An intelligent control method and system for environmental zoning of a livestock house of a large-scale farm

Abstract

The present application relates to the technical field of breeding house, and particularly relates to a livestock house environment partition intelligent regulation method and system for large-scale breeding farm. The method determines a reference position through temperature abnormal period analysis of a measurement position; determines actual influence factors based on the deviation of the persistence and occurrence of drinking water behavior between the measurement position and the reference position according to drinking water information analysis; and obtains regional regulation demand indexes for fan regulation by combining cage vertical distribution and influence factor changes, as well as measurement position abnormal frequency and influence factor trends. The present application determines the actual temperature influence by jointly analyzing the drinking water behavior of the livestock group, and then adjusts the current regulation demand strength by combining the cage distribution of multi-layer cage breeding houses and abnormal accumulation, so that the environmental regulation takes into account the health of the livestock group and resource conservation, and better meets and adapts to the ecological breeding goal.

Description

Technical Field

[0001] This invention relates to the field of livestock housing technology, specifically to a method and system for intelligent zoning and control of livestock housing environment in large-scale livestock farms. Background Technology

[0002] Most existing livestock sheds adopt a fully enclosed, multi-layered cage system, which inherently suffers from problems such as temperature stratification, ventilation dead spots, and localized heat accumulation. The core of ecological farming is respecting the natural growth patterns of animals. However, livestock and poultry commonly found in large-scale farming, such as pigs, chickens, ducks, and rabbits, are all homeothermic animals. Their thermoregulation ability determines that temperature is fundamental to their health, and environmental temperature is a crucial factor affecting livestock and poultry health and farming production.

[0003] Current livestock housing environmental control typically relies on temperature sensors at a few fixed points within the housing to collect data. This approach fails to consider the temperature differences distributed across multiple cages and the varying adaptability of livestock to temperature changes. Using average temperature as the basis for control not only fails to resolve the inherent temperature issues in multi-layer cage housing but may also lead to energy waste. Summary of the Invention

[0004] To address the technical problems in the prior art, the present invention aims to provide a method and system for intelligent zoning and control of livestock housing environment in large-scale farms. The specific technical solution adopted is as follows: This invention provides a method for intelligent zoning and control of livestock housing environment in large-scale farms, the method comprising: Acquire temperature data at each measurement location in the livestock cage at each time, as well as drinking water information for all cages in the preset control area corresponding to the measurement location at each time. For each measurement location, abnormal time periods are determined based on the deviation of temperature data from expectations over time; temperature fluctuation factors are obtained based on the degree of sustained temperature deviation during each abnormal time period; and reference locations for the corresponding measurement locations are determined based on the temperature fluctuation factors for other measurement locations during each abnormal time period. For each abnormal period, based on the drinking information of all cage locations in the corresponding measurement location and the preset control area, the drinking behavior characteristics are analyzed and compared with the reference location to determine the actual influence factor of each location in the abnormal period. Based on the vertical position distribution and actual influencing factor changes of each cage position in the preset control area corresponding to each measurement position, combined with the distribution characteristics of abnormal periods in the time sequence of the measurement position, the changing trend of actual influencing factors and temperature fluctuation factor, the regional control demand index of each measurement position is obtained. Wind turbine regulation is based on regional regulation demand indicators determined by the location.

[0005] Furthermore, the method for obtaining the abnormal time period includes: For any given measurement location, in the historical time series of that location, the moments when the temperature data is outside the expected temperature range are considered abnormal moments; and the time periods consisting of consecutively adjacent abnormal moments are considered abnormal time periods.

[0006] Furthermore, the method for obtaining the temperature fluctuation factor includes: For any abnormal time period at any measurement location, calculate the difference between the temperature data at each moment during the abnormal time period and the boundary value of the expected temperature range, and take the minimum value of the difference as the temperature deviation at each moment. Multiply the maximum temperature deviation during the abnormal period by the duration of the abnormal period and then normalize the result to obtain the temperature fluctuation factor for that abnormal period.

[0007] Furthermore, the method for obtaining the reference position includes: For any abnormal time period at any measurement location, for other measurement locations besides that measurement location, the product of the number of abnormal moments at the other measurement locations during the abnormal time period and the average value of the temperature fluctuation factor is calculated to obtain the abnormal probability index for each other measurement location; if there are no abnormal moments at any other measurement location, the abnormal probability index can be set to zero. Other measurement locations where the probability of anomalies is minimized are used as reference locations.

[0008] Furthermore, the method for obtaining the actual impact factor includes: For any abnormal time period at any measurement location, each cage position in the preset control area corresponding to the measurement location is sequentially taken as the analysis location; the drinking status of the analysis location is obtained from the drinking information at each moment during the abnormal time period; the moment when the drinking status of the analysis location is turned on is taken as the drinking time of the analysis location; each consecutive drinking time is taken as a drinking period of the analysis location during the abnormal time period. The proportion of drinking times during the abnormal period at the analysis location is used as the drinking occurrence indicator for the analysis location; the total number of drinking and non-drinking times at the analysis location during the abnormal period is negatively correlated and normalized to obtain the drinking continuity indicator for the analysis location. After calculating the absolute values ​​of the differences in drinking water occurrence indicators and drinking water sustainability indicators between the analysis location and the reference location, the mean of the two absolute values ​​of difference is taken as the actual influencing factor of the analysis location during the abnormal period.

[0009] Furthermore, the method for obtaining the regional regulation demand indicators includes: For any given measurement location, the current urgency of temperature control at that measurement location is obtained based on the interval between abnormal periods in time series, the growth trend of actual influencing factors during consecutive abnormal periods, and the temperature fluctuation factor. Based on the number of cages distributed vertically above each cage position in the preset control area corresponding to the measurement location, and combined with the adjacent deviation value in the vertical direction of the cage position, the ventilation requirement of each cage position is obtained. After calculating the product of ventilation demand and temperature control urgency for each cage location, the sum of all products is normalized to obtain the regional control demand index for that measurement location.

[0010] Furthermore, the method for obtaining the urgency of temperature control includes: After obtaining the interval duration between every two adjacent abnormal time periods in the time series of the measurement location, the mean of all interval durations is calculated and negatively correlated to be used as the frequency of abnormality at the measurement location. The minimum value of all temperature fluctuation factors over the time series at the measurement location is taken as the minimum fluctuation influence at that measurement location. All actual influencing factors at the measurement location are arranged in the order of distribution during abnormal time periods to obtain the influence sequence; the influence sequence is fitted with a straight line and the slope after fitting is used as the degree of improvement of the influence trend at the measurement location; The urgency of temperature control at the measurement location is obtained by weighted summation and normalization of the abnormal frequency, minimum fluctuation impact, and impact trend improvement of the measurement location.

[0011] Furthermore, the method for obtaining the ventilation demand includes: For any cage position in the preset control area of ​​the measurement location, in the most recent abnormal period from the current time, calculate the absolute value of the difference between the actual influence factor between the cage position and each adjacent cage position in the vertical direction, and then take the mean of all the absolute values ​​of the difference as the neighbor-to-neighbor deviation value of the cage position. The number of cages located vertically above the cage is normalized and used as the distribution influence value of the cage's location. The product of the distribution influence value of the cage location and the adjacent deviation value is taken as the ventilation demand of the cage location.

[0012] Furthermore, the wind turbine regulation based on the regional regulation demand index of the measured location includes: Based on the regional control demand index at the measured location, a new wind speed is determined between the preset base wind speed and the preset maximum wind speed, so that the new wind speed is positively correlated with the regional control demand index; the wind turbine at the measured location is then adjusted to operate at the new wind speed.

[0013] The present invention also provides an intelligent control system for the zoning of livestock housing environment in a large-scale farm, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the intelligent control method for the zoning of livestock housing environment in a large-scale farm as described above.

[0014] The present invention has the following beneficial effects: This invention analyzes abnormal temperature periods at measurement locations and then identifies reference locations during these abnormal periods as a baseline for analyzing the impact of temperature disturbances. Based on drinking water information analysis, the persistence and frequency of drinking behavior deviations between the measurement and reference locations are determined to identify actual influencing factors. This can be combined with livestock physiological responses to quantify the actual impact of temperature, allowing for comparison with livestock drinking behavior under normal environmental conditions, avoiding the limitations of relying solely on temperature data. By combining the vertical distribution of cages and changes in influencing factors, the possibility of airflow obstruction by upper and lower cage layers is analyzed at each location. The urgency of regulation at the measurement locations is analyzed by combining the frequency of abnormalities and the trend of influencing factors, resulting in a comprehensive regional regulation demand index. This index is adapted to the vertical temperature difference characteristics of multi-layer cage housing and considers the cumulative effect of impacts, making regulation needs more reliable. Finally, based on the regional regulation demand index, fan regulation is applied to achieve targeted temperature control, with results that better meet actual needs. This invention determines the actual temperature impact by jointly analyzing livestock drinking behavior, and then adjusts the current regulation intensity based on the cage distribution and abnormal accumulation in multi-layer cage housing. This ensures that environmental regulation balances livestock health and resource conservation, better meeting and aligning with ecological farming goals. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A flowchart illustrating an intelligent control method for zoning livestock housing environment in a large-scale farm, as provided in one embodiment of the present invention. Figure 2 This is a schematic diagram of the distribution structure of a livestock cage according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the control zone division of a livestock cage according to an embodiment of the present invention; Figure 4 The flowchart illustrates a method for obtaining regional regulation demand indicators according to an embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method and system for intelligent zoning control of livestock housing environment in a large-scale farm according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details the specific scheme of the intelligent control method and system for zoning livestock housing environment in large-scale farms provided by this invention.

[0020] Please see Figure 1 The diagram illustrates a flowchart of an intelligent zoning control method for livestock housing environment in a large-scale farm, according to an embodiment of the present invention. The method includes the following steps: S1: Obtain temperature data at each measurement location in the livestock cage at each time, as well as drinking water information for all cages in the preset control area corresponding to the measurement location at each time.

[0021] It should be noted that in this embodiment, the target livestock shed is a multi-layered, three-dimensional cage rearing model. This model effectively improves the utilization rate of space resources by utilizing multiple layers of space. However, because the multiple rows of cages create physical barriers, air circulation between rows and layers is poor, which can easily lead to environmental temperature stratification and localized heat spots, thereby reducing the overall environmental temperature control effect of the livestock shed.

[0022] In this step, based on the longitudinal ventilation characteristics of the livestock shed, several cages are selected at equal intervals along the cage frame from the air inlet to the air outlet as temperature measurement locations. A temperature sensor is installed at each of these locations. In this embodiment, a DS18B20 digital temperature sensor is used, with a temperature measurement range of -55℃ to +125℃ and an inherent measurement accuracy of ±0.5℃ within the range of -10℃ to +85℃. Each temperature sensor is connected to the corresponding row of area controllers via a fieldbus. The area controllers in each row collect temperature data to the central processing unit (CPU). Every minute, the CPU triggers all area controllers to collect temperature data synchronously via a broadcast command. Each collected temperature data point includes a local timestamp from the controller. After receiving the data, the CPU corrects for the timestamp deviation based on the clock, ensuring that the temperature data from all measurement locations are aligned in time. Temperature data is recorded every minute, and the temperature at each measurement location is collected within a preset time period.

[0023] The preset time period is set to 2 hours prior to the current time. This preset duration is based on the following: Under longitudinal ventilation conditions, the complete air exchange cycle from the air inlet to the air outlet in a large-scale livestock shed typically takes 8 to 15 minutes. The preset time of 2 hours covers 8 to 15 complete air exchange cycles, which not only fully reflects the steady-state or dynamic changes in the thermal environment of the livestock shed under the current ventilation conditions, but also takes into account the controllability of data storage and processor computational overhead. Implementers can adjust this according to the specific implementation scenario.

[0024] Please see Figure 2 The diagram shows a distributed structure of livestock cages according to an embodiment of the present invention, wherein a row of cages has 5 layers, and the specific number of layers depends on the species being raised and the specifications of the cages (for example, laying hens are usually raised with 3 to 8 layers, broilers with 2 to 4 layers, rabbits with 2 to 3 layers, etc.). The thickened cage position in each row of cages is selected as the temperature measurement position.

[0025] It should be noted that livestock and poultry primarily rely on active contact for daily drinking. Therefore, the current layout of waterers in livestock sheds often uses nipple-type waterers installed parallel to the feed trough. This type of waterer only opens when the livestock or poultry touches the end of the nipple. Specifically, when a livestock or poultry needs water, it will actively touch the trigger position of the waterer, pushing the valve core to open, allowing it to drink from below the outlet. When the livestock or poultry stops touching the nipple, the valve core closes again under the return force of the spring, and the water flow immediately stops.

[0026] It should be understood that the drinking behavior of livestock and poultry (such as pigs, chickens, rabbits, etc.) under large-scale farming conditions is generally related to the temperature environment. When the ambient temperature deviates from the suitable range, livestock and poultry adjust their internal heat balance by increasing the frequency and amount of drinking. Therefore, changes in drinking behavior can be used as an indirect physiological indicator of temperature stress.

[0027] In this embodiment, a flow sensor is installed at the water outlet of the nipple drinker in each cage to determine the presence and flow rate of water. When a water flow signal is detected, i.e., the pulse count value of the flow sensor is ≥1 for one consecutive second, the drinking status in the drinking information is set to "on"; otherwise, the drinking status is set to "off". The drinking information of each cage is synchronously collected by the area controller at the same sampling time as the temperature data and uploaded to the central processing unit via the fieldbus. The central processing unit associates and stores the temperature data and drinking information at the same time according to the cage location index, forming a time-aligned data record.

[0028] Since localized temperature anomalies often reflect an overall imbalance in airflow and heat distribution within a given area, this invention uses each measurement location as a center and divides the area into regions through preset control zones for regional control analysis to meet regional ventilation control requirements. In this embodiment, the preset control zone is set as a rectangular area of ​​M×N centered on the measurement location, thereby ensuring that the preset control zones at different measurement locations do not overlap and that all cages are divided.

[0029] In this embodiment, the preset control area is set to a 3×5 range, that is, centered on the cage where the measurement location is located, extending horizontally by one column and vertically by two layers. Considering that the effective air supply coverage of a single inter-row duct fan is approximately 2.5m×4.0m, the size of this area matches the effective air supply coverage of a single inter-row duct fan, ensuring the uniformity of airflow distribution within the control area.

[0030] In another embodiment of the present invention, the preset control area can also be divided in the following alternative manner: Based on the simulation results of the airflow organization in the livestock house using computational fluid dynamics (CFD), the velocity field distribution of the longitudinal section of the livestock house is extracted. The local extreme value line of the airflow velocity gradient is used as the boundary line of the control area, and the livestock house is divided into natural control areas of varying sizes with boundaries matching the airflow distribution characteristics. A measurement position is set in each area. This alternative division method can further reduce the airflow crosstalk rate between adjacent areas, and is especially suitable for large-scale livestock houses with large cage row spacing, such as spacing ≥1.2m or non-uniform fan layout.

[0031] Please see Figure 3 The illustration shows a schematic diagram of the control area division of a livestock cage according to an embodiment of the present invention. The bolded outer frame of measurement position 1 in the illustration is a representative preset control area configuration.

[0032] S2: For each measurement location, determine the abnormal time period based on the deviation of the temperature data from the expected value over time; obtain the temperature fluctuation factor based on the degree of continuous temperature deviation during each abnormal time period; and determine the reference location of the corresponding measurement location based on the temperature fluctuation factor of other measurement locations during each abnormal time period.

[0033] In multi-level cage rearing, the airflow distribution in livestock sheds is uneven due to ventilation differences between different rows and layers of cages. For example, the vertical obstruction of the multi-level cages causes the upper cages to block the middle and lower layers, resulting in slow air circulation in the middle and lower layers and thus creating a temperature difference with the upper layers. Therefore, the temperature distribution in different locations within the same livestock shed is not uniform.

[0034] Livestock and poultry commonly found in large-scale farming, such as pigs, chickens, and rabbits, are all homeothermic animals. Homeothermic animals maintain their internal heat balance through thermoregulation. They can regulate their body temperature through behavioral and physiological adjustments, such as changing drinking frequency, adjusting activity levels, and dissipating heat through respiration and evaporation. However, prolonged exposure to unsuitable temperature ranges can disrupt physiological homeostasis, leading to decreased feed intake, reduced feed conversion ratio, and other performance impairments. Therefore, the first step is to analyze temperature fluctuations at each measurement location to identify periods of abnormal temperature.

[0035] In this embodiment of the invention, the method for obtaining abnormal time periods includes: for any measurement location, in the historical time sequence of the measurement location, the time when the temperature data at that time is outside the expected temperature range is taken as an abnormal time, and the time period consisting of consecutively adjacent abnormal times is taken as an abnormal time period.

[0036] The desired temperature range is a preset suitable temperature interval based on the breed and growth stage of the livestock and poultry being raised. In this embodiment, the reference desired temperature range for each livestock and poultry breed under large-scale farming conditions is determined as follows: According to the "NY / T 388-1999 Livestock and Poultry Farm Environmental Quality Standard," the desired temperature range for growing chickens is set at 16 to 22℃; according to "GB / T 17824.1-2008 Construction of Large-Scale Pig Farms," ​​the desired temperature range for fattening pigs is 15 to 22℃, and for adult rabbits, it is 15 to 25℃. For other livestock and poultry breeds or special growth stages, such as chicks requiring 32 to 35℃ during the brooding period and piglets requiring 25 to 30℃ during the nursery period, the desired temperature range is automatically determined by looking up a preset breed-age-temperature reference table in the livestock and poultry management system. This reference table is based on local livestock and poultry farming standards, and implementers can also customize its configuration based on actual farming experience and breed characteristics. When the temperature data is outside the desired temperature range, anomaly analysis is performed.

[0037] It should be understood that while livestock and poultry have a certain capacity to adapt to and buffer against changes in ambient temperature, when the magnitude and rate of temperature changes are too large and exceed their own thermoregulation capabilities, they will experience heat or cold stress responses. This will significantly reduce their feed intake, daily weight gain, egg production rate, feed efficiency, and other production performance indicators. Therefore, when the temperature fluctuation at the measurement location is large, it indicates that the temperature control at that location is relatively unstable, possibly due to poor ventilation, and adjustments should be made as soon as possible.

[0038] The temperature fluctuation factor is determined based on the degree of sustained deviation during abnormal periods. In this embodiment of the invention, for any abnormal period at any measurement location, the difference between the temperature data at each moment during the abnormal period and the boundary values ​​of the expected temperature range is calculated, and the minimum value of the difference is taken as the temperature deviation at each moment. It can be understood that the expected temperature range has two boundary values: an upper limit and a lower limit. Therefore, after calculating the absolute value of the difference between the temperature data at a moment and each boundary value, the minimum of the two is taken as the temperature deviation, reflecting the degree to which the temperature at the abnormal moment deviates from the ideal temperature range. For example, if the expected temperature range is [16℃, 22℃], and the temperature at a certain moment is 25℃, then the difference from the upper limit is |25-22|=3℃, and the difference from the lower limit is |25-16|=9℃. In this case, the minimum value of 3℃ is taken as the temperature deviation at that moment.

[0039] Furthermore, the maximum temperature deviation during the abnormal period is multiplied by the duration of the abnormal period and then normalized to obtain the temperature fluctuation factor for that abnormal period. The longer the abnormal temperature period at the measurement location and the greater the temperature deviation at the abnormal moment, the longer the livestock and poultry stay in the unsuitable environment, the more prominent the negative impact on their health and production performance, and the more necessary it is for the control system to make adjustments as soon as possible; therefore, the larger the temperature fluctuation factor.

[0040] It should be noted that, unless otherwise explicitly stated, all normalization processes in this invention employ the Min-Max normalization method to map the data to the [0,1] interval.

[0041] In this embodiment of the invention, for any abnormal time period at any measurement location, for other measurement locations besides that location, the product of the number of abnormal moments at those other measurement locations during the abnormal time period and the average value of the temperature fluctuation factor is calculated to obtain the abnormality probability index for each other measurement location. The fewer the number of abnormal moments at other measurement locations, the shorter the time the temperature deviates from the suitable range; the smaller the temperature fluctuation factor, the less the temperature fluctuation interference. Since other measurement locations may have multiple consecutive abnormal moments during the abnormal time period, meaning multiple abnormal time periods have occurred at other measurement locations, corresponding to multiple temperature fluctuation factors, the average value of the temperature fluctuation factor is calculated for analysis. If other measurement locations do not have any abnormal moments during the abnormal time period, meaning no abnormal time period has occurred at other measurement locations, then there is no temperature fluctuation factor, and the abnormality probability index can be directly set to zero.

[0042] Therefore, other measurement locations where the probability of anomaly is minimized can be used as reference locations. If there are multiple other measurement locations where the probability of anomaly is minimized, the one closest to the current measurement location is selected as the reference location by default. In other embodiments of the present invention, any one of these other measurement locations can also be selected as the reference location. In this case, the behavioral characteristics of the livestock and poultry population at the reference location are closer to the normal state without temperature stress, making it reasonable to use their drinking behavior as a normal benchmark.

[0043] S3: For each abnormal time period, based on the drinking information of all cage positions in the corresponding measurement location and the preset control area, analyze the drinking behavior characteristics and compare them with the reference position to determine the actual influence factor of each position during the abnormal time period.

[0044] In large-scale farming environments, although temperature fluctuations can affect the growth of livestock and poultry, differences in breed, age, body condition, and other physiological conditions among individuals lead to varying degrees of response to the same temperature fluctuations. Therefore, relying solely on temperature sensor data is insufficient to comprehensively assess the impact of temperature on livestock and poultry populations; further analysis of changes in livestock and poultry behavior is necessary.

[0045] It should be understood that the thermoregulation mechanisms of livestock and poultry under heat stress share commonalities: when the ambient temperature rises, the livestock and poultry first engage in behavioral thermoregulation, including increasing the frequency and amount of drinking, reducing feeding activity, and adjusting lying posture to increase the effective heat dissipation area; simultaneously, physiological thermoregulation is initiated, such as increasing respiratory rate to promote evaporative cooling through the respiratory tract and dilating skin blood vessels to promote convective cooling through the body surface. Increased drinking directly replenishes the water lost through respiration and evaporation from the body surface, and also helps lower body temperature by ingesting cooler drinking water. Therefore, when the ambient temperature deviates from the suitable range, it disrupts the water balance in livestock and poultry, leading to detectable changes in drinking frequency and patterns.

[0046] Based on the above mechanism, the impact of temperature fluctuations on livestock populations at each cage location can be analyzed by comparing the differences in drinking behavior between abnormal and reference locations. It should be noted that although decreased feed intake and reduced activity are also important behavioral indicators of heat stress in livestock populations, drinking behavior can be automatically monitored at low cost using flow sensors; its response time to temperature changes is relatively short, typically appearing within 10 to 30 minutes after the onset of heat exposure, meeting the timeliness requirements of environmental control; furthermore, drinking behavior is not limited by the photoperiod, allowing for continuous monitoring data. Therefore, in this embodiment, drinking behavior is used as the core analytical indicator.

[0047] Preferably, in this embodiment, the method for obtaining the actual impact factor includes: For any abnormal time period at any measurement location, each cage position in the preset control area corresponding to that measurement location is sequentially taken as the analysis location, and all cage positions in the control area are analyzed. The drinking status of the water dispenser is obtained from the drinking information at each moment during the abnormal time period at the analysis location, reflecting the usage of the water dispenser. Furthermore, the moment when the drinking status at the analysis location is "on" is taken as the drinking time of the analysis location, and each consecutive drinking time is taken as a drinking period at the analysis location during the abnormal time period, reflecting continuous drinking behavior.

[0048] Furthermore, the proportion of drinking times during the abnormal period at the analysis location is used as the drinking occurrence index for the analysis location. That is, the ratio of the number of drinking times to the total number of times during the abnormal period is used as the drinking occurrence index. The more drinking times there are, the greater the proportion of the total time during which drinking behavior occurs, reflecting an overall increase in the drinking demand of livestock and poultry groups, which is consistent with the trend of increased water loss caused by heat stress.

[0049] Statistical analysis was performed to negatively correlate the total number of drinking and non-drinking periods at the analyzed location during the abnormal period, and the data was normalized to obtain a drinking continuity index for the analyzed location. The number of periods reflects the frequency of drinking behavior switching. The smaller the total number of drinking and non-drinking periods, the better the continuity of the livestock and poultry group's drinking behavior, the longer the duration of each drinking session, and the more stable the behavior, indicating that the livestock and poultry are in a relaxed drinking state and are less affected by fluctuations in ambient temperature.

[0050] It should be noted that the negative correlation mapping described in this invention uses a mapping function. , where x is the original value, referring to the total number of drinking and non-drinking periods in this step. This mapping function maps the original value to the interval (0,1). The normalization process further employs the Min-Max normalization method to map the mapped value to the interval [0,1]. The larger the normalized drinking continuity index value, the better the drinking behavior of the livestock and poultry group during the abnormal period.

[0051] Since the reference location has the fewest anomalous moments and the smallest temperature fluctuation factor during the anomalous period, the disturbance caused by temperature to livestock and poultry populations is the weakest. Therefore, the drinking behavior of livestock and poultry populations at the reference location is almost unaffected by temperature fluctuations and can be approximated as the baseline of normal drinking behavior during this anomalous period. By comparing the drinking behavior of livestock and poultry populations at each analysis location with that at the reference location, the degree of impact of temperature fluctuations on livestock and poultry populations at each analysis location can be further analyzed. Similarly, the methods for obtaining the drinking occurrence and persistence indicators at the reference location during this anomalous period are consistent with those described above.

[0052] Finally, after calculating the absolute values ​​of the differences in water occurrence and persistence indicators between the analysis location and the reference location, the mean of these two absolute values ​​was taken as the actual influencing factor for the analysis location during the abnormal period. The greater the differences between the water occurrence and persistence indicators, i.e., the greater the actual influencing factor, the higher the degree to which the overall drinking behavior of the livestock and poultry population deviates from the normal baseline. This indicates that temperature fluctuations are more likely to have disrupted the water balance and metabolic homeostasis of the livestock and poultry, resulting in significant physiological stress responses. In such cases, the temperature at the analysis location needs to be regulated as soon as possible.

[0053] To facilitate the use of actual impact factors for grading in practical control systems, this invention further provides reference grading thresholds for actual impact factors: when the actual impact factor < 0.2, it indicates that the drinking behavior of livestock and poultry groups is only slightly different from the normal baseline, and temperature fluctuations have not yet caused significant physiological stress responses; when the actual impact factor is between 0.2 and 0.5, it indicates that livestock and poultry groups have experienced mild to moderate physiological stress, and observable deviations in drinking behavior have occurred; when the actual impact factor > 0.5, it indicates that livestock and poultry groups have experienced significant physiological stress responses, and the control system needs to prioritize responding to the temperature control needs of this area.

[0054] It should be noted that the above thresholds were obtained based on the following experimental method: In a large-scale farm, a 3-row × 5-layer cage structure was selected. Under natural high-temperature conditions in summer, with an outside temperature of 28 to 35°C, monitoring was conducted continuously for 14 days, recording temperature data, drinking behavior data, and respiratory frequency of livestock and poultry at each cage location. A 30% increase in respiratory frequency compared to normal levels was used as the criterion for significant physiological stress. ROC curve analysis was used to determine the optimal cutoff threshold for the actual influencing factor, which was then used as the above-mentioned grading threshold. The area under the curve was 0.87 ± 0.05, indicating that the influencing factor has a good ability to distinguish the physiological stress state of livestock and poultry.

[0055] S4: Based on the vertical position distribution and actual influencing factor changes of each cage position in the preset control area corresponding to each measurement position, combined with the distribution characteristics of abnormal time periods in the time sequence of the measurement position, the changing trend of actual influencing factors and temperature fluctuation factors, obtain the regional control demand index for each measurement position.

[0056] For the overall ambient temperature of large-scale livestock sheds, temperature anomalies are not isolated phenomena in a single location, but rather signals of an imbalance in airflow and heat distribution within the area. To achieve the overall health of the livestock population, further analysis of the cages surrounding each measurement location is needed to assess the ventilation requirements of the entire area.

[0057] In this embodiment of the invention, the method for obtaining regional regulation demand indicators is described in detail below. Figure 4The diagram illustrates a method for obtaining regional regulation demand indicators according to an embodiment of the present invention. The method includes the following steps: S401: For any measurement location, based on the interval between abnormal periods in time series, the growth trend of actual influencing factors during consecutive abnormal periods, and the temperature fluctuation factor, the current urgency of temperature control at the measurement location is obtained.

[0058] Because livestock and poultry populations have a cumulative effect on their tolerance to temperature stress, even if the impact of temperature fluctuations is small in the short term, the cumulative effect of temperature fluctuations over multiple historical periods can lead to a long-term sub-healthy state for the livestock and poultry populations. Therefore, it is necessary to analyze the distribution of historical abnormal periods at the measurement location to determine the urgency of temperature regulation at that location at the current moment, so as to intervene in a timely manner and avoid insufficient regulation that would reduce the productivity of livestock and poultry populations.

[0059] In this embodiment of the invention, after obtaining the interval duration between every two adjacent abnormal time periods in the time sequence of the measurement location, the mean of all interval durations is calculated and a negative correlation mapping is performed as the frequency of the abnormality at the measurement location. The smaller the interval duration, the more frequent the abnormality, the greater the current abnormality threat, and the higher the need for regulation. The negative correlation mapping is consistent with the method described in section S3.

[0060] Furthermore, the minimum value of all temperature fluctuation factors over the time series at the measurement location is taken as the minimum fluctuation impact at that measurement location. The larger the minimum temperature fluctuation factor during historical anomalous periods, the higher the level of even the mildest state of the anomalous situation, indicating a severe overall threat from temperature fluctuations and a higher demand for regulation.

[0061] Furthermore, all actual influencing factors at the measurement location are arranged in chronological order according to their distribution during abnormal time periods to obtain an influence sequence. A linear regression is then performed on the influence sequence, and the slope of the fitted sequence is used as the degree of improvement in the influence trend at the measurement location. Specifically, the linear regression is performed using the least squares method, and the regression model is as follows: Where x is the chronological number of the abnormal period, y is the actual influencing factor in the corresponding abnormal period, and the fitting slope is... This refers to the degree of influence on the trend improvement. When the goodness of fit... When the slope is less than 0.5, it indicates that the actual influencing factor does not show a clear linear trend over time, and the degree of improvement in the influence trend is set to 0. The larger the slope, the more it indicates that the actual influencing factor is showing a continuous upward trend, and the degree of temperature stress experienced by livestock and poultry populations is intensifying, at which point the need for regulation is higher.

[0062] Finally, the frequency of anomalies, the minimum fluctuation impact, and the increase in impact trend at the measurement location are weighted, summed, and normalized to obtain the urgency of temperature control at that location. For example, the sum of the weighting coefficients in the weighted summation is 1, where the weight of the frequency of anomalies is 0.35, the weight of the minimum fluctuation impact is 0.35, and the weight of the increase in impact trend is 0.30; implementers can adjust these weighting coefficients according to the actual aquaculture scenario and control strategy.

[0063] The normalization process uses the Min-Max normalization method to map the weighted sum to the [0,1] interval. The greater the urgency of temperature control, the more persistent the threat to the health of livestock and poultry at the measurement location is to the continuous and superimposed temperature stimulation, and the higher the need for temperature control to avoid losses such as reduced feed conversion rate, decreased egg production or milk production, or failure to reach the target slaughter weight.

[0064] S402: Based on the number of cages distributed vertically above each cage position in the preset control area corresponding to the measurement location, and combined with the adjacent deviation value in the vertical direction of the cage position, the ventilation requirement of each cage position is obtained.

[0065] It should be noted that multi-layered cages exhibit a natural vertical temperature gradient. Due to rising hot air and the cages' own shading, the temperature in the upper cages is generally higher than in the middle and lower layers; for example, the temperature difference between the upper and lower layers can reach 3 to 5°C in summer. Furthermore, due to different ventilation paths, ammonia and carbon dioxide concentrations also exhibit vertical stratification across different cage layers. The lower layers, where airflow is obstructed, typically have significantly higher ammonia concentrations than the upper layers. Excessively high ammonia concentrations can exacerbate the adverse effects of temperature stress on the respiratory mucosa of livestock and poultry. Therefore, even within the same area, the actual environmental conditions of cages with different layers can vary considerably. To improve the accuracy of overall environmental temperature control in this area, further analysis of the ventilation requirements of each cage location is necessary.

[0066] In this embodiment of the invention, for any cage location within the preset control area of ​​the measurement location, during the most recent abnormal time period from the current time, the absolute value of the difference between the actual influencing factor of that cage location and each adjacent cage location in the vertical direction is calculated. The mean of all absolute values ​​of the differences is then used as the neighbor-to-neighbor deviation value of that cage location. The difference in the actual influencing factor between the cage location and the two adjacent layers above and below reflects the difference in drinking behavior between the livestock and poultry population at that cage location and the livestock and poultry populations on the upper and lower layers. A larger difference indicates a potentially greater difference in the actual environment between the cage location and the adjacent layers, and a higher ventilation requirement.

[0067] Meanwhile, the number of cages vertically above the cage is normalized and used as the distribution influence value of that cage location. The normalization process uses the Min-Max normalization method. The more cage layers there are above, the more cages the airflow needs to penetrate and bypass the livestock group before reaching that cage, resulting in more severe airflow obstruction and thus lower ventilation efficiency at that location.

[0068] In one embodiment of the present invention, specifically, when there are no other cages above the cage (i.e., it is at the top layer), the distribution influence value of the cage's position can be set to 0.1. This value means that although there are no cages directly obstructing the top layer cage, it is still affected by roof heat radiation and the convergence of rising hot air currents, and there is a certain degree of ventilation demand baseline value of 0.1, rather than no ventilation demand at all. The specific value can be adjusted by the implementer according to the structure of the livestock shed.

[0069] Finally, the product of the distribution influence value and the adjacent deviation value of the cage location is taken as the ventilation demand for that location. The larger the distribution influence value and the adjacent deviation value, the higher the ventilation demand for that cage location. It should be noted that a product rather than a weighted sum is used here because the distribution influence value reflects the physical obstruction dimension of ventilation, while the adjacent deviation value reflects the actual ventilation demand dimension. When there is no obstruction above the top cage, one dimension will be extremely low. Even if another dimension is high, the overall ventilation demand should be significantly suppressed, meaning the product approaches 0. If a weighted sum were used, the ventilation demand might be overestimated due to a high value in a single dimension, leading to excessive ventilation and cooling of the top cage.

[0070] S403: By combining the ventilation demand and temperature control urgency of all cages in the preset control area, the regional control demand index of the measurement location is obtained.

[0071] It should be noted that when the ventilation requirement of each cage within the area corresponding to the measurement location is higher, and the urgency of temperature control at that measurement location is higher, it indicates that the temperature in that area has seriously deviated from the suitable range for livestock and poultry groups at the current moment, and needs to be controlled as soon as possible. However, due to poor air circulation, the overall need for control is greater, and more temperature control intervention is required in that area.

[0072] In this embodiment of the invention, after calculating the product of ventilation demand and temperature control urgency at each cage location, the sum of all products is normalized to obtain the regional control demand index for that location. The normalization process uses the Min-Max normalization method, mapping the sum to the [0,1] interval. The larger the regional control demand index, the higher the degree of control.

[0073] S5: Wind turbine regulation is performed based on the regional regulation demand indicators of the measured location.

[0074] It should be noted that large-scale livestock sheds are multi-layered, three-dimensional cage-raising models. The sheds typically have multiple rows of cages, and the aisles between the rows use inter-row duct fans to direct airflow into the cages through the ducts, rather than blowing air throughout the entire shed. This tunnel-like longitudinal ventilation improves airflow circulation within the shed.

[0075] In one specific embodiment of the present invention, based on the regional control demand index of the measured location, a new wind speed is determined between a preset base wind speed and a preset maximum wind speed using linear interpolation. The calculation formula is: New wind speed = Base wind speed + Regional control demand index × (Maximum wind speed - Base wind speed). Wherein, the preset base wind speed is the minimum wind speed required to maintain normal ventilation in the livestock shed, which can be determined according to the minimum ventilation volume standard for the livestock shed; the preset maximum wind speed is the wind speed corresponding to the rated maximum output of the inter-row duct fan in the measured location area.

[0076] When the regional regulation demand index is zero, the replacement wind speed equals the base wind speed, and the fan maintains basic ventilation with minimal energy consumption. When the regional regulation demand index increases, the replacement wind speed smoothly rises to the maximum wind speed to achieve on-demand regulation. When the temperature returns to normal and the regulation demand index falls back, the replacement wind speed synchronously falls back to the base wind speed, thus possessing bidirectional regulation capability, taking into account both regulation response speed and energy conservation.

[0077] Because the livestock shed adopts a "row-to-row duct fan + directional airflow" mode, the system first converts the refresh wind speed of each area into the target speed of the corresponding area's fan, and then sends an operating command to the row-to-row duct fan in that area, adjusting the fan at that measurement location to operate at the refresh wind speed. Subsequently, the fan speed of each area can be updated every two hours, thereby achieving intelligent control of the livestock shed environment.

[0078] In summary, this invention, after analyzing abnormal temperature periods at the measurement locations, further determines reference locations at these abnormal periods as a normal baseline for analyzing the impact of temperature interference. Based on drinking water information analysis, the persistence and occurrence of deviations in drinking behavior between the measurement locations and reference locations are used to determine actual influencing factors. This can be combined with livestock physiological responses to quantify the actual impact of temperature, analyzing and comparing drinking behavior with that under normal environmental conditions, avoiding the limitations of relying solely on temperature data. Combining the vertical distribution of cages and changes in influencing factors, the possibility of airflow obstruction by upper and lower cage layers is analyzed at each location. Combining the frequency of abnormalities at the measurement locations and the trend of influencing factors, the urgency of regulation at the measurement locations is analyzed, comprehensively obtaining regional regulation demand indicators. This adapts to the vertical temperature difference characteristics of multi-layer cage housing and considers the cumulative effect of impact, making regulation needs more reliable. Finally, based on the regional regulation demand indicators, fan regulation is implemented to achieve targeted temperature control, and the regulation results better meet actual needs. This invention determines the actual temperature impact by jointly analyzing livestock drinking behavior, and then adjusts the current regulation demand intensity based on the cage distribution and abnormal accumulation in multi-layer cage housing, making environmental regulation consider both livestock health and resource conservation, better meeting and aligning with ecological farming goals.

[0079] The present invention also provides an intelligent control system for the zoning of livestock housing environment in a large-scale farm, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the intelligent control method for the zoning of livestock housing environment in a large-scale farm as described above.

[0080] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0081] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for intelligent control of livestock house environment zoning in a large-scale farm, characterized in that, The method includes: Acquire temperature data at each measurement location in the livestock cage at each time, as well as drinking water information for all cages in the preset control area corresponding to the measurement location at each time. For each measurement location, abnormal time periods are determined based on the deviation of temperature data from expectations over time; temperature fluctuation factors are obtained based on the degree of sustained temperature deviation during each abnormal time period; and reference locations for the corresponding measurement locations are determined based on the temperature fluctuation factors for other measurement locations during each abnormal time period. For each abnormal period, based on the drinking information of all cage locations in the corresponding measurement location and the preset control area, the drinking behavior characteristics are analyzed and compared with the reference location to determine the actual influence factor of each location in the abnormal period. Based on the vertical position distribution and actual influencing factor changes of each cage position in the preset control area corresponding to each measurement position, combined with the distribution characteristics of abnormal periods in the time sequence of the measurement position, the changing trend of actual influencing factors and temperature fluctuation factor, the regional control demand index of each measurement position is obtained. Wind turbine regulation is based on regional regulation demand indicators determined by the location; The method for obtaining the regional regulation demand index includes: for any measurement location, obtaining the current temperature regulation urgency of the measurement location based on the interval between abnormal periods in time series, the growth trend of actual influencing factors during consecutive abnormal periods, and the temperature fluctuation factor; obtaining the ventilation demand of each cage location based on the number of cages vertically above each cage location in the preset regulation area corresponding to the measurement location, combined with the adjacent deviation value in the vertical direction of the cage location; calculating the product of the ventilation demand and the temperature regulation urgency of each cage location, and then normalizing the sum of all products to obtain the regional regulation demand index of the measurement location. The method for obtaining the urgency of temperature control includes: After obtaining the interval duration between every two adjacent abnormal time periods in the time series of the measurement location, the mean of all interval durations is calculated and negatively correlated to serve as the frequency of abnormalities at the measurement location. The minimum value of all temperature fluctuation factors in the time series of the measurement location is taken as the minimum fluctuation impact of the measurement location. All actual influencing factors of the measurement location are arranged in the order of distribution of abnormal time periods in the time series to obtain the impact sequence. The impact sequence is fitted with a straight line and the slope after fitting is taken as the improvement degree of the impact trend of the measurement location. The frequency of abnormalities, minimum fluctuation impact, and improvement degree of impact trend of the measurement location are weighted, summed, and normalized to obtain the urgency of temperature regulation at the measurement location.

2. The intelligent control method for zoning the livestock housing environment in a large-scale farm according to claim 1, characterized in that, The method for obtaining the abnormal time period includes: For any given measurement location, in the historical time series of that location, the moments when the temperature data is outside the expected temperature range are considered abnormal moments; and the time periods consisting of consecutively adjacent abnormal moments are considered abnormal time periods.

3. The intelligent control method for zoning the livestock housing environment in a large-scale farm according to claim 1, characterized in that, The method for obtaining the temperature fluctuation factor includes: For any abnormal time period at any measurement location, calculate the difference between the temperature data at each moment during the abnormal time period and the boundary value of the expected temperature range, and take the minimum value of the difference as the temperature deviation at each moment. Multiply the maximum temperature deviation during the abnormal period by the duration of the abnormal period and then normalize the result to obtain the temperature fluctuation factor for that abnormal period.

4. The intelligent control method for zoning the livestock housing environment in a large-scale farm according to claim 1, characterized in that, The method for obtaining the reference position includes: For any abnormal time period at any measurement location, for other measurement locations besides that measurement location, the product of the number of abnormal moments at the other measurement locations during the abnormal time period and the average value of the temperature fluctuation factor is calculated to obtain the abnormal probability index for each other measurement location; if there are no abnormal moments at any other measurement location, the abnormal probability index can be set to zero. Other measurement locations where the probability of anomalies is minimized are used as reference locations.

5. The intelligent control method for zoning the livestock housing environment in a large-scale farm according to claim 1, characterized in that, The methods for obtaining the actual impact factors include: For any abnormal time period at any measurement location, each cage position in the preset control area corresponding to the measurement location is sequentially taken as the analysis location; the drinking status of the analysis location is obtained from the drinking information at each moment during the abnormal time period; the moment when the drinking status of the analysis location is turned on is taken as the drinking time of the analysis location; each consecutive drinking time is taken as a drinking period of the analysis location during the abnormal time period. The proportion of drinking times during the abnormal period at the analysis location is used as the drinking occurrence indicator for the analysis location; the total number of drinking and non-drinking times at the analysis location during the abnormal period is negatively correlated and normalized to obtain the drinking continuity indicator for the analysis location. After calculating the absolute values ​​of the differences in drinking water occurrence indicators and drinking water sustainability indicators between the analysis location and the reference location, the mean of the two absolute values ​​of the differences is taken as the actual influencing factor of the analysis location during the abnormal period.

6. The intelligent control method for zoning the livestock housing environment in a large-scale farm according to claim 1, characterized in that, The method for obtaining the ventilation demand includes: For any cage position in the preset control area of ​​the measurement location, in the most recent abnormal period from the current time, calculate the absolute value of the difference between the actual influencing factor between the cage position and each adjacent cage position in the vertical direction, and take the mean of all absolute differences as the neighbor-to-neighbor deviation value of the cage position. The number of cages located vertically above the cage is normalized and used as the distribution influence value of the cage's location. The product of the distribution influence value of the cage location and the adjacent deviation value is taken as the ventilation demand of the cage location.

7. The intelligent control method for zoning livestock housing environment in a large-scale farm according to claim 1, characterized in that, The wind turbine regulation based on the regional regulation demand index of the measured location includes: Based on the regional control demand index at the measured location, a new wind speed is determined between the preset base wind speed and the preset maximum wind speed, so that the new wind speed is positively correlated with the regional control demand index; the wind turbine at the measured location is then adjusted to operate at the new wind speed.

8. A smart control system for zoning livestock housing environment in a large-scale livestock farm, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the intelligent control method for livestock housing environment zoning in a large-scale farm as described in any one of claims 1 to 7.

Images