

[0022] Based on the temperature trend change value, an abnormal point in the temperature time series is identified, and combined with the positive and negative directions of the temperature trend change value, it is determined whether the current temperature change needs to be adjusted, and a temperature adjustment signal is generated.

[0023] Preferably, the steps of obtaining the forecast adjustment scheme are:

[0024] Based on the temperature adjustment signal, the future temperature change is calculated using the following formula:

[0025]

[0026] Where, ΔT future Represents the predicted future temperature change, T last is the temperature value, T avg is the average temperature value of the time series column, represents the temperature change rate of the time series, T var is the temperature variability, β adjusts the effect of the difference between the warming temperature and the mean temperature, γ adjusts the effect of the square root of the rate of temperature change, and δ adjusts the effect of the logarithm of the temperature variability;

[0027] Based on the future temperature change, a temperature adjustment strategy is formulated and a predicted adjustment plan is generated.

[0028] Preferably, the steps of obtaining the adjusted temperature setting are:

[0029] Obtain the heart rate and body temperature data of the pigs, and combine them with the prediction and adjustment plan to obtain biofeedback data;

[0030] Based on the biofeedback data, the comfort score is calculated using the following formula:

[0031]

[0032] Where C represents the comfort score, T is the current body temperature, and T opt is the standard body temperature, T range is the normal fluctuation range of body temperature, H is the current heart rate, and H opt is the standard heart rate, H range It is the normal fluctuation range of heart rate;

[0033] Based on the comfort score, the temperature setting is adjusted to generate an adjusted temperature setting.

[0034] Preferably, the steps of obtaining the temperature control strategy are:

[0035] Based on the adjusted temperature setting, and extracting the pig's current heart rate and body temperature data, comprehensive biofeedback information is obtained;

[0036] Based on the comprehensive biofeedback information, the environmental adaptability index is calculated using the following formula:

[0037]

[0038] Among them, E represents the environmental adaptability index, T actual is the pig's body temperature, T set is the adjusted temperature setting value, H actual is the pig's heart rate, H target is the target heart rate value;

[0039] Based on the environmental adaptability index, the temperature control strategy is analyzed and optimized.

[0040] Preferably, the steps of obtaining the device operation record are:

[0041] According to the temperature control strategy, operate the heating and cooling equipment and perform temperature adjustment activities, including setting new temperature parameters and starting or stopping equipment operation;

[0042] Record each equipment operation data, including operation time, equipment response and temperature changes before and after, and generate equipment operation records.

[0043] Compared with the prior art, the advantages and positive effects of the present invention are:

[0044] In the present invention, by real-time collection of temperature data from multiple monitoring points inside a pig farm, discrete temperature data sequences are converted into basic environmental information such as average temperature and temperature difference, and trend and anomaly analysis is carried out on this basis to clarify the timing and needs of temperature adjustment, predict the future temperature change trend of the environment, and generate corresponding control plans; on the basis of the prediction, the real-time biofeedback information of the pigs, such as heart rate and body temperature, is integrated to dynamically evaluate the comfort of individual animals, and the environmental temperature control strategy is further optimized based on the comfort of the animals, so that the adjustment of the environmental temperature can more accurately match the growth needs of the pigs, and the personalized and intelligent control of the breeding environment can be achieved; in the actual control process, the heating or cooling equipment automatically operates according to the optimized temperature control strategy, and generates corresponding operation records to ensure the traceability of the temperature control effect; the environmental temperature control is realized in an integrated, intelligent, personalized and dynamic manner, the suitability of the pig growth environment is improved, the animal stress response is reduced, the animal health status is improved, and the breeding efficiency is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0047] See also Figure 1 The present invention provides a technical solution: a pig farm breeding environment temperature control system comprising:

[0048] The temperature acquisition module collects the temperature data of each monitoring point in the pig farm in real time, converts the temperature data into a temperature value sequence, and generates a temperature raw data sequence; based on the temperature raw data sequence, the average temperature value and temperature difference are calculated to obtain the basic temperature information of the environment;

[0049] The temperature analysis module receives basic ambient temperature information, analyzes temperature trends and abnormal points, determines whether the temperature needs to be adjusted through trend analysis, and generates a temperature adjustment signal; based on the temperature adjustment signal, predicts future temperature changes and formulates a prediction adjustment plan;

[0050] The biofeedback integration module receives the heart rate and body temperature of the pigs, combines the prediction adjustment plan, evaluates the comfort of the pigs, adjusts the temperature setting value, and generates the adjusted temperature setting; based on the adjusted temperature setting, analyzes the adaptability of the pigs to the current environment, optimizes the temperature control strategy, and obtains the temperature control strategy;

[0051] The execution control module receives the temperature control strategy, controls the heating or cooling equipment to adjust the temperature according to the control strategy, and generates equipment operation records.

[0052] The steps to obtain the original temperature data sequence are:

[0053] Use multiple temperature sensors located inside the pig farm to monitor the ambient temperature of each monitoring point in real time, and record the temperature data of each monitoring point in real time. The data of each sensor includes a timestamp and a temperature value, and obtain temperature data including a timestamp and a temperature value;

[0054] Based on the temperature data containing timestamps and temperature values, all temperature data are sorted by timestamps and uniformly converted into temperature value sequences to obtain the original temperature data sequence.

[0055] Specifically, multiple temperature sensors located inside the pig farm are used, combined with the temperature range determined in advance through historical climate data and expert experience. For example, the monitoring range is set between -10°C and 50°C. If the collected value exceeds this range within a certain period of time, it is regarded as abnormal data, and the corresponding temperature value and timestamp are recorded at each collection. The threshold setting process can be exemplified by first collecting statistics on the highest and lowest temperature data in different seasons in the past three years, and then adding and subtracting 5°C of safety boundaries to the actual distribution range of the highest and lowest temperatures, so as to obtain the effective detection upper and lower limits of the range. When collecting, the sensors are first numbered and their locations are marked. , then obtain the temperature readings at fixed time intervals and associate them with the corresponding timestamps. Each record is checked for a range, for example, the temperature value is compared between -10℃ and 50℃. If the value falls within this interval, it is marked as normal. If the value exceeds the range, it is marked as abnormal and the readings at adjacent time points are compared again to troubleshoot sensor failures. After the comparison is completed, all records are integrated into multiple numbered temperature information data rows, which contain sensor locations, temperature values, and timestamps, and are accumulated and stored in the data set after each update. In this way, data can be continuously collected and verified, and temperature data containing timestamps and temperature values ​​are finally obtained.

[0056] Based on the temperature data containing timestamps and temperature values, first disassemble and extract the corresponding timestamp fields and temperature value fields from the multiple numbered temperature information data rows obtained in the previous step, and sort them in ascending order according to the timestamps. When sorting, if it is found that some records corresponding to the timestamps are repeated or the intervals are too short, such as multiple records in the same second, the most accurate and unmarked abnormal data can be selected as the final retention value. If some non-compliant records need to be eliminated, refer to the range check results in the previous step. At this time, according to the set empirical threshold T range (For example, -10℃ to 50℃) to reconfirm the abnormal data row. If the recorded temperature value or timestamp itself is unreasonable, it will be marked separately and removed from the ordered sequence. After completing the effective data screening, each temperature reading is arranged in chronological order. Each temperature value is paired with the corresponding timestamp, and a corresponding index is established internally to facilitate subsequent retrieval and comparison. In this process, a minimum time difference threshold can be set to ensure the time sequence integrity of the data. For example, it is set that the interval between two records must be at least 1 second. If the interval between a record is too short, only the record with a more accurate timestamp is retained. Finally, all processed entries are uniformly converted into a temperature value sequence in timestamp order to obtain the original temperature data sequence.

[0057] The steps to obtain the basic ambient temperature information are as follows:

[0058] Based on the original temperature data sequence, the temperature values ​​recorded at each monitoring point are summed up, and the number of temperature records is counted to obtain the total temperature and the number of monitoring points;

[0059] Using the total temperature and the number of monitoring points, the average temperature value of the pig farm environment is calculated by dividing the total temperature by the number of monitoring points to generate an average temperature value;

[0060] Based on the original temperature data sequence, the highest temperature value and the lowest temperature value in the sequence are identified, the difference between the highest temperature value and the lowest temperature value is calculated to obtain the temperature difference, and the basic environmental temperature information is obtained by combining the average temperature value.

[0061] Specifically, based on the previously acquired temperature raw data sequence, when summarizing the temperature values ​​of each monitoring point, you can first assign unique numbers to different monitoring points and record their respective temperature readings, then add up the temperature values ​​collected by all monitoring points at the same time. In order to distinguish possible extreme abnormal readings, you can refer to the highest and lowest temperatures obtained in the same environment in previous years and add or subtract 5°C as an empirical threshold. For example, compare the temperature of the monitoring point with -10°C to 45°C. If it exceeds this range, it is marked as abnormal and continues to compare the records at adjacent times to determine whether it is fault data. If it is confirmed to be a fault, the point is removed from the current summation process, and then the temperature values ​​within the normal range are accumulated item by item and the total number of records is calculated. Assuming that the number of monitoring points is n, the temperature reading of each point is T i , then all T i The sum is recorded as The number of valid monitoring points is determined as the actual statistical value of n. For example, if there are 20 monitoring points, the validity of their temperature readings is checked one by one and the valid readings are superimposed one by one. If some of the points exceed the specified threshold multiple times, these readings are temporarily excluded. Finally, the number of monitoring points that meet the requirements is counted and output together with the superposition results to obtain the total temperature and number of monitoring points.

[0062] Using the total temperature and the number of monitoring points obtained in the previous steps, the average temperature can be obtained by directly performing division in subsequent processing. In order to make the calculation process more consistent, a basic comparison range can be set and the same monitoring point statistical strategy can be used each time. For example, the monitoring point temperature is compared with the interval between -10℃ and 50℃. If all readings are within this interval, they are retained for calculation. If there is a small amount of fault data that exceeds this interval and has been confirmed many times, they are marked separately and then excluded. Then let S temp represents the total temperature of the superposition, and m represents the number of actual effective monitoring points. Complete the division calculation. If S is set in the example temp=480, m=20, then 24℃ can be obtained as the average value of the current batch record. When completing this process, it is necessary to confirm the validity of all monitoring points once and summarize the average temperature in the result, so as to calculate the average temperature value of the pig farm environment and generate the average temperature value.

[0063] Based on the original temperature data sequence obtained previously, the highest and lowest temperature values ​​are identified by searching the temperature readings under all time indexes one by one. If the temperature value deviates greatly from the previous observation range at certain moments, it can be compared with the set upper and lower limits. For example, the pre-defined range is first defined as -10℃ to 50℃. If a reading falls below -15℃, additional confirmation is performed. If the confirmation is correct, the recording state is adjusted to prevent confusion. Then, the maximum and minimum values ​​are selected from all valid data and their difference is calculated, which is recorded as the temperature difference. If the maximum value is 40℃ and the minimum value is 10℃ in the actual example, the temperature difference is 30℃. If there are many monitoring points, an index table can be established first to quickly find the sensor number or time period corresponding to the maximum and minimum readings. For individual sensors that may repeatedly experience abnormal conditions, they are placed in an independent list for separate observation. In the whole process, no other external calculation models are introduced and the final result is confirmed in an intuitive calculation method to obtain the temperature difference, and the basic temperature information of the environment is obtained by combining the average temperature value.

[0064] The steps for obtaining the temperature adjustment signal are:

[0065] Parse the temperature data sequence in the basic ambient temperature information, and divide the time series data into fixed time windows to obtain the temperature time series data set;

[0066] Based on the temperature time series data set, the temperature trend change value is calculated using the following formula:

[0067]

[0068] Among them, M k is the temperature trend change value at the kth time point, P i is the temperature value at the i-th time point, value, P i-1 is the temperature value at the previous time point, t i is the timestamp of the i-th time point, t i-1 is the timestamp of the previous time point, m is the total number of observation points in the time window, b represents the impact scale of the data in the time window on the current calculation point, and t k is the timestamp of the kth time point;

[0069] Based on the temperature trend change value, identify the abnormal points in the temperature time series, and combine the positive and negative directions of the temperature trend change value to determine whether the current temperature change needs to be adjusted, and generate a temperature adjustment signal.

[0070] Specifically, after parsing the temperature data sequence in the basic environmental temperature information, it is necessary to compare the timestamp fields of these data one by one, and divide them continuously according to the determined time windows. When dividing the time windows, first refer to the temperature update frequency of each monitoring point obtained before, and determine the size of the fixed window in combination with the actual sampling frequency. For example, the window duration is set to ten minutes and divided into several segments throughout the day. Each segment will correspond to a batch of temperature data at the start and end times. To ensure the accuracy of the window division, you can start from a base time point and accumulate ten-minute intervals segment by segment. The temperature data collected in each time period is recorded and classified. If it is found that the temperature value of any monitoring point is abnormal in the current window, such as exceeding the preset range of -10°C to 50°C, it is necessary to further verify whether there is equipment failure or environmental interference at the monitoring point. If it is confirmed multiple times that the monitoring point continues to have readings that deviate significantly from the range, you can temporarily stop reading. Stop including it in this window statistics and continue the window division process. In this way, the temperature data groups within a fixed time length are obtained one by one and the corresponding start and end times are marked. In order to distinguish the temperature changes at different times, a number can be added to each time window and the timestamp and monitoring point number corresponding to each temperature record can be stored. If some records are repeated or missing at the window boundary, they can be classified into the window closest to the collection time according to the order of actual records. After completing the division of all time segments, a preliminary integrity verification is performed on each segment. The verification steps include verifying whether the number of data is consistent with the number of monitoring points and whether there are obviously unreasonable extreme readings in the numerical distribution. After these operations, the temperature data corresponding to all windows can be summarized and a structured time period index list can be formed to facilitate the subsequent analysis of temperature changes in different time intervals to obtain a temperature time series data set.

[0071] The benefit of the formula is that it performs a weighted summation of the changes in adjacent temperature readings with time differences in the numerator, and uses an exponential decay function in the denominator to measure the contribution of data at different times to the target time point, thereby taking into account both numerical differences and time distances when analyzing temperature trends.

[0072] P i The steps of obtaining the parameter are as follows: the parameter is the temperature value at the i-th time point, which needs to be continuously monitored by the previously established temperature acquisition device. During monitoring, the temperature of each monitoring point is periodically recorded at a uniformly set sampling interval, and each data is accompanied by a timestamp of the moment. Then, the temperature value is separated and stored in the database according to the monitoring point number and time series index. If P is to be obtained, i, the corresponding value is read from the database according to the i-th time index. In order to further quantify the accuracy of the temperature value, an observation range of -10℃ to 50℃ is set in combination with actual environmental factors. If the temperature reading exceeds this range for many times, it is considered that the sensor may be faulty. The reading of this monitoring point needs to be excluded or listed separately in the formal calculation. For legal data records, they can be directly used as P i Assuming that there is a window containing 5 valid observation points, the corresponding temperatures are 21℃, 22℃, 24℃, 23℃, and 25℃, then the i-th reading can be recorded as P i =21 or 22, etc.

[0073] P i-1 The steps for obtaining the parameter are as follows: the parameter represents the temperature value after one time unit is traced back from the i-th time point, and is used to calculate the temperature difference of adjacent moments. At the index level, it can be directly obtained by using i-1 as the subscript of the time series. If i=1, it means that this is the reading at the initial moment in the window. At this time, the record of i-1 cannot be found in the same window, and special processing is required during statistics. After arranging the temperature data in time series, each record can quickly locate the temperature value of the previous record. Missing data or abnormal records also need to be excluded before statistics, corresponding to the logical AND P i Similarly, when the temperature value at the previous moment can be obtained normally, the value is used as P i-1 , for example in the above example P 2 At 22℃, P 1 is 21°C, so P 2 -P 1 =1 can be used in the formula.

[0074] t i The steps for obtaining the parameter are as follows: This parameter is the timestamp of the i-th time point, which is usually automatically recorded by the real-time clock or system time and bound to the temperature reading when the temperature is collected. Each monitoring point has its own set of timestamp records, which can be managed uniformly according to the previously established index method. If the system sets a sampling interval of 10 seconds, a new t can be obtained every 10 seconds. i In actual applications, timestamps are converted to whole seconds or finer milliseconds for comparison. If some records are missed or extremely delayed during the entire window process, it is impossible to form a continuous t i Sequences need to be eliminated in the later statistical analysis. For normal time series records, their time values ​​can be directly extracted as t i For example, if the Dangdang window starts recording at 08:00:00 on the current day and collects temperature every 10 seconds, the i-th timestamp can be 08:00:10i and increase in steps.

[0075] t i-1 The steps to obtain the parameters are: the parameters and P i-1 Similarly, it is obtained by locating the i-1i-1th moment in the timestamp sequence, which is used to measure the time difference between adjacent observation points. If a record is invalid or missing due to a fault, it will not be able to correctly correspond to the timestamp i-1 during statistics. It is necessary to filter all data for validity in advance and only retain the records that can be fully aligned for calculation. The acquisition process can also be in the overall time series list. The time value is directly read according to the index i-1. If the window contains 5 valid observation data, corresponding to the timestamps of 08:00:00, 08:00:10, 08:00:20, 08:00:30, and 08:00:40, then when i=3, t 3 For 08:00:20 2 It is 08:00:10.

[0076] The steps to obtain the m parameter are as follows: This parameter refers to the total number of observation points in the time window, which needs to be obtained after the legitimacy of the data of all monitoring points in the window is determined, that is, the number of truly available time series records is counted to determine the value of m. If the monitoring points frequently have abnormalities, the number of valid records will be reduced. Therefore, when calculating M k The m actually used may be less than the theoretical number of records. Before obtaining m, all data can be arranged in chronological order and illegal records can be removed. The remaining is a list of valid observation points. After counting them, m can be determined. For example, under 10-second sampling intervals, a five-minute window should contain 30 records. If 2 of them are deleted due to failures, the final m is 28. In the formula, it is summed and normalized according to 28.

[0077] The steps to obtain the b parameter are as follows: this parameter represents the impact scale of the data in the time window on the current calculation point. It is used as a benchmark value in the exponential decay factor. It is necessary to first collect time and temperature data for a longer period, and determine the specific value by analyzing the impact intensity of different time differences on the trend calculation. If the value is small, all records can be divided into several intervals according to the time difference distribution, such as each interval covering 5 seconds, and then the impact of temperature changes in each interval on subsequent moments is counted. After forming multiple impact measurement values, the average is calculated and fine-tuned in combination with the change characteristics in the actual environment. Finally, this average value is regarded as the benchmark value of b. For example, in a four-hour monitoring cycle, the time difference for each collection is 10 seconds, and a total of 1440 records are collected. The correlation between the temperature fluctuations between adjacent records and the subsequent moments is calculated in turn, and then all correlations are averaged to determine b. If the measured average value is about 2.0, then b=2.0.

[0078] t kThe steps to obtain the parameters are as follows: the parameter is the timestamp of the kth time point, and the position is the same as t i The concept is similar, but in the formula it represents the current target moment under consideration, and it needs to be compared with all t in the window i Calculate the time difference and decay its effect exponentially to get t k When traversing the window, we can first select a record as the current target point, then traverse other records in the window to calculate the weighted value, and then repeat the similar process on the next record, so t k In the program, it will change with the difference of k, and the corresponding temperature value will also select the observation value at that moment synchronously. If there are m records in the current window, there will be multiple t records with k ranging from 1 to m. k For example, in the five records from 08:00:00 to 08:00:40, 08:00:00, 08:00:10, 08:00:20, 08:00:30, and 08:00:40 are used as t k .

[0079] Calculation process:

[0080] In an example scenario, set the window m=5 and the temperature sequence is P 1 =21, P 2 =22, P 3 =24, P 3 =24, P 4 =23, P 5 =25, the corresponding timestamp is t 1 =0, t 2 =10, t 3 =20, t 4 =30, t 5 = 40 seconds, and b = 2.0, when k = 3 is selected to calculate M 3 , taking 08:00:20 as the current target time; substituting it into the formula, the result is 0.597, which further characterizes the temperature trend change amplitude at the third time point.

[0081] The results show that when M k When M is larger, the temperature fluctuation near the considered time point is more obvious. k If it is smaller, it means that the temperature changes slowly. If a series of M k The larger value can be used to determine the period of short-term drastic temperature fluctuations, thus providing a basis for subsequent abnormal point identification and adjustment.

[0082] Based on the temperature trend change value calculated in the previous step, the entire temperature time series can be traversed. In the process, a preliminary temperature trend change value needs to be assigned to each time point and its positive and negative directions need to be recorded. Then, a comparison is made within a set reference range. For example, when most M k If the values ​​are all between 0 and 2, and a value greater than 5 or a negative drop to less than -1 suddenly appears at some moments, it means that the temperature difference in this period is quite significant. At this time, these values ​​can be marked as potential abnormal points. In order to determine whether these abnormal points are indeed related to abnormal environmental conditions, they need to be compared with the previously recorded average temperature level or equipment operation conditions. When comparing, a special threshold range can be set. For example, M k If the value is greater than 4.5, it is considered a strong fluctuation. Based on this empirical threshold, we can quickly screen out the time periods that are worth focusing on, and further check whether there are sensor failures or abnormal operation records of the heating and cooling devices in the corresponding time periods. If it is confirmed that there are no failures, it means that sudden temperature fluctuations did occur during this period. This situation needs to be promptly incorporated into the subsequent control logic of the system. At the same time, combined with the positive and negative directions of the temperature trend change value, if it is found that the current M k A large positive rise indicates that the temperature is rising rapidly, and a significant negative value indicates that the temperature is dropping rapidly. Then the next adjustment process can be initiated, such as issuing a cooling command when the positive fluctuation is too high or triggering a heating plan when the negative drop occurs. Finally, after a series of judgments, it is concluded whether the ambient temperature needs to be adjusted at this moment, and a temperature adjustment signal is generated.

[0083] The steps to obtain the forecast adjustment plan are:

[0084] Based on the temperature adjustment signal, the future temperature change is calculated using the following formula:

[0085]

[0086] Where, ΔT future Represents the predicted future temperature change, T last is the temperature value, T avg is the average temperature value of the time series column, represents the temperature change rate of the time series, T var is the temperature variability, β adjusts the effect of the difference between the warming temperature and the mean temperature, γ adjusts the effect of the square root of the rate of temperature change, and δ adjusts the effect of the logarithm of the temperature variability;

[0087] Based on future temperature changes, a temperature adjustment strategy is formulated and a predicted adjustment plan is generated.

[0088] Specifically, the benefit of the formula is that by weighted addition of the difference between the temperature and the average temperature, the square root of the temperature change rate, and the logarithm of the temperature variability, the current temperature deviation, the rate of change, and the overall degree of fluctuation can be comprehensively considered in the same expression, thus avoiding the deviation caused by relying on a single indicator.

[0089] T last The steps for obtaining the parameter are as follows: This parameter represents the temperature value of the most recent valid observation, which is derived from the temperature collection data of the actual monitoring point. The monitoring system will send query instructions to each sensor in the pig farm in each sampling period and record the temperature value of each monitoring point. In order to ensure that the most recent temperature value can be accurately extracted, all monitoring points will be sorted by timestamp and the legal temperature record at the latest moment will be found. If some sensors have multiple abnormal temperature values, for example, exceeding the empirical range of -10℃ to 50℃, the data of the monitoring point will be excluded after judging its fault, and the latest temperature readings of the remaining sensors will select the value corresponding to the maximum timestamp as T according to the time sequence. last For example, in a certain collection cycle, after sorting, it is found that there are data with the latest timestamp marked as 10:23:06, and the temperature reading is 26.5℃, then 26.5℃ is determined as T last .

[0090] T avg The steps to obtain the parameter are as follows: This parameter represents the average temperature value of a time series. It is necessary to count the temperature readings of all sensors within the continuous monitoring period, sum up the data that meets the valid range and divide it by the corresponding number of records to obtain it. When using it, first merge the temperature data of a period of time (for example, the last 30 minutes or 1 hour), then exclude the confirmed fault data and add up the remaining data, and then divide it by the number of valid records to obtain the average temperature value of the global or local time period. To enhance the accuracy, 5 minutes or 10 minutes can be used as the minimum statistical unit. After each unit obtains a local average value, it is summarized step by step. For example, in a certain area, a total of 180 legal temperature values ​​are collected in the past 30 minutes, and the total sum is 4100℃. 4100℃, then

[0091] The steps to obtain the parameter are as follows: This parameter represents the temperature change rate of the time series. It is necessary to perform differential calculations on the temperature data of the same sensor or multiple sensors at different times. First, select a reference sensor that can represent the overall trend, or use the average change rate of multiple sensors as a reference, and record a series of temperature values ​​T at adjacent times. k With T k-1 , timestamp t k With t k-1 ,by The form of point-by-point calculation and taking the average or median over a period of time to obtain For example, at a sampling interval of 10 seconds, the temperature at the last moment was 25.2°C, and the current temperature is 25.8°C. The temperature change rate is

[0092] T var The steps to obtain the parameter are as follows: This parameter represents the variability of temperature data within a certain time window. It is necessary to collect temperature records within a period of time and then calculate the variance or standard deviation, and then use the variance or standard deviation to measure the overall fluctuation degree. For example, a series of temperature values ​​are collected within 30 consecutive minutes.

[0093] {T 1 ,T 2 ,...,T n}, μ represents the mean of these temperature values, then the variance of the temperature can be If the variance result is relatively large, it means that the temperature fluctuates greatly. If the value is small, it means that the temperature is stable. The variance or the standard deviation obtained by taking the square root of the variance is used as T var For example, when n = 50 records, the mean μ = 23 °C, and the variance is about 3.61 after calculating the cumulative term. Therefore, 3.61 can be regarded as the temperature variability T of the current window. var .

[0094] The steps for obtaining the β parameter are as follows: this parameter is used to adjust the weight of the impact of the difference between temperature and the average temperature. It is necessary to determine a reasonable value range in combination with long-term observation data. For example, when the temperature difference between summer and winter is obvious, β can be moderately increased to highlight the impact of temperature deviation from the average value. When obtaining β, you can first make statistics on the difference between the measured temperature and the average temperature of the same period from the temperature monitoring in recent weeks, and then observe the impact of the difference distribution on the system control results. The initial value range of β is formulated based on the subsequent energy consumption caused by the difference change, and then the final β is selected through localized debugging. For example, after comparative analysis of 1,000 sets of data, it is concluded that when β is close to 0.3, the system's response to temperature deviation is more appropriate.

[0095] The steps to obtain the γ parameter are as follows: This parameter is used to adjust the proportion of the square root of the temperature change rate in the overall prediction. A series of reference values ​​are collected, and the actual temperature fluctuation results of the corresponding time period are used to verify the perception of rapid temperature rise and fall. In the case of frequent heating or cooling, γ can be appropriately increased. If the average temperature change rate of the monitoring point every 10 seconds is 0.05℃ / second, and in some local time periods it can reach up to 0.15℃ / second, then it can be compared based on multiple time periods, and the impact of the high change rate can be measured one by one to determine γ. For example, after one week of monitoring, it is found that γ is more appropriately set at 0.4.

[0096] The steps to obtain the δ parameter are as follows: This parameter is used to adjust the position of the logarithmic term of temperature variability in the overall prediction, and needs to be calculated based on the T var As a result, the temperature fluctuations in different aquaculture areas were quantitatively evaluated. When the variability between regions was significantly different, it was necessary to appropriately adjust δ to reflect the weight in the prediction. To this end, multiple groups of T var The actual control situation performed by the system is recorded, and then the specific value of δ is selected by comparing the difference in the results. For example, after 10 days of continuous collection, it is found that a higher δ is more sensitive to local fluctuations and a lower δ has a greater tolerance. Finally, δ is established as 0.2.

[0097] Calculation process:

[0098] Now let's take a specific numerical calculation example and let T last =26.5℃(obtained from the most recent collection),

[0099] T avg =22.8℃ (averaged from 30 minutes of data), (calculated according to the previous temperature difference), T var =3.61 (estimated from the 30-minute temperature variance), β = 0.3, γ = 0.4, δ = 0.2 (determined by the above analysis of long-term observation data). Substitute the above parameters into the formula:

[0100]

[0101] First calculate each addition:

[0102] (26.5-22.8)=3.7

[0103] 0.3 3.7 = 1.11

[0104]

[0105] 0.4 · 0.2449 ≈ 0.09796

[0106] ln(3.61)≈1.281

[0107] 0.2 1.281 ≈ 0.2562

[0108] Then add them together to get:

[0109] ΔT future =1.11+0.09796+0.2562≈1.46416

[0110] Combining the above formula, we can get ΔT future ≈1.46416, indicating that the subsequent temperature may rise by about 1.46℃.

[0111] This result shows that the subsequent temperature still has a certain upward trend compared with the overall average level in the current measurement stage. future When it further increases to 2 or 3 or above, it may cause a more dramatic temperature rise. When the value is small or even negative, it means that the temperature may tend to stabilize or decrease. This calculation result can provide a numerical basis for the next temperature adjustment strategy, and then help formulate subsequent environmental temperature control execution actions.

[0112] When formulating a temperature adjustment strategy based on future temperature changes, we first need to calculate the ΔT future To clarify how different temperature changes match corresponding operations, multiple threshold intervals are set within the recording range. For example, ΔT future 0℃ to 1℃ is considered a small range, 1℃ to 2℃ is considered a medium range, and 2℃ to 5℃ is considered a large range. Reasonable boundaries are added to the area. For example, the difference in summer temperature in the breeding area in the past three years usually does not exceed 5℃. If the monitored ΔT future If it exceeds 5, it is considered as extreme temperature rise. Combined with the actual power data of the on-site heating and cooling equipment, the corresponding execution time and equipment output are specified for each interval, thereby generating a step-by-step adjustment strategy. future Only at 1°C to 2°C can a short observation be maintained and the cooling equipment power be slightly reduced or the ventilation be moderately increased. When the value is larger, the mandatory temperature control process is directly started, for example, at ΔT future = 3, turn on the maximum air volume cooling device and continue to observe for five minutes and then re-evaluate. If the ΔT calculated again after five minutes is future If it still exceeds 2, the current cooling power is maintained. If it falls back below 1, it will enter the normal maintenance mode. In order to prevent the device from frequently flickering between high power and low power, the execution records of the current moment and the previous moment will be compared. If two ΔT future If the difference is not big, maintain the current adjustment intensity. If the difference is large, switch to the corresponding power gear immediately. Finally, the monitoring link collects the actual temperature changes and the operating conditions of the heating or cooling equipment and summarizes them, so as to use the determined operation output as the predicted adjustment plan.

[0113] The steps to obtain the adjusted temperature setting are:

[0114] Obtain the pig's heart rate and body temperature data, and combine it with the predicted adjustment plan to obtain biofeedback data;

[0115] Based on the biofeedback data, the comfort score is calculated using the following formula:

[0116]

[0117] Where C represents the comfort score, T is the current body temperature, and T opt is the standard body temperature, T range is the normal fluctuation range of body temperature, H is the current heart rate, and H opt is the standard heart rate, H range This is the normal fluctuation range of heart rate;

[0118] Based on the comfort score, the temperature setting is adjusted to generate an adjusted temperature setting.

[0119] Specifically, when obtaining the heart rate and body temperature data of the pigs and combining them with the prediction and adjustment plan, it is necessary to first arrange a monitoring device that can measure the heart rate and body temperature in the pig farm. The monitoring device detects the heart rate fluctuation through the sensor close to the pig's body and records the readings at fixed time intervals. If some samples have too high or too low values, it is necessary to compare them with the normal heart rate range. For example, the normal heart rate range is set between 60 beats / minute and 120 beats / minute. If the collected reading is higher than 120 beats / minute, it is marked as a high value and the reading of the next period is recorded again to confirm whether there is a continuous abnormality. If it deviates significantly from 60 beats / minute, it is marked as a low value and continued to be concerned. The body temperature data is usually obtained by placing the thermometer close to the pig's torso or armpit. The corresponding timestamp is stored each time it is recorded and compared with the heart rate data. In order to make effective use of the real-time data, it is necessary to combine the previously obtained prediction and adjustment plan and record each heart rate data. A matching index is added to the body temperature data. The matching index will associate the predicted adjustment plan at that moment with the actually measured physiological indicators. If the heart rate and body temperature are within the specified range and there is no drastic fluctuation, the state is judged to be relatively stable. If the heart rate or body temperature exceeds the preset normal range for multiple times in a row, the historical monitoring records can be compared to determine whether there is equipment failure or physiological abnormalities of pigs. If it is confirmed that there is no failure, the state will be marked as an abnormal situation in the data, and the subsequent heart rate and body temperature changes will continue to be recorded. Then all samples will be matched one-to-one with the predicted adjustment plan in chronological order. For example, thousands of heart rate and body temperature readings are accumulated in the 24-hour monitoring link every day to form a large-scale time series data sequence. Each record contains the physiological indicators and the adjustment plan entries corresponding to that moment. The combination of these records is the biofeedback data, which can provide input for the next calculation after the data storage and identification are completed.

[0120] The benefit of the formula is that it reflects the degree of deviation of body temperature and heart rate in a segmented quantitative manner, and uses a linear subtraction mechanism to make larger deviations account for a more obvious proportion in the calculation results, which can comprehensively measure the stability of the pig's current body temperature and heart rate in a single score.

[0121] The steps for obtaining the T parameter are as follows: this parameter represents the current body temperature value, which needs to be regularly detected and recorded by a temperature measurement sensor in the actual breeding environment. During the data acquisition process, the sensor is first calibrated for zero point and range, and the frequency of collection is set to be once every five minutes. Then, the latest, most timely and not marked as faulty data is selected from all monitoring records as the current body temperature. The selection of sensors can refer to common infrared thermometers or surface contact temperature measurement devices. If a body temperature is detected to be outside the normal range during the measurement, such as 33°C to 42°C, it will be additionally marked as a suspicious reading when recording and subsequent inspection will be performed. If the possibility of fault is ruled out, this value will be normally included in T. For example, if 37.8°C is detected at 10:15 on a certain day and it is confirmed that there is no abnormal fault, it will be recorded as T at the current moment.

[0122] T opt The steps to obtain the parameter are as follows: This parameter represents the standard body temperature value, which is used to measure the difference between the actual body temperature and the ideal level. It needs to be obtained by intensively analyzing the body temperature of a large number of pigs of the same breed and the same growth stage. The specific method can be to conduct multiple body temperature tests on pigs in the target breeding area every day within one month. After excluding sick individuals or individuals in special stress states, the body temperature values ​​of the remaining samples are averaged or the most common segment is taken, and then a stable mean value is obtained as T opt In most breeds of pigs, a temperature of around 39°C is common. This can be further corrected by combining genetic and environmental factors. For example, after collecting 1,000 temperature records of healthy pigs, the average value is calculated to be 39.0°C, and 39.0°C can be used as T opt Example value of .

[0123] T range The steps to obtain the parameter are as follows: This parameter represents the normal fluctuation range of body temperature and is used to regulate the degree of deviation of the current body temperature from the standard body temperature. It is necessary to collect the upper and lower limits of the body temperature of a large number of pigs under non-abnormal conditions over a period of time, and then calculate the difference between the maximum and minimum values, or use the standard deviation method to weight and determine the upper and lower thresholds, so as to form a T range If the fluctuation is determined to be between 38.5℃ and 39.5℃, the range is about 1.0℃. In order to get a more accurate range, we can count the temperature records of about 10,000 pigs in the same batch, take the 95% or 99% confidence interval as the fluctuation limit, and then use the width of the limit as T. range The final value of the temperature can be substituted into the formula assuming that the final value is 1.0°C.

[0124] The steps to obtain the H parameter are as follows: this parameter represents the current heart rate value, which needs to be achieved by monitoring the number of heartbeats. The conventional practice is to install or briefly attach an electrocardiogram sensor to the pig's trunk, collect signals once a second and average them over a period of time, and obtain more stable heart rate results after eliminating occasional noise points. If multiple measurements within a day give roughly similar results, the last legal value can be extracted as the current heart rate. If some records exceed 160 beats / minute or are lower than 50 beats / minute and the equipment has been confirmed to be fault-free, it means that the pig does have extreme fluctuations. It is necessary to retain this value in the record, mark it as an extreme value, and make additional observations. For example, if the heart rate is measured at 98 beats / minute in the same time period at 10:15, it is confirmed that this is the current H value.

[0125] H opt The steps to obtain the parameter are as follows: This parameter represents the standard heart rate value. It is necessary to collect the heart rates of healthy pigs on a large scale in a breeding environment and perform statistical analysis to obtain an average or median value that best reflects the normal state. If most of the records in several weeks of monitoring are concentrated in the range of 70 to 110 beats per minute, the center of the interval or the distribution segment with the highest frequency can be regarded as the standard heart rate. Further fine-tuning can be performed based on the physical fitness and daily activities of the pigs on site, and the determined value can be used as H opt For example, based on thousands of valid heart rate readings, the average heart rate is about 85 beats / minute. In this case, 85 beats / minute can be designated as H opt .

[0126] H range The steps to obtain the parameter are as follows: This parameter represents the normal fluctuation range of heart rate. It is necessary to determine the upper and lower limits of the heart rate of pigs under quiet or normal activity conditions. Continuous monitoring can be used to obtain heart rate readings for a whole day, and then eliminate high fluctuation periods such as feeding, strenuous exercise, and individuals with diseases, and then calculate the difference between the minimum and maximum heart rate values ​​under a safe state. This value can be regarded as H range If a certain statistics shows that many pigs fall within the normal range of 70 to 110 beats / minute, then 110-70=40 can be regarded as H range .

[0127] Calculation process:

[0128] In the example calculation, let T = 37.8℃ (obtained from actual monitoring), T opt =39.0℃ (obtained through the statistics of thousands of body temperatures in the early stage), T range =1.0℃ (obtained after centralized analysis of body temperature fluctuations), H = 98 beats / min (recorded by a heart rate monitoring device), H opt = 85 beats / min (established after statistics of a large number of normal heart rate samples), H range=40 (difference between upper and lower limits of heart rate), substitute the parameters in turn:

[0129]

[0130] First calculate the temperature difference:

[0131] |37.8-39.0|=1.2

[0132] Multiply by the factor:

[0133]

[0134] Then calculate the heart rate difference:

[0135] |98-85|=13

[0136] Divide by your heart rate range:

[0137]

[0138] Comprehensive substitution:

[0139] C = 100-120-0.325 = -20.325

[0140] The results show that the calculated comfort score is already lower than 0, indicating that the current temperature deviation and heart rate deviation are both large. It is also possible to further determine at the breeding site whether there is inappropriate ambient temperature or abnormal physical condition of the pigs. The threshold of the score value can be classified between 0 and 100 based on daily monitoring habits. If it is greater than 80, it means the condition is ideal. If it is between 50 and 80, it means a certain degree of deviation. If it is lower than 50, it is necessary to be alert to whether the pigs are uncomfortable. When it is less than 0, it can be considered that the condition has deviated significantly.

[0141] When adjusting the temperature setting value based on the comfort score, it is necessary to first compare the calculated score with multiple pre-set levels. For example, you can set 80, 60 and 40 integer values ​​as the normal state dividing line on site. Whenever the comfort score is higher than 80 points, the temperature will be maintained at the previously executed setting value. When the score falls between 60 and 80 points, it indicates that the pig's current body temperature and heart rate are somewhat different from the standard value, and it is necessary to fine-tune the current temperature setting value appropriately, such as raising or lowering the power level of the heating or cooling equipment by one level and observing for a few minutes, then measuring the heart rate and body temperature again and recalculating the score. When the score range continues to drop to 40 to 60 points, it means that the pigs are less adaptable to the current temperature, and the temperature setting value needs to be significantly increased or decreased to improve the on-site environmental conditions. In this case, the output proportion of the corresponding equipment will be increased or the cooling or heating time will be extended. If the comfort score has dropped below 40 points, it will be regarded as a serious deviation. The most powerful cooling or heating measures need to be directly started and the heart rate and body temperature need to be frequently monitored to see if there are large fluctuations or sensor failures. If it is confirmed that the sensor is working normally and the pigs continue to have abnormal heart rate and body temperature, the relevant maintenance personnel will be notified to intervene to check or isolate the pigs. During the whole process, after each change in the temperature setting value, it is necessary to record the environmental information at this time, such as the upper and lower temperature limits and the temperature readings of each monitoring point. These records are recorded in the system to form a continuous historical data sequence in order to track the adjustment content of each time period. Finally, the temperature setting value is updated through this score-based adjustment method to generate an adjusted temperature setting.

[0142] The steps to obtain the temperature control strategy are:

[0143] Based on the adjusted temperature setting, the pig’s current heart rate and body temperature data are extracted to obtain comprehensive biofeedback information;

[0144] Based on the comprehensive biofeedback information, the environmental adaptability index is calculated using the following formula:

[0145]

[0146] Among them, E represents the environmental adaptability index, T actual is the pig's body temperature, T set is the adjusted temperature setting value, H actual is the pig's heart rate, H target is the target heart rate value;

[0147] Analyze and optimize temperature control strategies based on environmental adaptability indicators.

[0148] Specifically, based on the adjusted temperature setting, the heart rate and body temperature of each pig in the farm need to be recorded in correspondence according to the identification number. When recording, the heart rate measuring device is first periodically tested, and the real-time heart rate value is read through the sensor attached to the pig's chest or ventral position, and the collection time and pig number are attached to each reading. If the current heart rate is found to be over 120 beats / minute or lower than 50 beats / minute, it is compared with the common intervals in the previous statistics to determine whether excessive tension or depression occurs and retain the record for further comparison. Body temperature detection can be carried out through continuous observation by a close-fitting body temperature measuring instrument or an infrared probe, and a reading is obtained every five minutes. Each reading is sorted according to the timestamp and then Compare with the normal body temperature range of 38.5℃ to 40.0℃. If the data appears to be higher than 40.0℃ or lower than 38.5℃ for multiple times, it is marked as a deviation and combined with the previously set fault detection procedure to rule out the possibility of equipment failure. If the equipment is confirmed to be normal, the abnormal value is included in the record for subsequent comparison. Then all heart rate and body temperature records are indexed and matched with the previously adjusted temperature settings. The temperature setting at this point in time is indicated by adding an associated tag to each record. If pigs from different batches are in different temperature zones, the area numbers need to be distinguished and managed separately. Finally, all valid heart rate data and body temperature data are summarized to form a complete time series table, and a set of comprehensive biofeedback information is formed through continuous integration and identification.

[0149] The benefit of the formula is that it measures the degree of deviation of body temperature through the exponential function part, and measures the degree of deviation of heart rate through the absolute difference in the denominator, so that body temperature and heart rate jointly affect the final indicator value.

[0150] T actual The steps to obtain the parameter are as follows: This parameter represents the actual body temperature of the pig at the current moment, and needs to be recorded during the continuous monitoring process of the farm. First, select a body temperature measurement device with high credibility, such as a surface contact sensor, set the acquisition frequency to once every five minutes, and accumulate temperature values ​​for multiple periods of time within a day. Then, mark and troubleshoot the extreme values ​​that are faulty or significantly deviate from the normal range of 38.5℃ to 40.0℃. After confirming that the sensor is normal, extract the most recent temperature reading of the pig that is not marked as abnormal as T actual For example, if the monitoring result at 10:20 on the same day is 39.2℃, then 39.2 can be used as the T of this period. actual .

[0151] T setThe steps for obtaining the parameters are as follows: This parameter is the adjusted temperature setting value, which comes from the ambient temperature adjustment result made in the previous step based on multiple references such as comfort or predicted adjustment schemes. It is necessary to query the currently executed target temperature in the constant temperature device control panel of the farm, and verify it with the value and the sensor reading. If the set temperature is obviously inconsistent with the measured ambient air temperature, it is necessary to identify whether there is a fault or hysteresis effect. If the system is in an error-free state, the current set value can be directly extracted as T set For example, if the temperature in a local area is set to 28°C, it is recorded as T set =28.

[0152] H actual The steps to obtain the parameter are as follows: This parameter represents the current heart rate of the pig. Data needs to be obtained through a heart rate sensor or a short-term attached electrocardiogram sensor. The data is continuously monitored at a fixed rhythm of collecting data every few minutes, and the latest legal record is taken as H after removing noise. actual If a certain acquisition is 95 times / min and the measurement failure is eliminated, it can be determined that 95 times / min is the current H actual .

[0153] H target The steps to obtain the parameter are as follows: This parameter represents the target heart rate value, which is used to measure the degree of deviation of the current heart rate. It is necessary to find the heart rate distribution of healthy pigs during routine activities in the data accumulated in the early stage, select the peak or median of this distribution as the target heart rate, and then make corrections to the genetics and daily growth conditions of a specific pig group to form a standard heart rate value that adapts to the characteristics of the group. For example, a large number of statistics show that the stable heart rate of this batch of pigs is mostly concentrated at around 85 beats / minute, so 85 can be set as H target .

[0154] Calculation process:

[0155] In the example scenario, let T actual =39.2℃, T set =28℃,H actual =95 times / min,

[0156] H target =85 times / min, now substitute the above values ​​into:

[0157]

[0158] Calculate the exponential part first:

[0159] (39.2-28)=11.2

[0160]

[0161] exp(-2.24)≈0.106

[0162] 1+0.106=1.106

[0163]

[0164] Then calculate the heart rate deviation part:

[0165] |95-85|=10

[0166] 1+10=11

[0167]

[0168] Finally, the comprehensive multiplication:

[0169] E=0.904×9.09≈8.22

[0170] The results show that the environmental adaptability index is approximately 8.22, indicating that at this temperature setting and heart rate level, the pigs' overall adaptability to the environment is not high. If the subsequent collected indicators are closer to 20 or 30, it indicates that the degree of adaptation has improved. If it continues to be below 10, it usually indicates that the current temperature control strategy needs to be re-evaluated.

[0171] Based on the environmental adaptability index, it is necessary to first compare the values ​​calculated at each moment with several reference intervals previously established based on field monitoring experience. For example, the environmental adaptability index can be divided into different levels such as 0 to 10, 10 to 20, and 20 to 30. Each level corresponds to a specific temperature control strategy adjustment step. The actual measured index values ​​are then substituted into these levels to determine whether they exceed the set thresholds. If the environmental adaptability index is below 10 during a certain period of time, it means that the current temperature difference and heart rate deviation are both obvious. The adjustment range can be directly increased or decreased and the time interval for re-evaluation can be shortened. If the index is between 10 and 20, it means that the temperature setting value can be slightly adjusted first and observed patiently for several minutes. If it is observed and measured again, If the obtained indicator is still in this range or further drops, the adjustment will be strengthened in the subsequent links. If the indicator exceeds 20, it is considered that the current temperature is close to the required level. In this case, it is only necessary to make fine adjustments based on the actual temperature change trend and pay attention to the heart rate trend. The analysis process can also refer to the changing pattern of pig activity at different times of the day and fine-tune the threshold range accordingly. For example, the adjustment sensitivity can be reduced accordingly at night, and a stricter indicator upper limit can be set during feeding or daytime active periods to track drastic fluctuations and respond in time. Finally, these judgment results are summarized into a temperature control strategy description containing multiple rules, so that managers can perform corresponding adjustment actions in subsequent steps to obtain an optimized temperature control strategy.

[0172] The steps to obtain equipment operation records are as follows:

[0173] Operate heating and cooling equipment according to temperature control strategies and perform temperature adjustment activities, including setting new temperature parameters and starting or stopping equipment operation;

[0174] Record each equipment operation data, including operation time, equipment response and temperature changes before and after, and generate equipment operation records.

[0175] Specifically, according to the temperature control strategy, when operating the heating and cooling equipment, it is necessary to first select the instruction items corresponding to the current environmental adaptability index and temperature deviation from the multiple rules obtained in the previous step, and then perform the corresponding adjustment activities in the prescribed order. The execution process includes comparing the difference between the current temperature reading and the most recent set temperature one by one, and matching the difference with the reference range. For example, when the difference is higher than 5°C, the power value of the heating or cooling equipment is increased by one level and maintained for three minutes before observation. When the difference is in the range of 1°C to 5°C, only the output power of the equipment is fine-tuned and the temperature of the monitoring point is collected again after five minutes. If the difference continues to decrease, it will continue to be adjusted here. The gear remains stable. If the difference is still large, check the working condition of the equipment and recalculate the adaptability index to decide whether to switch to a higher or lower power level immediately. Then, during the equipment execution process, it is necessary to combine the temperature summary of multiple monitoring points in the breeding area to determine whether different locations are close to the preset values. If there is still a large deviation in some corners, extend the heating or cooling time of the area and observe whether the heart rate record fluctuates abnormally with the temperature adjustment. If it is confirmed that all data are stable, maintain the current operation plan unchanged, perform additional temperature setting value modifications or extend the operating time for the monitoring points that have deviated significantly, and obtain the final temperature adjustment activity.

[0176] When recording each equipment operation data, you can first read the ambient temperature at this time before each heating or cooling equipment is started and record the timestamp of the current moment, and then collect the operating response of the equipment in real time. For example, the heating speed of the heating equipment or the cooling speed of the cooling equipment is included in the key observation range. If it is monitored that the temperature difference drops from 5°C to 2°C within the set five-minute test period, its response speed is quantitatively recorded. If the temperature still does not drop significantly within the same period, it is marked that the output capacity of the equipment is attenuated or has a tendency to fail. Then, after the operation is completed, the final temperature change amplitude at this time must be recorded again and the corresponding timestamp and equipment number must be attached so that the two temperature comparison results can be included in the comprehensive query. If the temperature rises in the opposite direction after the equipment is started, this phenomenon is marked as an abnormal record and the line or control valve of the equipment is re-inspected in the subsequent collection. During the whole process, each operation must be summarized according to the three key fields of operation time, equipment response, and temperature change before and after. After the summary is completed, the equipment operation record is obtained.

Claims

1. A pig farm breeding environment temperature control system, characterized in that: The system comprises: The temperature acquisition module collects the temperature data of each monitoring point in the pig farm in real time, converts the temperature data into a temperature value sequence, and generates a temperature original data sequence; based on the temperature original data sequence, the average temperature value and temperature difference are calculated to obtain the basic temperature information of the environment; A temperature analysis module receives the basic ambient temperature information, analyzes temperature trends and abnormal points, determines whether the temperature needs to be adjusted through trend analysis, and generates a temperature adjustment signal; predicts future temperature changes based on the temperature adjustment signal and formulates a prediction adjustment plan; The biofeedback integration module receives the heart rate and body temperature of the pig, evaluates the comfort of the pig in combination with the prediction adjustment plan, adjusts the temperature setting value, and generates an adjusted temperature setting; based on the adjusted temperature setting, analyzes the adaptability of the pig to the current environment, optimizes the temperature control strategy, and obtains the temperature control strategy; The execution control module receives the temperature control strategy, controls the heating or cooling equipment to adjust the temperature according to the control strategy, and generates equipment operation records.

2. The pig farm breeding environment temperature control system according to claim 1, characterized in that: The steps for obtaining the original temperature data sequence are as follows: Use multiple temperature sensors located inside the pig farm to monitor the ambient temperature of each monitoring point in real time, and record the temperature data of each monitoring point in real time. The data of each sensor includes a timestamp and a temperature value, and obtain temperature data including a timestamp and a temperature value; Based on the temperature data including the timestamp and the temperature value, all the temperature data are sorted by the timestamp and uniformly converted into a temperature value sequence to obtain a temperature original data sequence.

3. The pig farm breeding environment temperature control system according to claim 1, characterized in that: The steps for obtaining the basic ambient temperature information are as follows: Based on the original temperature data sequence, the temperature values ​​recorded at each monitoring point are summed up, and the number of temperature records is counted to obtain the total temperature and the number of monitoring points; Using the total temperature and the number of monitoring points, the average temperature value of the pig farm environment is calculated by dividing the total temperature by the number of monitoring points to generate an average temperature value; Based on the temperature original data sequence, the highest temperature value and the lowest temperature value in the sequence are identified, the difference between the highest temperature value and the lowest temperature value is calculated to obtain the temperature difference, and the ambient basic temperature information is obtained in combination with the average temperature value.

4. The pig farm breeding environment temperature control system according to claim 1, characterized in that: The steps of obtaining the temperature adjustment signal are: Parsing the temperature data sequence in the basic ambient temperature information, and dividing the time series data into fixed time windows to obtain a temperature time series data set; Based on the temperature time series data set, the temperature trend change value is calculated, and the calculation formula is: Among them, M k is the temperature trend change value at the kth time point, P i is the temperature value at the i-th time point, P i-1 is the temperature value at the previous time point, t i is the timestamp of the i-th time point, t i-1 is the timestamp of the previous time point, m is the total number of observation points in the time window, b represents the impact scale of the data in the time window on the current calculation point, and t k is the timestamp of the kth time point; Based on the temperature trend change value, an abnormal point in the temperature time series is identified, and combined with the positive and negative directions of the temperature trend change value, it is determined whether the current temperature change needs to be adjusted, and a temperature adjustment signal is generated.

5. The pig farm breeding environment temperature control system according to claim 1, characterized in that: The steps for obtaining the forecast adjustment plan are: Based on the temperature adjustment signal, the future temperature change is calculated using the following formula: Where, ΔT future Represents the predicted future temperature change, T last is the temperature value, T avg is the average temperature value of the time series, represents the temperature change rate of the time series, T var is the temperature variability, β adjusts the effect of the difference between the temperature and the mean temperature, γ adjusts the effect of the square root of the rate of temperature change, and δ adjusts the effect of the logarithm of the temperature variability; Based on the future temperature change, a temperature adjustment strategy is formulated and a predicted adjustment plan is generated.

6. The pig farm breeding environment temperature control system according to claim 1, characterized in that: The steps for obtaining the adjusted temperature setting are: Obtaining the heart rate and body temperature data of the pigs, and combining the prediction and adjustment scheme to obtain biofeedback data; Based on the biofeedback data, the comfort score is calculated using the following formula: Where C represents the comfort score, T is the current body temperature, and T opt is the standard body temperature, T range is the normal fluctuation range of body temperature, H is the current heart rate, and H opt is the standard heart rate, H range This is the normal fluctuation range of heart rate; Based on the comfort score, the temperature setting is adjusted to generate an adjusted temperature setting.

7. The pig farm breeding environment temperature control system according to claim 1, characterized in that: The steps for obtaining the temperature control strategy are: Based on the adjusted temperature setting, and extracting the pig's current heart rate and body temperature data, comprehensive biofeedback information is obtained; Based on the comprehensive biofeedback information, the environmental adaptability index is calculated using the following formula: Among them, E represents the environmental adaptability index, T actual is the pig's body temperature, T set is the adjusted temperature setting value, H actual is the pig's heart rate, H target is the target heart rate value; Based on the environmental adaptability index, the temperature control strategy is analyzed and optimized.

8. The pig farm breeding environment temperature control system according to claim 1, characterized in that: The steps for obtaining the device operation record are: According to the temperature control strategy, operate the heating and cooling equipment and perform temperature adjustment activities, including setting new temperature parameters and starting or stopping equipment operation; Record each equipment operation data, including operation time, equipment response and temperature changes before and after, and generate equipment operation records.

Images