WO2022166919A1 - 一种畜禽舍养殖环境温度预测控制***及其调控方法 - Google Patents

一种畜禽舍养殖环境温度预测控制***及其调控方法 Download PDF

Info

Publication number
WO2022166919A1
WO2022166919A1 PCT/CN2022/075157 CN2022075157W WO2022166919A1 WO 2022166919 A1 WO2022166919 A1 WO 2022166919A1 CN 2022075157 W CN2022075157 W CN 2022075157W WO 2022166919 A1 WO2022166919 A1 WO 2022166919A1
Authority
WO
WIPO (PCT)
Prior art keywords
temperature
livestock
prediction
model
control
Prior art date
Application number
PCT/CN2022/075157
Other languages
English (en)
French (fr)
Inventor
李保明
王阳
郑炜超
童勤
Original Assignee
中国农业大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国农业大学 filed Critical 中国农业大学
Priority to US18/264,747 priority Critical patent/US20240000046A1/en
Publication of WO2022166919A1 publication Critical patent/WO2022166919A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01K—ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
    • A01K29/00—Other apparatus for animal husbandry
    • A01K29/005—Monitoring or measuring activity, e.g. detecting heat or mating
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D27/00—Simultaneous control of variables covered by two or more of main groups G05D1/00 - G05D25/00
    • G05D27/02—Simultaneous control of variables covered by two or more of main groups G05D1/00 - G05D25/00 characterised by the use of electric means
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01K—ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
    • A01K1/00—Housing animals; Equipment therefor
    • A01K1/0047—Air-conditioning, e.g. ventilation, of animal housings
    • A01K1/0052—Arrangement of fans or blowers
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01K—ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
    • A01K1/00—Housing animals; Equipment therefor
    • A01K1/0047—Air-conditioning, e.g. ventilation, of animal housings
    • A01K1/007—Arrangement of curtain systems
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
    • Y02A40/70—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in livestock or poultry

Definitions

  • the invention relates to the technical field of agricultural environment control, in particular to a predictive control system and a control method for the breeding environment temperature of livestock and poultry houses.
  • a suitable breeding environment is a key factor affecting the health and production performance of livestock and poultry, and it is also the basis for ensuring their genetic potential and production efficiency.
  • Complex and changeable climatic conditions, building facilities of livestock and poultry houses, ventilation and air flow organization and feeding mode all directly affect the environment in the house, resulting in the complexity and diversity of the environment in the house, with uneven thermal environment, large temperature fluctuation, and thermal stress. Excessive stress and other factors all affect the health, production performance, disease resistance and feed conversion rate of livestock and poultry.
  • a large-scale livestock and poultry house environmental control system is the key to ensuring that the environmental conditions in the house meet the living environment of livestock and poultry.
  • the existing environmental control system of livestock and poultry houses mainly monitors the environmental conditions in the house by arranging sensors such as temperature and humidity in the house. Then, the operation and switching of the ventilation mode, the operation of the wet curtain cooling system and the heating device are regulated to regulate the small environment in the house through control strategies such as control algorithms.
  • control strategies such as control algorithms.
  • These control systems are mainly used in the environmental control system of livestock and poultry houses with control strategies such as PID control, fuzzy control, neural network and expert system control algorithms.
  • the internal environment of the poultry house is within a range that is relatively suitable for the growth of livestock and poultry, which reduces the labor intensity, improves the economic and social benefits of breeding, and promotes the development of the livestock and poultry breeding industry.
  • the control of the environmental parameters is based on the feedback environment regulation method based on the environmental parameter detection results.
  • the feedback environmental regulation method based on the environmental parameter detection results has the following problems:
  • the control temperature is a limited value, which is difficult to control accurately in real time, and cannot meet the needs of environmental temperature in different growth stages: the set value of the environmental temperature in the livestock and poultry house is generally a few or more specific temperature critical values, which cannot be Scientifically set the temperature value for the needs of different stages of livestock and poultry growth;
  • the climate environment is complex and changeable. When the temperature outside the house drops and increases sharply in a short period of time, monitoring the changes in the house environment and feedback adjustment cannot make control strategies before the environment suddenly changes, and the house environment cannot completely avoid the influence of the harsh environment. Makes the house a certain period of time in the heat and cold stress, temperature fluctuations and other phenomena.
  • the environmental regulation accuracy and strategy are limited by the sensors in the house, which are directly affected by the monitoring accuracy of the sensors in the house, and the abnormality of the sensor cannot be detected in time and continues to be the input module of the control system;
  • the invention discloses a livestock and poultry house breeding environment temperature prediction control system.
  • the control system comprises: a temperature and humidity sensor, a breeding environment temperature dynamic demand module, an environment controller and an environment regulation and control actuator; wherein, the environment controller is connected to the breeding environment respectively.
  • the environmental temperature dynamic demand module and the temperature and humidity sensor, the environmental regulation actuator are connected to the environmental controller, and the corresponding environmental regulation is performed according to the commands of the environmental controller.
  • the environmental controller includes a data acquisition module, an environmental temperature prediction module, a predictive fuzzy control module, a predictive fuzzy decision-making module and an early warning maintenance module.
  • the ambient temperature prediction module is connected to the data acquisition module, and the ambient temperature prediction module outputs the livestock and poultry house breeding based on a comprehensive GM(1,1) model of cumulative generation sequence, residual model correction and isodimensional innovation processing The predicted temperature of the environment at the Nth time;
  • the prediction fuzzy control module is connected to the ambient temperature prediction module, the input signal of the prediction fuzzy control module is the change rate of the prediction error and the prediction error of the breeding environment temperature, and a fuzzy decision rule is established to determine the output of the prediction fuzzy control.
  • the environmental control actuator performs control and adjustment according to the output of the predicted fuzzy control;
  • the predictive fuzzy decision-making module is connected to the predictive fuzzy control module, and the input signal of the predictive fuzzy decision-making module is the error of input and output and the rate of change of the error to determine an appropriate prediction step size;
  • the early warning maintenance module performs early warning based on the accumulated number of deviations, and the early warning maintenance module is connected to the aquaculture environment temperature dynamic demand module, the temperature and humidity sensor and the data acquisition module.
  • the breeding environment temperature dynamic demand module outputs the breeding environment temperature demand parameters of the livestock and poultry at different growth stages based on the livestock and poultry breeding characteristics, behavioral characteristics, stress mechanism, livestock and poultry species, quality and age parameters.
  • the ambient temperature dynamic demand module is connected to the data acquisition module.
  • the environmental regulation executive mechanism includes a fan, a wet curtain, a small ventilation window, a heating device and a spray device, and the environmental regulation executive mechanism controls the ventilation, cooling and heating of the livestock and poultry breeding environment according to the commands of the environmental controller. and humidification.
  • the number of the temperature and humidity sensors is greater than or equal to two, and they are respectively arranged inside and outside the livestock and poultry house.
  • the invention also discloses a control method of a livestock and poultry house breeding environment temperature prediction control system.
  • the prediction system and the gray prediction fuzzy system based on the GM(1,1) model control the temperature of the breeding environment.
  • the specific operation is carried out according to the following steps:
  • S1 Determine the required temperature T x of the environment in the livestock and poultry house at time N based on the dynamic demand model of the livestock and poultry breeding environment temperature; , The temperature parameters of the breeding environment are dynamically set for the quality and age parameter model;
  • S2 Predict the ambient temperature of the livestock and poultry house based on the GM(1,1) model, and obtain the predicted temperature T y of the livestock and poultry house environment at time N;
  • a system and method for predicting the environment in a livestock and poultry house integrating the GM(1,1) model first, the historical temperature data of the data acquisition module is used to accumulate and generate an accumulation sequence, and a GM(1,1) model is established and the model is solved to obtain the time Response function, and then use the GM(1,1) model to obtain the predicted value for residual test. Combined with the characteristics of the residual sequence change and the advantages of the residual GM(1,1) model, the model test is unqualified or the accuracy is low.
  • the residual GM(1,1) model is established to modify the original model to improve the accuracy of the model, and then the residual or accuracy test of the model is used to select the dimension of the prediction model. Updates predict the GM(1,1) model, and repeat the test and correction of the model;
  • the invention proposes a livestock and poultry house breeding environment temperature prediction control system and a control method thereof. Based on the dynamic demand model of livestock and poultry breeding environment temperature, the invention determines the required temperature of the environment in the livestock and poultry house at different stages; The temperature change in the poultry house has been mastered, and the temperature change law in the livestock and poultry house has been grasped. Based on a small amount of temperature data output during the acquisition process, the control module quickly, automatically and real-timely establishes a temperature prediction model, and processes the temperature data and other updates, and continuously changes the temperature from the new temperature. The data automatically updates the prediction model. During the prediction process, the temperature parameters change continuously, and the parameters of the controller change with time.
  • the residual model is used to correct the accuracy of the prediction model, and the actuator is regulated based on the gray prediction-fuzzy control system.
  • the output prediction value is obtained by fuzzy inference method, and the prediction error and the change rate of prediction error are used as input variables of the gray prediction-fuzzy control system.
  • Adaptability and regulation accuracy in addition, the present invention predicts the trend of the peak and valley points of the temperature in the livestock and poultry house, predicts the appearance time of the peak and valley points of the temperature in the livestock and poultry house in advance, and automatically enters the temperature according to the temperature difference between the demanded temperature and the measured temperature.
  • the breeding manager can overhaul the sensor and the environmental control execution equipment according to the early warning, which can prevent problems before they occur; the prediction function of the prediction fuzzy controller module of the present invention can achieve the prediction value in advance without increasing the overshoot.
  • the set value reduces the control amount in advance, avoids negative overshoot, improves the control effect, enhances the robustness, and reduces the energy consumption caused by overshoot.
  • the present invention is a predictive control system and a control method for the breeding environment temperature of livestock and poultry houses (especially closed livestock and poultry houses).
  • livestock and poultry houses especially closed livestock and poultry houses.
  • the stability and uniformity of the environment reduce the thermal stress and temperature fluctuation range, and reduce the environmental regulation and energy consumption of livestock and poultry houses.
  • the quality and age parameters of the breeding environment temperature dynamic demand model output the breeding environment temperature demand parameters of livestock and poultry in different growth stages, and determine the demand for environmental temperature in different growth stages;
  • the present invention seeks and grasps the temperature change law in the livestock and poultry house based on the temperature change in the livestock and poultry house.
  • Real-time modeling no need to manually establish a mathematical model for the temperature in the livestock and poultry house, establish a GM(1,N) model to predict the temperature value in the future, the modeling method is simple, and the temperature data and other maintenance information are processed without increasing Under the calculation amount, the information in the control system is kept "fresh" in real time, and the residual model is used to correct the accuracy of the prediction model.
  • the house control system regulates the temperature environment in the house in time;
  • the ambient temperature prediction module of the present invention processes the temperature data and other information, and adds a new temperature data, then a new prediction model is established, and the new temperature data is continuously updated.
  • the prediction model is automatically updated, and the temperature parameters change continuously during the prediction process, that is, the parameters of the controller change with time, which can take into account the dynamic and static performance of the system, and overcome the difficulty of parameter adjustment to meet the response speed and tracking accuracy.
  • Random nonlinear disturbances such as changes in temperature and human activities have self-adaptive ability
  • the model is corrected by the residual model, which improves the accuracy of the temperature prediction model of the livestock and poultry house
  • the system control of the prediction-fuzzy controller module of the present invention is prediction fuzzy Control, has the ability to adapt to various disturbances such as randomness, nonlinearity and uncertainty of the controlled environment and controlled parameters, and can achieve timely control and strong real-time performance;
  • the data acquisition module and the environmental temperature prediction module in the environmental controller of the present invention carry out trend prediction on the peak and valley points of the temperature in the livestock and poultry house, which overcomes the problem that the livestock and poultry houses are based on the detection results. Due to the limitations of the feedback environment regulation, it is possible to predict the peak and valley time of the temperature in the livestock and poultry house in advance, so as to prevent problems before they occur. And the prediction-fuzzy controller module and the early warning maintenance module, the system can automatically enter the accumulation mode. When the cumulative sum is greater than the specified value, the system enters the early warning. suffer in the future;
  • the prediction function of the prediction-fuzzy controller module of the present invention improves the response time without increasing the overshoot, forwards the response to the drop, and the predicted value reaches the preset value in advance.
  • the fixed value reduces the control amount in advance and avoids the negative overshoot.
  • the robustness of the gray prediction fuzzy control is stronger than that of the fuzzy control, which improves the control effect and reduces the energy consumption caused by the overshoot.
  • Fig. 1 is a structural schematic diagram of a poultry house breeding environment temperature prediction control system
  • Fig. 2 is the flow chart of the temperature prediction method in the livestock and poultry house based on the GM(1,1) model
  • Figure 3 is a schematic diagram of a temperature prediction and control system in a livestock house based on the GM(1,1) model
  • Figure 4 is a graph showing the comparison between the measured value and the predicted value of the temperature in the house
  • Fig. 6 is a graph of the temperature change graph in the predicted control house.
  • a system and method for predicting and controlling the environment in a livestock and poultry house based on the GM(1,N) grey model of the present invention includes the following main steps:
  • S1 Determine the required temperature of the environment in the livestock and poultry house at time N based on the dynamic demand model of livestock and poultry breeding environment.
  • Most modern large-scale high-density livestock and poultry breeding production adopts high-yield varieties, and livestock and poultry have high demand for the breeding environment, and the breeding environment The ability to adapt to stress such as temperature fluctuations is weak, and the breeding environment temperature at different stages of growth of livestock and poultry meets the needs, which is the basis for ensuring that high-yielding livestock and poultry breeds can exert their genetic potential and production efficiency.
  • the control target of the ambient temperature of the livestock and poultry house is set as the model of livestock and poultry breeding characteristics, behavioral characteristics, stress mechanism, livestock and poultry species, quality and age parameters, etc.
  • Environmental requirements T x regulating the temperature inside the house is the ideal temperature T x for livestock welfare and healthy growth.
  • the predicted temperature Ty of the environment in the livestock and poultry house at time N is predicted by the livestock and poultry house breeding environment temperature prediction system based on the GM(1,1) model;
  • Step (1) first use the weather forecast value of the location of the livestock and poultry house building of the environmental controller data acquisition module or the ambient temperature inside and outside the house collected by the temperature and humidity sensor as historical temperature data:
  • the weather forecast value of the location of the livestock and poultry house building of the data acquisition module of the environmental controller or the ambient temperature inside and outside the house collected by the temperature and humidity sensors inside and outside the house are used as the historical temperature as the data source, and the "starting point" time and the previous n equal time intervals are used.
  • the data constructs a time series as the initial operation data set;
  • the original temperature sequence T (0) is:
  • T (0) (T (0) (1), T (0) (2), T (0) (3)...T (0) (n));
  • Step (2) Accumulate historical temperature data to generate an accumulation sequence:
  • the historical temperature data is accumulated to generate the accumulated sequence, and the accumulated generation can reduce the volatility and randomness of the random sequence of thermal environment parameters, thereby improving the prediction accuracy of the model;
  • T (1) (T (1) (1), T (1) (2), T (1) (3)...T (1) (n)), where,
  • Step (3) establish a GM(1,1) model and solve the model to obtain a time response function, establish a whitening differential equation of the GM(1,1) model on the transformed data, and solve the model to obtain a time response function;
  • T (1) establish a first-order variable differential equation to form a temperature grey prediction model GM(1,1), where a and u are the parameters to be solved. Assume Use the least squares method to solve the parameters a, u, where B and Y are respectively:
  • Step (4) Use the GM(1,1) model to obtain the predicted value for residual test, and carry out the series residual test and reduction series test for the predicted temperature series. If the predicted value is within the accuracy range, when the simulation relative error and average When the relative error is less than 1%, the mean square error ratio is less than 0.35, and the probability of small error is greater than 0.95, it is considered that the temperature prediction model of the livestock and poultry house meets the accuracy requirements, and the predicted temperature sequence is output;
  • the relative error sequence is:
  • the variance ratio c of the posterior difference test is:
  • the small probability error value p is:
  • Step (5) Combine the characteristics of the variation of the residual sequence and the advantages of the residual GM(1,1) model to establish a residual GM(1,1) model for the sequences that fail the model test or have low precision. Correction to improve the accuracy of the model: If the model fails to pass the test, establish a residual correction model, take the residual sequence of the sequence with lower precision and sort it, and process the residual sequence of the sequence with lower precision 1-AGO to generate the cumulative sequence; Establish a residual GM(1,N) model, solve the model to obtain the time response sequence, obtain a new residual model and superimpose it into the revision of the original temperature prediction model, modify the original model to obtain a new temperature prediction model, and test the prediction accuracy of the model , if the accuracy requirements are met, the corrected temperature sequence is output, otherwise, the correction is continued until the predicted temperature sequence meets the inspection requirements.
  • Step (6) Use the residual or accuracy test of the model to select the dimension of the prediction model: Use the residual or accuracy test of the model to select the dimension m, and the smaller the mean absolute error, the better, the mean absolute error is: When the mean absolute error is the smallest, i.e. The value of m is the dimension of the prediction model.
  • Step (7) Do equal-dimensional innovation processing, build an equal-dimensional innovation prediction GM(1,1) model, and repeat the model inspection and correction: with the increase of the temperature sequence of the livestock and poultry house, the previous step is based on the prediction of the temperature prediction model. The temperature is sorted according to the time series, the operation data set is updated, the sequence is processed in equal dimensions, the equal dimension innovation sequence is obtained, and the equal dimension innovation GM(1,1) model is established; repeat the steps (3), (4) and (5) step process.
  • the corresponding time response model is:
  • the predicted output value of the system at step m is
  • Step (8) Repeat the process of step (7), adopt the recursive method, output the predicted temperature value at the mth time in turn, and use the equation get the predicted value of the original series
  • S4 Gray prediction-fuzzy control system based on GM(1,1) model, the predicted value output by the gray model is used to obtain the control amount by fuzzy inference, and the prediction error and the rate of change of the prediction error are used as the gray prediction-fuzzy control system input variable, prediction error and the rate of change of forecast error respectively Determine the advanced control amount of the system, and the controlled object performs advanced control to predict the error and the rate of change of forecast error and the control quantity ⁇ U(t i ) are standardized to the basic universe by the scale factor, and the corresponding fuzzy subsets are defined to establish the fuzzy decision-making rules. Fuzzy algorithm, the input and output membership functions of the fuzzy controller are all triangles, which are simple to calculate and occupy less space.
  • Adjacent fuzzy numbers intersect at the point where the membership angle is equal to 1/2, and there are at most two rules at a certain time. controller output.
  • the predicted value of the temperature series in the house can be obtained as:
  • the environmental control level and control strategy parameters of the case test house are shown in Table 1.
  • the changes of the measured temperature and predicted value of the test point of the test house are shown in Figure 4. It can be seen from the figure that the maximum difference between the measured value and the predicted value of the temperature in the chicken house is 0.5 °C, and the measured value of the temperature in the house is different from the predicted value. There is no significant difference between the values (P>0.05).
  • the percentage of temperature difference between the measured value and the predicted value is shown in Figure 5.
  • the percentage of difference between the measured temperature and the predicted value is 0-1.9%, and the change between the predicted value and the measured value of the temperature The trend is consistent, and the temperature prediction value can well represent the temperature change trend in the house.
  • the temperature gray predictive control strategy is used to control the indoor environment.
  • the maximum and minimum temperature differences between the measuring points at different positions in the house are 1°C and 0°C, respectively.
  • the system oscillation and overshoot are weakened under the predictive control.
  • Table 1 Test environmental control level and strategy parameter table

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Environmental Sciences (AREA)
  • Animal Husbandry (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Zoology (AREA)
  • Biophysics (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Housing For Livestock And Birds (AREA)

Abstract

一种畜禽舍养殖环境温度预测控制***及其调控方法，控制***包括：温湿度传感器、养殖环境温度动态需求模块、环境控制器和环境调控执行机构；其中，环境控制器分别连接养殖环境温度动态需求模块和温湿度传感器，环境调控执行机构连接环境控制器，并根据环境控制器的命令执行相应环境调控；调控方法根据畜禽养殖环境温度动态设定模型、基于GM(1,1)模型的畜禽舍养殖环境温度预测***和基于GM(1,1)模型的灰色预测模糊***控制养殖环境温度。解决了畜禽舍内温度难以及时调控、调控适应性差及不能做到防患于未然的问题，克服了畜禽舍基于检测结果的反馈式调控的局限。

Description

一种畜禽舍养殖环境温度预测控制***及其调控方法
本申请要求于2021年02月08日提交中国专利局、申请号为202110170467.0、发明名称为“一种畜禽舍养殖环境温度预测控制***及其调控方法”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及农业环境控制技术领域，特别是涉及畜禽舍养殖环境温度预测控制***及其调控方法。
背景技术
高密度、大规模舍饲环境下，适宜的养殖环境是影响畜禽健康和生产性能的关键因素，也是保障其发挥遗传潜力和生产效率的基础。复杂多变的气候条件、畜禽舍建筑设施、通风气流组织和饲养模式等都直接影响舍内环境，造成舍内环境的复杂和多样性，舍内热环境不均匀、温度波幅大、冷热应激等都影响畜禽机体的健康、生产性能、抗病能力和饲料转化率，规模化畜禽舍环境调控***是保障舍内环境条件满足畜禽生活环境的关键。
近年来，国内学者对畜禽舍小气候环境调控***提出了新的调控方法，如李立峰、武佩等的《基于组态软件和模糊控制的分娩母猪舍环境监控***》(2011年)，王新政、韩玉杰等的《畜禽舍温度模糊自适应PID控制与仿真研究》(2012年)，宣传忠、武佩等的《基于自适应模糊神经网络的畜禽舍环境控制***的研究》(2013年)，马从国、胡应占等的《国内猪舍小气候环境调控***研究进展》(2014年)，柴钰、于全刚的《基于模糊PID解耦算法的多功能禽舍环境控制***设计》(2014年)，马从国、胡应占等的《国内鸡舍小气候环境调控***研究》(2015年)，刘艳昌、张志霞等的《基于EPGA的畜禽舍环境模糊智能监控***》(2017年)。
现有的畜禽舍环境调控***主要通过在舍内布置温、湿度等传感器监测舍 内环境状况，用传感器监测到的舍内某点温度或几个点的平均温度信号与舍内设定值比较，然后通过控制算法等调控策略来调控通风模式的运行及切换、湿帘降温***和加热装置的运行来调控舍内小环境。这些调控***主要以PID控制、模糊控制、神经网络和专家***控制算法等调控策略应用于畜禽舍环控***中，这些畜禽舍环境调控***的应用促进了环境调控的自动化水平，使畜禽舍内环境在一个相对适宜畜禽生长的范围内，降低了人工劳动强度，提高了养殖的经济和社会效益，促进了畜禽养殖产业的发展。
然而，随畜禽饲养规模的不断增加，且高产品种畜禽对养殖环境温度波动等应激的适应能力弱，通过舍内定点传感器检测到的温湿度信号是否符合舍内环境要求来进行下一步的控制，均是基于环境参数检测结果的反馈式环境调控方法，基于环境参数检测结果的反馈式环境调控方法存在以下问题：
(1)调控温度为限定值，难以实时精准调控，不能满足不同生长阶段对环境温度的需求：畜禽舍内环境温度的设定值一般为几个或多个特定的温度临界值，不能按照畜禽生长不同阶段的需求科学设置温度值；
(2)难以及时调控舍内环境，存在滞后性，控制效率差。用传感器监测到的舍内该时刻的环境参数为指标，反馈调控下一时刻的舍内环境状况，研究表明，畜禽舍热环境***变化具有动态性、时变性和不确定性等特点，根据可测定的被控量温度信息确定控制量，对于畜禽舍环境滞后过程这一复杂对象，被控量的变化不能及时反应控制量的变化，环境调控策略与舍内环境状况相比具有明显的滞后性，无法根据被控量温度变化趋势的预测来相应的调整控制策略，控制效率差。舍内局部环境已经不在适宜畜禽生长的范围内，但环境调控***接收到的信号还在适宜环境范围内，环境调控***未能进行及时调控，致使舍内局部环境下的畜禽长时间在冷热应激环境下，易导致呼吸道疾病的发生；
(3)调控精度低。PID控制响应时间短但超调量大，畜禽舍环境温度控制***不允许有过大的超调量；对于模糊控制，阶跃上升响应虽超调量小，但响应时间较长，阶跃下降有负向超调，且模糊控制的精度尚不够高，只能实现畜禽舍环境的粗略调控；基于神经网络算法的自调控策略，对畜禽舍环境变化有适应性，但需要专家经验；
(4)不能做到防患于未然，适应性差，调控品质难以保证。气候环境复杂多变，舍外环境短时间内温度骤降、骤增下，监测舍内环境的变化反馈调节不能在环境骤变前做出调控策略，舍内环境无法完全避免恶劣环境的影响，使得舍内一定时间内冷热应激大、温度波幅大等现象。环境调控精度及策略受限于舍内传感器，受舍内传感器监测精度的直接影响，且传感器异常不能及时发现并持续作为控制***的输入模块；
(5)不利于节能，增大了运行能耗。畜禽舍内环境具有非线性特点，研究表明，基于环境参数检测结果的反馈式环境调控算法下，存在***震荡和超调现象，畜禽舍风机等调控设备超调运行，且环控设备启停次数多等问题，环境调控运行模式下势必增大了运行能耗。
畜禽舍外环境时刻变化，如何通过饲养环境调控技术实现舍内无冷热应激、温度波幅小、热环境均匀等目标，已经成为畜禽养殖产业需要亟待解决的一个重要的问题。
因此希望有一种畜禽舍养殖环境温度预测控制***及其调控方法能够解决现有技术中存在的问题。
发明内容
本发明公开了一种畜禽舍养殖环境温度预测控制***，所述控制***包括：温湿度传感器、养殖环境温度动态需求模块、环境控制器和环境调控执行机构；其中，环境控制器分别连接养殖环境温度动态需求模块和温湿度传感器，环境调控执行机构连接环境控制器，并根据环境控制器的命令执行相应环境调控。
优选地，所述环境控制器包括数据采集模块、环境温度预测模块、预测模糊控制模块、预测模糊决策模块和预警检修模块。
优选地，所述环境温度预测模块连接所述数据采集模块，所述环境温度预测模块基于累加生成数列、残差模型修正和等维新息处理的综合GM(1,1)模型输出畜禽舍养殖环境第N时刻的预测温度；
所述预测模糊控制模块连接所述环境温度预测模块，所述预测模糊控制模块的输入信号为养殖环境温度预测误差与预测误差的变化率，建立模糊决策规 则确定预测模糊控制的输出量，所述环境调控执行机构根据预测模糊控制的输出量进行控制调节；
所述预测模糊决策模块连接所述预测模糊控制模块，所述预测模糊决策模块的输入信号为输入输出的误差及其误差的变化率，确定适宜的预测步长；
所述预警检修模块基于偏差累加次数进行预警，所述预警检修模块连接所述养殖环境温度动态需求模块、所述温湿度传感器和所述数据采集模块。
优选地，所述养殖环境温度动态需求模块基于畜禽饲养特性、行为特性、应激机制、畜禽品种、质量与日龄参数，输出畜禽不同生长阶段的养殖环境温度需求参数，所述养殖环境温度动态需求模块连接所述数据采集模块。
优选地，所述环境调控执行机构包括风机、湿帘、通风小窗、加热设备和喷雾装置，所述环境调控执行机构根据所述环境控制器的命令控制畜禽养殖环境的通风、降温、加热和加湿。
优选地，所述温湿度传感器数量大于等于两个，且分别设置在畜禽舍内部和外部。
本发明还公开了一种畜禽舍养殖环境温度预测控制***的调控方法，所述调控方法根据畜禽养殖环境温度动态设定模型、基于GM(1,1)模型的畜禽舍养殖环境温度预测***和基于GM(1,1)模型的灰色预测模糊***控制养殖环境温度，具体运行按照如下步骤进行：
S1：基于畜禽养殖环境温度动态需求模型确定N时刻畜禽舍内环境的需求温度T x；畜禽养殖环境温度动态设定模型基于畜禽饲养特性、行为特性、应激机制、畜禽品种、质量与日龄参数模型的养殖环境温度动态设定温度参数；
S2：基于GM(1,1)模型对畜禽舍养殖环境温度进行预测，获取N时刻的畜禽舍环境预测温度T y；该模型基于累加生成数列、残差模型修正和等维新息处理的综合GM(1,1)模型的畜禽舍内环境预测***及方法：首先用所述数据采集模块的历史温度数据作累加生成累加序列，建立GM(1,1)模型并求解模型，得时间响应函数，其次采用GM(1,1)模型得到预测值进行残差检验，结合残差序列变化的特点及其残差GM(1,1)模型的优点，对模型检验不合格或精度较低的序列建立残差GM(1,1)模型对原模型修正，以提高模型的精度，再者采用模型的残差或精度检验来选择预测模型的维数，最后，作等维新息处理建等维新息 预测GM(1,1)模型，并重复模型的检验及修正；
S3：判断畜禽养殖环境的需求温度T x与预测温度T y是否相等，若T x＝T y，保持原环境调控执行机构的调控策略；若T x≠T y，基于GM(1,1)模型的灰色预测模糊控制***对执行机构进行调控，执行S4；
S4：将畜禽养殖环境预测模块输出的预测误差、预测误差的变化率作为所述预测模糊控制模块的输入量，输入输出的误差及其误差的变化率为所述预测模糊决策模块的输入信号，确定适宜的预测步长，建立模糊决策规则，建立基于GM(1,1)模型的灰色预测模糊控制模型确定最终模糊控制的输出值，根据预测模糊控制的输出量对所述环境调控装置进行控制，当T y<T x时，***自动进入加热模式；当T y>T x时，***自动进入湿帘降温模式；使得畜禽舍内温度满足畜禽温度要求T x＝T y，重复S1和S2步骤；
S5：判断需求温度T x与实测温度T c是否相等，若|T x-T c|>0.5℃，***自动进入累计模式，当累计合大于5次，***进入预警，饲养管理者根据预警对所述温湿度传感器和环境调控执行机构进行检修。
本发明提出了一种畜禽舍养殖环境温度预测控制***及其调控方法，本发明基于畜禽养殖环境温度动态需求模型，确定了不同阶段畜禽舍内环境的需求温度；并且本发明基于畜禽舍内温度变化，掌握了畜禽舍内温度变化规律，基于采集过程输出的少量温度数据由控制模块快速、自动、实时建立温度预测模型，并对温度数据等维新息处理，不断由新温度数据自动更新预测模型，预测过程中温度参数不断变化，控制器的参数随时间变化，利用残差模型进行修正预测模型的精度，并基于灰色预测-模糊控制***对执行机构进行调控，将灰色模型输出的预测值用模糊推理方式求得控制量，将预测误差和预测误差的变化率作为灰色预测-模糊控制***输入变量，调控***可实时、及时调控舍内温度环境，并提高了控制***的适应性及调控精度；此外，本发明对畜禽舍内温度的峰、谷点进行趋势预测，提前预知畜禽舍内温度的峰、谷点出现时间，并根据需求温度与实测温度温差自动进入累计模式，饲养管理者根据预警进行传感器、环控执行设备的检修，可做到防患于未然；本发明预测模糊控制器模块的预测功能，在不增加超调的情况下，预测值提前达到设定值，提前减小了控制量，避免了负向超调，改善了控制效果，鲁棒性增强，减少了超调造成的 能耗。
本发明是一种针对畜禽舍(特别是密闭式畜禽舍)养殖环境温度的预测控制***及其调控方法，该***和控制方法改善畜禽舍内的温度环境，提高畜禽舍内温度环境的稳定及均匀性，减少冷热应激及温度波动幅度，并降低畜禽舍的环境调控能耗。
1.针对畜禽舍内调控温度为限定值，难以实时精准调控，不能满足不同生长阶段对环境温度的需求问题，本发明提出基于畜禽饲养特性、行为特性、应激机制、畜禽品种、质量与日龄参数养殖环境温度动态需求模型，输出畜禽不同生长阶段的养殖环境温度需求参数，确定了不同生长阶段对环境温度的需求问题；
2.针对调控***难以及时调控舍内温度环境问题，本发明基于畜禽舍内温度变化，寻找并掌握畜禽舍内温度变化规律，基于采集过程输出的少量温度数据由控制模块快速、自动、实时建模，不需对畜禽舍内温度人工建立数学模型，建立GM(1,N)模型对未来时刻的温度值进行预测，建模手段简单，并对温度数据等维新息处理，不增加计算量下，实时保持控制***中信息“新鲜”性，利用残差模型进行修正预测模型的精度，适用于工况复杂、干扰频繁而难以精准数学建模的畜禽舍环境，可实现畜禽舍调控***及时调控舍内温度环境；
3.针对适应性差，调控精度低问题，调控品质难以保证问题，本发明环境温度预测模块中对温度数据等维新息处理，增加一个新温度数据，便建立一个新预测模型，不断由新温度数据自动更新预测模型，预测过程中温度参数不断变化，即控制器的参数随时间变化，能兼顾***的动态和静态性能，克服了满足响应速度和跟踪精度而面临的参数调节困难问题，对环境参数的变化、人为活动等随机的非线性扰动有自适应能力，模型利用残差模型进行修正，提高了畜禽舍温度预测模型的精度，且本发明预测-模糊控制器模块的***控制为预测模糊控制，对被控环境、被控参数的随机性、非线性、不确定性等各种扰动具有自适应能力能达到控制及时，实时性强；
4.针对不能做到防患于未然问题，本发明的环境控制器中数据采集模块和环境温度预测模块对畜禽舍内温度的峰、谷点进行趋势预测，克服了畜禽舍基于检测结果的反馈式环境调控的局限，提前预知畜禽舍内温度的峰、谷点出现 时间，可以做到防患于未然。且预测-模糊控制器模块及预警检修模块，***可自动进入累计模式，当累计合大于规定值时，***进入预警，饲养管理者根据预警进行传感器、环控执行设备的检修，进一步做到防患于未然；
5.针对能耗大，不利于节能问题，本发明预测-模糊控制器模块的预测功能，在不增加超调的情况下，提高了响应时间，对下降响应前向预测，预测值提前达到设定值，提前减小了控制量，避免了负向超调，灰色预测模糊控制的鲁棒性比模糊控制的鲁棒性增强，改善了控制效果，减少了超调造成的能耗。
附图说明
图1为禽舍养殖环境温度预测控制***的结构示意图；
图2为基于GM(1,1)模型的畜禽舍内温度预测方法流程图；
图3为基于GM(1,1)模型的畜禽舍内温度预测调控***示意图；
图4为舍内温度实测值与预测值对比曲线图；
图5舍内温度实测值与预测值间差异百分比图；
图6预测调控舍内温度变化图曲线图。
具体实施方式
为使本发明实施的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行更加详细的描述。在附图中，自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。所描述的实施例是本发明一部分实施例，而不是全部的实施例。下面通过参考附图描述的实施例是示例性的，旨在用于解释本发明，而不能理解为对本发明的限制。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
如图1-3所示，本发明的一种基于GM(1,N)灰色模型的畜禽舍内环境预测调控***及方法，包含以下的主要步骤：
S1：基于畜禽养殖环境温度动态需求模型确定N时刻畜禽舍内环境的需求温度，现代规模化高密度畜禽养殖生产大多采用高产品种，畜禽对养殖环境 的需求高，且对养殖环境温度波动等应激的适应能力较弱，畜禽生长不同阶段的养殖环境温度满足需求，是保障高产性能畜禽品种发挥遗传潜力和生产效率的基础。本***中，将畜禽舍环境温度的控制目标设置为畜禽饲养特性、行为特性、应激机制、畜禽品种、质量与日龄参数等的模型，动态设定畜禽不同生长阶段对温度环境的需求T x，调控舍内温度为畜禽福利和健康生长的理想温度T x。
S2：基于GM(1,1)模型的畜禽舍养殖环境温度预测***预测N时刻畜禽舍内环境的预测温度T y；
步骤(1)：首先用环境控制器数据采集模块的畜禽舍建筑所在地天气预报值或采用舍内外温湿度传感器采集的舍内外环境温度为历史温度数据：
将环境控制器数据采集模块的畜禽舍建筑所在地天气预报值或采用舍内外温湿度传感器采集的舍内外环境温度为历史温度作为数据源，使用“起点”时刻及之前的n个等时间间隔的数据构造时间序列，作为初始运算数据集合；
对畜禽舍建筑所在地天气预报值或采用舍内外温湿度传感器采集的舍内外环境温度等时矩取样，原始温度序列T (0)为：
T (0)＝(T (0)(1),T (0)(2),T (0)(3)……T (0)(n))；
步骤(2)：历史温度数据作累加生成累加序列：
将历史温度数据作累加生成累加序列，累加生成以减弱热环境参数随机序列的波动性和随机性，从而提高模型的预测精度；
对温度序列作1-AGO，以减弱随机干扰的影响，得新数据序列T (1)：T (1)＝(T (1)(1),T (1)(2),T (1)(3)……T (1)(n))，式中，
Figure PCTCN2022075157-appb-000001
步骤(3)：建立GM(1,1)模型并求解模型，得时间响应函数，将经过变换的数据建立GM(1,1)模型的白化形式微分方程，求解模型得时间响应函数；
T (1)建立一阶变量微分方程构成温度灰色预测模型GM(1,1)，
Figure PCTCN2022075157-appb-000002
Figure PCTCN2022075157-appb-000003
其中a，u为待解参数。设
Figure PCTCN2022075157-appb-000004
利用最小二乘法解出参数a，u，
Figure PCTCN2022075157-appb-000005
其中B和Y分别为：
Figure PCTCN2022075157-appb-000006
Y＝[T (0)(2),T (0)(3),T (0)(4),……,T (0)(m)] T
取T (1)(0)＝T (0)(1)，当序列T (1)的变化是平滑的，GM(1,1)模型的时间响应序列为：
Figure PCTCN2022075157-appb-000007
Figure PCTCN2022075157-appb-000008
得数据列
Figure PCTCN2022075157-appb-000009
步骤(4)：采用GM(1,1)模型得到预测值进行残差检验，对预测温度序列进行数列残差检验及还原数列检验，若预测值在精度范围内，当满足模拟相对误差及平均相对误差小于1％、均方差比值小于0.35及小误差概率大于0.95时，则认为畜禽舍温度预测模型满足精度要求，则输出预测温度序列；
畜禽舍内原始温度数据列T (0)(k)，预测温度序列
Figure PCTCN2022075157-appb-000010
T (0)(k)与
Figure PCTCN2022075157-appb-000011
之差为残差序列ε (0)(n)，
Figure PCTCN2022075157-appb-000012
Figure PCTCN2022075157-appb-000013
相对误差序列为：
Figure PCTCN2022075157-appb-000014
当k≤n时，k点的模拟相对误差为：
Figure PCTCN2022075157-appb-000015
k点的平均相对误差为：
Figure PCTCN2022075157-appb-000016
原始数据X (0)的均值
Figure PCTCN2022075157-appb-000017
为：
Figure PCTCN2022075157-appb-000018
原始数据X (0)的方差
Figure PCTCN2022075157-appb-000019
Figure PCTCN2022075157-appb-000020
残差数列ε (0)(n)的均值
Figure PCTCN2022075157-appb-000021
为：
Figure PCTCN2022075157-appb-000022
残差数列ε (0)(n)的方 差
Figure PCTCN2022075157-appb-000023
为：
Figure PCTCN2022075157-appb-000024
后验差检验的方差比值c为：
Figure PCTCN2022075157-appb-000025
小概率误差值p为：
Figure PCTCN2022075157-appb-000026
根据计算的4个指标值Δδ k、
Figure PCTCN2022075157-appb-000027
c、p进行精度检验。
步骤(5)：结合残差序列变化的特点及其残差GM(1,1)模型的优点，对模型检验不合格或精度较低的序列建立残差GM(1,1)模型对原模型修正，以提高模型的精度：若模型检验不合格，建立残差修正模型，取精度较低数列的残差序列并排序，将精度较低数列的残差序列1-AGO处理，生成累加序列；建立残差GM(1,N)模型，求解模型得时间响应序列，得新残差模型并叠加入到原温度预测模型的修正中，对原模型进行修正得温度预测新模型，检验模型预测精度，若满足精度要求则输出修正后的温度序列，否则继续进行修正，直到预测温度序列满足检验要求。
当建立的畜禽舍温度预测模型GM(1,1)精度不符合舍内环控要求时，需对原模型进行修正，用残差序列建模以提高预测模型GM(1,1)的精度。
残差数列
Figure PCTCN2022075157-appb-000028
取精度较低数列的残差序列并排序，并作1-AGO处理，得序列ε (1)(n)，ε (1)建立一阶变量微分方程构成残差灰色预测模型GM(1,1)，
Figure PCTCN2022075157-appb-000029
其中a'，u'为待解参数。设
Figure PCTCN2022075157-appb-000030
利用最小二乘法解出参数a'，u'，
Figure PCTCN2022075157-appb-000031
其中B和Y分别为：
Figure PCTCN2022075157-appb-000032
Y＝[ε (0)(2)，ε (0)(3),ε (0)(4),……,ε (0)(m)] T
残差模型的有时间响应函数为
Figure PCTCN2022075157-appb-000033
为便于表示，时间响应函数改写为
Figure PCTCN2022075157-appb-000034
对畜禽舍温度预测模型的时间响应函数
Figure PCTCN2022075157-appb-000035
的倒数修正，当
Figure PCTCN2022075157-appb-000036
时，
Figure PCTCN2022075157-appb-000037
Figure PCTCN2022075157-appb-000038
将残差GM(1,N)模型
Figure PCTCN2022075157-appb-000039
加入到原预测模型中修正，有
Figure PCTCN2022075157-appb-000040
其中
Figure PCTCN2022075157-appb-000041
步骤(6)：采用模型的残差或精度检验来选择预测模型的维数：采用模型的残差或精度检验来选择维数m，且平均绝对误差越小越好，平均绝对误差为：
Figure PCTCN2022075157-appb-000042
平均绝对误差最小时，即
Figure PCTCN2022075157-appb-000043
m的取值为预测模型的维数。
步骤(7)：作等维新息处理建等维新息预测GM(1,1)模型，并重复模型的检验及修正：随畜禽舍温度序列的递增，将上一步骤根据温度预测模型的预测温度按时间序列排序，更新运算数据集合，对序列作等维处理，得等维新息数列，建立等维新息GM(1,1)模型；重复第(3)、(4)和(5)的步骤过程。
畜禽舍温度灰色预测GM(1,1)模型中，T (0)＝(T (0)(1),T (0)(2),T (0)(3)，……，T (0)(n-1)，T (0)(n))作等维新息处理，去掉T (0)(1)，增加T (0)(n+1)后，得：T (0)＝(T (0)(2),T (0)(3)，……，T (0)(n-1)，T (0)(n)，T (0)(n+1))。
时刻t i输出序列分别为T i为，T i＝(T i(1)，T i(2)，T i(3)，……，T i(n))，时刻t i+1输出序列为T i+1，T i+1＝(T i+1(1)，T i+1(2)，T i+1(3)，……，T i+1(n))；且保持T i+1(k)＝T i(k+1)，T i+1(n-1)＝T i(n)。
对等维新息处理后的数据序列作1-AGO，累加生成新序列，建立一阶变量微分方程构成温度灰色预测模型GM(1,1)，
Figure PCTCN2022075157-appb-000044
其中a i+1，u i+1为待解参数。设
Figure PCTCN2022075157-appb-000045
利用最小二乘法解出参数a i+1，u i+1，
Figure PCTCN2022075157-appb-000046
其中B i+1和Y i+1,N分别为：
Figure PCTCN2022075157-appb-000047
Figure PCTCN2022075157-appb-000048
Figure PCTCN2022075157-appb-000049
对应的时间响应模型为：
Figure PCTCN2022075157-appb-000050
Figure PCTCN2022075157-appb-000051
***m步的预测输出值为
Figure PCTCN2022075157-appb-000052
Figure PCTCN2022075157-appb-000053
步骤(8)：重复第(7)的步骤过程，采用递推的方法，依次输出第m时刻的预测温度值，运用方程
Figure PCTCN2022075157-appb-000054
得原始序列的预测值
Figure PCTCN2022075157-appb-000055
S3：判断畜禽养殖环境的需求温度T x与预测温度T y是否相等，若T x＝T y，保持原环境调控执行机构的调控策略；若T x≠T y，基于GM(1,1)模型的灰色预测-模糊控制***对执行机构进行调控；
S4：基于GM(1,1)模型的灰色预测-模糊控制***，将灰色模型输出的预测值用模糊推理方式求得控制量，将预测误差和预测误差的变化率作为灰色预测-模糊控制***输入变量，预测误差
Figure PCTCN2022075157-appb-000056
和预测误差的变化率
Figure PCTCN2022075157-appb-000057
分别为
Figure PCTCN2022075157-appb-000058
Figure PCTCN2022075157-appb-000059
确定***的超前控制量，被控对象进行超前控制，将预测误差
Figure PCTCN2022075157-appb-000060
和预测误差的变化率
Figure PCTCN2022075157-appb-000061
和控制量ΔU(t i)经比例因 子规范至基本论域，并定义相应的模糊子集，建立模糊决策规则，并对控制规则采用Mamdani推理方法，确定规则的适应度，并采用加权平均去模糊算法，模糊控制器的输入、输出的隶属函数形状均采用三角形，计算简单，占用空间少，相邻模糊数在隶属角等于1/2处交叉，则在某一时刻最多只有两条规则决定控制器的输出。预测误差、误差变化和控制量的量化等级分为7级，确定适宜的预测步长，确定最终模糊控制的输出值，控制相应通风模式的运行及切换、湿帘降温***、加热装置和侧墙小窗等装置，使得畜禽舍内温度T x＝T y＝T c满足要求。
S5：判断需求温度T x与实测温度T c是否相等，若|T x-T c|>0.5℃，***自动进入累计模式，当累计合大于5次，***进入预警，饲养管理者根据预警进行传感器、环控执行设备的检修。
案例：
选山东日照市某养殖场的舍内环境温度预测调控案例进行分析，不同维数下舍内温度的预测值和误差表如表1所示，本案例研究中选取温度灰色预测模型的维数为7，得GM(1,1)模型的时间响应序列白化形式微分方程为：
Figure PCTCN2022075157-appb-000062
可得鸡舍内温度序列的预测值为：
Figure PCTCN2022075157-appb-000063
Figure PCTCN2022075157-appb-000064
案例试验舍的环控等级及调控策略参数如表1所示。利用预测值
Figure PCTCN2022075157-appb-000065
参与***控制决策。试验舍测点的温度实测值、预测值变化如图4所示，从图中可看出，鸡舍内温度实测值与预测值间最大差值为0.5℃，且舍内温度实测值与预测值间无显著性差异(P＞0.05)，实测值与预测值间温度差异百分比如图5所示，温度实测值与预测值间差异百分比为0～1.9％，温度预测值与实测值间变化趋势一致，且温度预测值可很好的表示舍内温度变化趋势。如图6所示，用温度灰色预测控制策略调控舍内环境，舍内不同位置测点间最大、最小温差分别为1℃、0℃，预测调控下***震荡和超调现象减弱。
表1试验环控等级及策略参数表
预测温度(℃) 级别 风量(×10 3m 3/h)
19.5±0.5 D1 70
20.1±0.6 D2 105
20.8±0.7 D3 140
21.6±0.8 D4 175
22.1±1.0 D5 210
24.1±1.5 D6 280
25.6±1.5 D7 350
27.1±1.5 D8 420
29.1±2.0 D9 560
31.4±2.3 D10 700
33.8±2.4 D11 910
36.8±3.0 D12 1015
最后需要指出的是：以上实施例仅用以说明本发明的技术方案，而非对其限制。尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。

Claims (7)

  1. 一种畜禽舍养殖环境温度预测控制***，其特征在于，所述控制***包括：温湿度传感器、养殖环境温度动态需求模块、环境控制器和环境调控执行机构；其中，环境控制器分别连接养殖环境温度动态需求模块和温湿度传感器，环境调控执行机构连接环境控制器，并根据环境控制器的命令执行相应环境调控。
  2. 根据权利要求1所述的畜禽舍养殖环境温度预测控制***，其特征在于：所述环境控制器包括数据采集模块、环境温度预测模块、预测模糊控制模块、预测模糊决策模块和预警检修模块。
  3. 根据权利要求2所述的畜禽舍养殖环境温度预测控制***，其特征在于：所述环境温度预测模块连接所述数据采集模块，所述环境温度预测模块基于累加生成数列、残差模型修正和等维新息处理的综合GM(1,1)模型输出畜禽舍养殖环境第N时刻的预测温度；
    所述预测模糊控制模块连接所述环境温度预测模块，所述预测模糊控制模块的输入信号为养殖环境温度预测误差与预测误差的变化率，建立模糊决策规则确定预测模糊控制的输出量，所述环境调控执行机构根据预测模糊控制的输出量进行控制调节；
    所述预测模糊决策模块连接所述预测模糊控制模块，所述预测模糊决策模块的输入信号为输入输出的误差及其误差的变化率，确定适宜的预测步长；
    所述预警检修模块基于偏差累加次数进行预警，所述预警检修模块连接所述养殖环境温度动态需求模块、所述温湿度传感器和所述数据采集模块。
  4. 根据权利要求2所述的畜禽舍养殖环境温度预测控制***，其特征在于：所述养殖环境温度动态需求模块基于畜禽饲养特性、行为特性、应激机制、畜禽品种、质量与日龄参数，输出畜禽不同生长阶段的养殖环境温度需求参数，所述养殖环境温度动态需求模块连接所述数据采集模块。
  5. 根据权利要求1所述的畜禽舍养殖环境温度预测控制***，其特征在于：所述环境调控执行机构包括风机、湿帘、通风小窗、加热设备和喷雾装置，所述环境调控执行机构根据所述环境控制器的命令控制畜禽养殖环境的通风、降温、加热和加湿。
  6. 根据权利要求1所述的畜禽舍养殖环境温度预测控制***，其特征在于：所述温湿度传感器数量大于等于两个，且分别设置在畜禽舍内部和外部。
  7. 根据权利要求2所述的畜禽舍养殖环境温度预测控制***的调控方法，其特征在于：所述调控方法根据畜禽养殖环境温度动态设定模型、基于GM(1,1)模型的畜禽舍养殖环境温度预测***和基于GM(1,1)模型的灰色预测模糊***控制养殖环境温度，具体运行按照如下步骤进行：
    S1：基于畜禽养殖环境温度动态需求模型确定N时刻畜禽舍内环境的需求温度T x；畜禽养殖环境温度动态设定模型基于畜禽饲养特性、行为特性、应激机制、畜禽品种、质量与日龄参数模型的养殖环境温度动态设定温度参数；
    S2：基于GM(1,1)模型对畜禽舍养殖环境温度进行预测，获取N时刻的畜禽舍环境预测温度T y；该模型基于累加生成数列、残差模型修正和等维新息处理的综合GM(1,1)模型的畜禽舍内环境预测***及方法：首先用所述数据采集模块的历史温度数据作累加生成累加序列，建立GM(1,1)模型并求解模型，得时间响应函数，其次采用GM(1,1)模型得到预测值进行残差检验，结合残差序列变化的特点及其残差GM(1,1)模型的优点，对模型检验不合格或精度较低的序列建立残差GM(1,1)模型对原模型修正，以提高模型的精度，再者采用模型的残差或精度检验来选择预测模型的维数，最后，作等维新息处理建等维新息预测GM(1,1)模型，并重复模型的检验及修正；
    S3：判断畜禽养殖环境的需求温度T x与预测温度T y是否相等，若T x＝T y，保持原环境调控执行机构的调控策略；若T x≠T y，基于GM(1,1)模型的灰色预测模糊控制***对执行机构进行调控，执行S4；
    S4：将畜禽养殖环境预测模块输出的预测误差、预测误差的变化率作为所述预测模糊控制模块的输入量，输入输出的误差及其误差的变化率为所述预测模糊决策模块的输入信号，确定适宜的预测步长，建立模糊决策规则，建立基于GM(1,1)模型的灰色预测模糊控制模型确定最终模糊控制的输出值，根据预测模糊控制的输出量对所述环境调控装置进行控制，当T y<T x时，***自动进入加热模式；当T y>T x时，***自动进入湿帘降温模式；使得畜禽舍内温度满足畜禽温度要求T x＝T y，重复S1和S2步骤；
    S5：判断需求温度T x与实测温度T c是否相等，若|T x-T c|>0.5℃，*** 自动进入累计模式，当累计合大于5次，***进入预警，饲养管理者根据预警对所述温湿度传感器和环境调控执行机构进行检修。
PCT/CN2022/075157 2021-02-08 2022-01-30 一种畜禽舍养殖环境温度预测控制***及其调控方法 WO2022166919A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/264,747 US20240000046A1 (en) 2021-02-08 2022-01-30 Predictive control system and regulatory method for temperature of livestock house

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110170467.0 2021-02-08
CN202110170467.0A CN112817354B (zh) 2021-02-08 2021-02-08 一种畜禽舍养殖环境温度预测控制***及其调控方法

Publications (1)

Publication Number Publication Date
WO2022166919A1 true WO2022166919A1 (zh) 2022-08-11

Family

ID=75863730

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/075157 WO2022166919A1 (zh) 2021-02-08 2022-01-30 一种畜禽舍养殖环境温度预测控制***及其调控方法

Country Status (3)

Country Link
US (1) US20240000046A1 (zh)
CN (1) CN112817354B (zh)
WO (1) WO2022166919A1 (zh)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115309203A (zh) * 2022-10-11 2022-11-08 山东华邦农牧机械股份有限公司 养殖用智能温度控制***
CN115830757A (zh) * 2022-12-02 2023-03-21 江苏锦花电子股份有限公司 一种基于大数据的显示设备性能监测***及方法
CN115856232A (zh) * 2023-02-28 2023-03-28 北京市农林科学院信息技术研究中心 畜禽舍臭气浓度确定方法、装置、电子设备及存储介质
CN116029435A (zh) * 2023-01-10 2023-04-28 盐城工学院 一种生猪设施养殖环境舒适度预警***
CN116484226A (zh) * 2023-05-04 2023-07-25 乳山市海洋经济发展中心 一种通过神经网络建立渔业养殖模型的方法及装置
CN116720713A (zh) * 2023-08-07 2023-09-08 广州城建职业学院 一种用于畜牧业禽类养殖管理方法、***及存储介质
CN116774761A (zh) * 2023-08-22 2023-09-19 山东华邦农牧机械股份有限公司 一种养殖用设备的节能优化方法
CN117077288A (zh) * 2023-08-21 2023-11-17 浙江大学 一种航空飞行器元件的适应性设计分析方法及***
CN117270403A (zh) * 2023-11-22 2023-12-22 四川中物技术股份有限公司 一种龙门式制坨机的优化控制方法
CN117970978A (zh) * 2024-03-29 2024-05-03 中发建筑技术集团有限公司 一种适用于实验的恒温恒湿实验箱自动调控***

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112817354B (zh) * 2021-02-08 2023-06-30 中国农业大学 一种畜禽舍养殖环境温度预测控制***及其调控方法
CN113703504B (zh) * 2021-09-26 2022-12-30 中国热带农业科学院湛江实验站 一种基于物联网的畜牧养殖用温湿度控制***及方法
CN114018322A (zh) * 2021-10-26 2022-02-08 中国农业大学 一种小型动物产热产湿测量方法及测量设备
CN113867451B (zh) * 2021-10-29 2022-08-23 江苏大学 一种云平台与控制器相结合的畜禽养殖智能温控方法
CN114200987A (zh) * 2021-12-03 2022-03-18 天津农学院 一种基于强化学习的密闭式畜禽舍环境智能调控方法
CN114115403A (zh) * 2021-12-03 2022-03-01 天津农学院 一种笼养肉鸡养殖环境智能控制装置
CN114371752B (zh) * 2022-03-22 2022-05-24 广东博创佳禾科技有限公司 一种精密仪器实验室的环境调控方法及***
CN115454177B (zh) * 2022-09-16 2023-05-23 江苏省家禽科学研究所 一种禽用通风降温湿帘的自动控制方法及***
CN116400600B (zh) * 2023-04-23 2023-11-03 重庆市畜牧科学院 一种基于智能全局优化的猪场通风动态调控***
CN116384979B (zh) * 2023-04-27 2023-09-26 圣麦克思智能科技(江苏)有限公司 一种idc运维服务支持***及其方法
CN117369572B (zh) * 2023-12-07 2024-03-15 广东凯得智能科技股份有限公司 一种用于恒温酒柜的智能加湿控制方法及其相关设备
CN117970986B (zh) * 2024-04-01 2024-06-14 广东热矩智能科技有限公司 一种冷热***的温湿度控制方法、装置及介质
CN118013448B (zh) * 2024-04-10 2024-06-21 湖南共农生物科技有限公司 基于物联网的数字化农业饲养数据采集方法及***
CN118112937B (zh) * 2024-04-29 2024-07-02 南京亥客技术有限公司 基于多源融合感知的养殖场微正压智能控制***和方法

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1996009576A1 (en) * 1994-09-19 1996-03-28 Georgia Tech Research Corporation Poultry environmental control systems and methods
CN204536850U (zh) * 2015-04-03 2015-08-05 陈恩保 养殖场环境监测控制***
CN104950948A (zh) * 2015-05-21 2015-09-30 淮阴工学院 牛舍温度智能控制***
US20160120144A1 (en) * 2014-10-30 2016-05-05 Electronics And Telecommunications Research Institute Livestock house management system and management method thereof
CN107203239A (zh) * 2017-07-31 2017-09-26 东北农业大学 一种密闭畜禽养殖环境解耦模糊调控方法
CN107289998A (zh) * 2017-05-12 2017-10-24 淮阴工学院 基于can总线的猪舍环境温度智能监测***
CN107491124A (zh) * 2017-08-29 2017-12-19 南阳华诚智能有限公司 物联网智能猪舍控制***
CN108445747A (zh) * 2018-01-30 2018-08-24 中国农业大学 一种畜禽养殖环境参数模糊优化控制方法
CN112817354A (zh) * 2021-02-08 2021-05-18 中国农业大学 一种畜禽舍养殖环境温度预测控制***及其调控方法

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4700887A (en) * 1986-12-29 1987-10-20 Cornell Research Foundation, Inc. Environmental control system for poultry houses
CN101480143B (zh) * 2009-01-21 2010-11-03 华中科技大学 一种预测灌区作物单产量的方法
CN107644276A (zh) * 2017-10-27 2018-01-30 常州工学院 基于二次项拟合gm(1，1)的社会总抚养比预测模型
CN110140662A (zh) * 2019-04-30 2019-08-20 青岛沃蒙斯畜牧设备有限公司 一种实现温度自动控制的畜禽养殖业专用智能发热板

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1996009576A1 (en) * 1994-09-19 1996-03-28 Georgia Tech Research Corporation Poultry environmental control systems and methods
US20160120144A1 (en) * 2014-10-30 2016-05-05 Electronics And Telecommunications Research Institute Livestock house management system and management method thereof
CN204536850U (zh) * 2015-04-03 2015-08-05 陈恩保 养殖场环境监测控制***
CN104950948A (zh) * 2015-05-21 2015-09-30 淮阴工学院 牛舍温度智能控制***
CN107289998A (zh) * 2017-05-12 2017-10-24 淮阴工学院 基于can总线的猪舍环境温度智能监测***
CN107203239A (zh) * 2017-07-31 2017-09-26 东北农业大学 一种密闭畜禽养殖环境解耦模糊调控方法
CN107491124A (zh) * 2017-08-29 2017-12-19 南阳华诚智能有限公司 物联网智能猪舍控制***
CN108445747A (zh) * 2018-01-30 2018-08-24 中国农业大学 一种畜禽养殖环境参数模糊优化控制方法
CN112817354A (zh) * 2021-02-08 2021-05-18 中国农业大学 一种畜禽舍养殖环境温度预测控制***及其调控方法

Cited By (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115309203B (zh) * 2022-10-11 2023-03-14 山东华邦农牧机械股份有限公司 养殖用智能温度控制***
CN115309203A (zh) * 2022-10-11 2022-11-08 山东华邦农牧机械股份有限公司 养殖用智能温度控制***
CN115830757A (zh) * 2022-12-02 2023-03-21 江苏锦花电子股份有限公司 一种基于大数据的显示设备性能监测***及方法
CN115830757B (zh) * 2022-12-02 2023-11-17 江苏锦花电子股份有限公司 一种基于大数据的显示设备性能监测***及方法
CN116029435A (zh) * 2023-01-10 2023-04-28 盐城工学院 一种生猪设施养殖环境舒适度预警***
CN116029435B (zh) * 2023-01-10 2023-07-07 盐城工学院 一种生猪设施养殖环境舒适度预警***
CN115856232B (zh) * 2023-02-28 2023-09-12 北京市农林科学院信息技术研究中心 畜禽舍臭气浓度确定方法、装置、电子设备及存储介质
CN115856232A (zh) * 2023-02-28 2023-03-28 北京市农林科学院信息技术研究中心 畜禽舍臭气浓度确定方法、装置、电子设备及存储介质
CN116484226B (zh) * 2023-05-04 2023-10-03 乳山市海洋经济发展中心 一种通过神经网络建立渔业养殖模型的方法及装置
CN116484226A (zh) * 2023-05-04 2023-07-25 乳山市海洋经济发展中心 一种通过神经网络建立渔业养殖模型的方法及装置
CN116720713A (zh) * 2023-08-07 2023-09-08 广州城建职业学院 一种用于畜牧业禽类养殖管理方法、***及存储介质
CN116720713B (zh) * 2023-08-07 2023-12-19 广州城建职业学院 一种用于畜牧业禽类养殖管理方法、***及存储介质
CN117077288A (zh) * 2023-08-21 2023-11-17 浙江大学 一种航空飞行器元件的适应性设计分析方法及***
CN117077288B (zh) * 2023-08-21 2024-03-29 浙江大学 一种航空飞行器元件的适应性设计分析方法及***
CN116774761A (zh) * 2023-08-22 2023-09-19 山东华邦农牧机械股份有限公司 一种养殖用设备的节能优化方法
CN116774761B (zh) * 2023-08-22 2023-11-14 山东华邦农牧机械股份有限公司 一种养殖用设备的节能优化方法
CN117270403A (zh) * 2023-11-22 2023-12-22 四川中物技术股份有限公司 一种龙门式制坨机的优化控制方法
CN117270403B (zh) * 2023-11-22 2024-01-26 四川中物技术股份有限公司 一种龙门式制坨机的优化控制方法
CN117970978A (zh) * 2024-03-29 2024-05-03 中发建筑技术集团有限公司 一种适用于实验的恒温恒湿实验箱自动调控***
CN117970978B (zh) * 2024-03-29 2024-06-04 中发建筑技术集团有限公司 一种适用于实验的恒温恒湿实验箱自动调控***

Also Published As

Publication number Publication date
CN112817354A (zh) 2021-05-18
CN112817354B (zh) 2023-06-30
US20240000046A1 (en) 2024-01-04

Similar Documents

Publication Publication Date Title
WO2022166919A1 (zh) 一种畜禽舍养殖环境温度预测控制***及其调控方法
CN108386971B (zh) 中央空调节能自控***
CN111580382B (zh) 基于人工智能的单元级供热调节方法及***
CN109634098A (zh) 一种育肥猪舍环境调控***及方法
CN112418498B (zh) 用于智能温室的温度预测方法及***
CN112527037B (zh) 一种具有环境因子预测功能的温室环境调控方法和***
CN103912966A (zh) 一种地源热泵制冷***优化控制方法
CN112413831A (zh) 一种中央空调节能控制***及方法
CN107247407B (zh) 一种基于云架构的大数据自我学习修正控制***及方法
CN117217407A (zh) 一种用于预评估电力***碳排放量的方法、介质及装置
CN112906856A (zh) 一种基于云平台的畜禽舍环境温度检测***
Xu et al. Review on control methods and strategies of greenhouse microclimate
Kang et al. Semi-autonomous greenhouse environment control by combining expert knowledge and machine learning
CN117469774B (zh) 空调***调控方法、装置、电子设备及存储介质
CN117519042B (zh) 基于ai技术的水泥生产智能控制方法、***及存储介质
CN117339980B (zh) 一种聚乳酸酶海洋垃圾降解控制方法及***
CN117408170B (zh) 一种适用于数据中心水冷***的节能预测控制方法
CN116224870B (zh) 一种基于多网络融合和植物光合作用反馈的智慧种植环境控制***及方法
Pucheta et al. A neuro-dynamic programming-based optimal controller for tomato seedling growth in greenhouse systems
Lysenko et al. Intelligent Algorithms for the Automation of Complex Biotechnical Objects
CN116954058B (zh) 一种锅炉NOx浓度预测与智能控制方法及***
Jiangqi et al. Research on intelligent supply control method of nutrient solution based on fuzzy neural network
Glad Neural Network Predictive Control for Greenhouse Climate
Patino et al. Approximate Optimal Control-Based Neurocontroller with a State Observation System for Seedlings Growth in Greenhouse
CN118393898A (zh) 生物隔离装置的智能控制方法及***

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22749198

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 18264747

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22749198

Country of ref document: EP

Kind code of ref document: A1