A greenhouse environment intelligent control method
Through intelligent control methods based on environmental characteristics, combined with historical data and fuzzy control rule tables, the greenhouse equipment status is dynamically adjusted, which solves the problem of difficult correlation between temperature, humidity and solar radiation in the greenhouse environment, and achieves more efficient greenhouse environmental control .
Patent Information
- Application Number
- CN202411267023.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-09-11
AI Technical Summary
Existing greenhouse environmental control technology is difficult to accurately regulate the correlation between temperature, humidity and solar radiation, resulting in unsatisfactory control results.
Using an intelligent control method based on the current value of environmental characteristics and historical data, the working status of the greenhouse equipment is dynamically adjusted by calculating control parameters and fuzzy control rule tables to accurately control the greenhouse environment.
It improves the accuracy and effect of greenhouse environmental control, can more accurately describe the complex relationships and uncertainties between multiple variables, provides smooth membership changes, and enhances the flexibility and robustness of control.
Smart Images

Figure CN118778747B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of greenhouse environment control, and in particular to a greenhouse environment intelligent control method. Background Art
[0002] With the development of modern agriculture, greenhouse planting technology has played an important role in improving crop yield and quality. As an important part of modern agriculture, greenhouse planting provides an ideal environment for crop growth by artificially controlling environmental factors in the greenhouse, such as temperature, humidity, solar radiation, etc. The control of the greenhouse environment directly affects the growth of crops, among which temperature, humidity and solar radiation are the most important environmental factors. Accurate control of temperature, humidity and solar radiation can not only improve crop yield and quality, but also reduce the incidence of diseases and energy consumption. However, due to the complexity and variability of the greenhouse environment, it is difficult to accurately control the temperature, humidity and solar radiation of the greenhouse.
[0003] At present, fuzzy control technology is often used to intelligently control the temperature, humidity, and solar radiation of the greenhouse. The traditional fuzzy control technology is to collect the current values of the environmental characteristics inside the greenhouse, determine the corresponding fuzzy subsets according to the current values of the environmental characteristics, find the fuzzy subsets corresponding to the working status of each greenhouse equipment from the fuzzy control rule table, and defuzzify to obtain the precise control values of the working status of each greenhouse equipment. However, this control method does not take into account the correlation between the temperature, humidity, and solar radiation in the greenhouse. The temperature, humidity, and solar radiation in the greenhouse do not exist independently, but are mutually influenced and related. In addition, the membership function of this control method is fixed and cannot be changed dynamically according to the actual situation, so it is difficult to ensure that the control effect is always in a relatively ideal state. Summary of the invention
[0004] In order to solve the above technical problems, the present invention provides a greenhouse environment intelligent control method, which controls the greenhouse environment according to the current values and historical data of the environmental characteristics inside the greenhouse, improves the control accuracy, and effectively improves the control effect.
[0005] In order to solve the above problems, the present invention adopts the following technical solutions:
[0006] A greenhouse environment intelligent control method of the present invention comprises the following steps:
[0007] S1: Collecting current values and historical data of environmental characteristics inside the greenhouse, wherein the environmental characteristics include air temperature, air humidity and solar radiation intensity;
[0008] S2: preprocessing the collected historical data of environmental characteristics;
[0009] S3: Calculate control parameters based on current values, historical data, and target values of environmental characteristics, and control the operation of corresponding greenhouse equipment based on the control parameters.
[0010] Preferably, the step S3 comprises the following steps:
[0011] S31: Calculate control parameters based on current values, historical data, and target values of environmental characteristics;
[0012] S32: Calculate the membership value of the control parameter to each fuzzy subset on the control parameter domain, and use the fuzzy subset with a membership value greater than 0 as the input fuzzy subset;
[0013] S33: according to the current season and the input fuzzy subset, the membership function of the corresponding greenhouse equipment working state fuzzy subset is found from the preset fuzzy control rule table;
[0014] S34: using the membership value of the control parameter to the input fuzzy subset as the activation intensity to cut the membership function of the corresponding greenhouse equipment working state fuzzy subset;
[0015] S35: Aggregating the membership functions of the cut greenhouse equipment working state fuzzy subsets belonging to the same greenhouse equipment working state domain to obtain a corresponding aggregated fuzzy set;
[0016] S36: Defuzzify the aggregated fuzzy set to obtain an accurate control value corresponding to the working state of the greenhouse equipment;
[0017] S37: Control the operation of the corresponding greenhouse equipment according to the precise control value of the working state of the greenhouse equipment.
[0018] Preferably, the greenhouse equipment includes a fan, a wet curtain, an external sunshade curtain, and an internal sunshade curtain, and the working status of the greenhouse equipment includes the fan opening degree, the fan opening time, the wet curtain opening degree, the wet curtain opening time, the external sunshade curtain opening degree, and the internal sunshade curtain opening degree.
[0019] Preferably, the calculation formula of the control parameter is as follows:
[0020] ,
[0021] ,
[0022] Where z is the control parameter, T is the current value of the air temperature, and T0 is the target value of the air temperature. is the standard deviation of the historical data of air temperature, L is the current value of solar radiation intensity, L0 is the target value of solar radiation intensity, is the standard deviation of the historical data of solar radiation intensity, S is the current value of air humidity, S0 is the target value of air humidity, is the standard deviation of the historical data of air humidity, , , is the coefficient.
[0023] represents the interaction strength between air temperature and solar radiation intensity, represents the interaction strength between air temperature and air humidity, Indicates the strength of the interaction between air humidity and solar radiation intensity.
[0024] Preferably, the coefficient The Kendall rank correlation coefficient between air temperature and solar radiation intensity is calculated based on the historical data of air temperature and solar radiation intensity, and the Kendall rank correlation coefficient is used as the coefficient .
[0025] The calculation of Kendall's rank correlation coefficient is based on the consistency and inconsistency of data pairs, so it is particularly suitable for small samples or situations with complex data relationships, such as air temperature and solar radiation intensity.
[0026] Preferably, the coefficient It is obtained by the following method: the Spearman rank correlation coefficient between air temperature and air humidity is calculated based on the historical data of air temperature and air humidity, and the Spearman rank correlation coefficient is used as the coefficient .
[0027] The Spearman rank correlation coefficient is suitable for analyzing the monotonic relationship between air temperature and air humidity, not just a linear relationship.
[0028] Preferably, the coefficient It is obtained by the following method: the Pearson correlation coefficient between air humidity and solar radiation intensity is calculated based on the historical data of air humidity and solar radiation intensity, and the Pearson correlation coefficient is used as the coefficient .
[0029] Air humidity and solar radiation intensity may have a more direct linear relationship driven by physical or chemical reactions, and the Pearson correlation coefficient can be used to calculate the relationship between the two.
[0030] Preferably, there are three fuzzy subsets in the control parameter domain, namely {low K}, {medium K}, and {high K}.
[0031] Membership function corresponding to {low K} for:
[0032] ,
[0033] The membership function corresponding to {medium K} is as follows:
[0034] ,
[0035] The membership function corresponding to {high K} is as follows:
[0036] ,
[0037] where z is a control parameter.
[0038] The control parameter is calculated based on the current value, historical data, and target value of the environmental characteristics. The environmental characteristics include air temperature, air humidity, and solar radiation intensity. Therefore, the membership functions of {low K}, {medium K}, and {high K} consider the interaction effects among temperature, solar radiation, and humidity, which can help simulate the combined effects of these three factors more accurately, thereby describing the system state more accurately.
[0039] Preferably, the method of using the membership degree value of the control parameter for a certain input fuzzy subset as the activation intensity to cut the membership function of the corresponding greenhouse equipment working state fuzzy subset includes the following steps:
[0040] N1: Use the membership degree value of the control parameter for this input fuzzy subset as the activation intensity, denoted as a;
[0041] N2: Use the membership function corresponding to this input fuzzy subset found from the fuzzy control rule table in step S33 as the membership function to be cut;
[0042] N3: Cut the membership function to be cut with the activation intensity so that the value of the membership function to be cut is within the interval [0, a].
[0043] Preferably, the membership function of the greenhouse equipment working state fuzzy subset is a triangular membership function. The triangular membership function is used as the membership function of the greenhouse equipment working state fuzzy subset.
[0044] There are three fuzzy subsets on the domain of the fan opening degree, namely {off F}, {medium speed F}, and {high speed F},
[0045] The membership function corresponding to {off F} is as follows:
[0046] ,
[0047] The membership function corresponding to {medium speed F} is as follows:
[0048] ,
[0049] The membership function corresponding to {High speed F} is:
[0050] ,
[0051] where x1 is the degree of fan opening, 0 ≤ x1 ≤ 100, and the fan opening degree x1 represents that the fan operating wind speed is x1% of the maximum wind speed.
[0052] There are three fuzzy subsets on the fan opening time domain, namely {Short F}, {Medium F}, and {Long F},
[0053] The membership function corresponding to {Short F} is:
[0054] ,
[0055] The membership function corresponding to {Medium F} is:
[0056] ,
[0057] The membership function corresponding to {Long F} is:
[0058] ,
[0059] where x2 is the fan opening time, 0 ≤ x2 ≤ 60, and the fan opening time x2 represents that the fan needs to operate for x2 minutes.
[0060] There are three fuzzy subsets on the wet curtain opening degree domain, namely {Closed S}, {Partially opened S}, and {Fully opened S},
[0061] The membership function corresponding to {Closed S} is:
[0062] ,
[0063] The membership function corresponding to {Partially opened S} is:
[0064] ,
[0065] The membership function corresponding to {Fully opened S} is:
[0066] ,
[0067] Among them, \(x3\) is the opening degree of the wet curtain, \(0\leq x3\leq100\), and the opening degree \(x3\) of the wet curtain represents that the opening of the wet curtain is \(x3\%\) of the maximum opening.
[0068] There are three fuzzy subsets on the domain of the wet curtain opening time, namely \{Short S\}, \{Medium S\}, \{Long S\},
[0069] The membership function corresponding to \{Short S\} is:
[0070] ,
[0071] The membership function corresponding to \{Medium S\} is:
[0072] ,
[0073] The membership function corresponding to \{Long S\} is:
[0074] ,
[0075] where \(x4\) is the wet curtain opening time, \(0\leq x4\leq60\), and the wet curtain opening time \(x4\) represents that the wet curtain needs to run for \(x4\) minutes.
[0076] There are three fuzzy subsets on the domain of the external sunshade curtain opening degree, namely \{Closed W\}, \{Partially Open W\}, \{Fully Open W\},
[0077] The membership function corresponding to \{Closed W\} is:
[0078] ,
[0079] The membership function corresponding to \{Partially Open W\} is:
[0080] ,
[0081] The membership function corresponding to \{Fully Open W\} is:
[0082] ,
[0083] where \(x5\) is the external sunshade curtain opening degree, \(0\leq x5\leq100\), and the external sunshade curtain opening degree \(x5\) represents that the opening of the external sunshade curtain is \(x5\%\) of the maximum opening.
[0084] There are three fuzzy subsets on the domain of the internal sunshade curtain opening degree, namely \{Closed N\}, \{Partially Open N\}, \{Fully Open N\},
[0085] The membership function corresponding to \{Closed N\} for:
[0086] ,
[0087] Membership function corresponding to partial opening N} for:
[0088] ,
[0089] Membership function corresponding to {all open N} for:
[0090] ,
[0091] Among them, x6 is the opening degree of the inner sunshade curtain, 0≤x6≤100, and the opening degree of the inner sunshade curtain x6 means that the opening of the inner sunshade curtain is x6% of the maximum opening.
[0092] Preferably, the fuzzy control rule table is established based on expert experience.
[0093] Preferably, the preprocessing method is data cleaning.
[0094] The beneficial effects of the present invention are as follows: control parameters are calculated based on current values, historical data, and target values of environmental characteristics; the membership function corresponding to each greenhouse equipment is selected based on the control parameters, the current season, and a fuzzy control rule table; the fuzzification is performed to obtain precise control values corresponding to the working state of the greenhouse equipment, and the greenhouse environment is controlled; the control parameters are used to more accurately describe the complex relationships and uncertainties between multiple variables in the system, provide smooth membership changes, and fully utilize historical data for regulation, thereby improving control accuracy and effectively improving control effects. BRIEF DESCRIPTION OF THE DRAWINGS
[0095] Figure 1 is a flow chart of an embodiment. DETAILED DESCRIPTION
[0096] The technical solution of the present invention is further specifically described below through embodiments and in conjunction with the accompanying drawings.
[0097] Embodiment: A greenhouse environment intelligent control method of this embodiment, such as Figure 1 As shown, the following steps are included:
[0098] S1: Collect current values and historical data of environmental characteristics inside the greenhouse, including air temperature, air humidity and solar radiation intensity.
[0099] The greenhouse is equipped with a temperature sensor for collecting air temperature, a humidity sensor for collecting air humidity, and a solar radiation sensor for collecting solar radiation intensity.
[0100] S2: Preprocess the collected historical data of environmental characteristics.
[0101] The preprocessing method is data cleaning. Data cleaning includes removing outliers and filling missing values. The method of filling missing values is to use the mean of the two data points before and after the missing value as the missing value.
[0102] S3: Calculate control parameters based on current values, historical data, and target values of environmental characteristics, and control the operation of corresponding greenhouse equipment according to the control parameters (i.e., control the execution actions of corresponding greenhouse equipment in the current control cycle according to the control parameters).
[0103] Step S3 includes the following steps:
[0104] S31: Calculate the control parameter according to the current value, historical data and target value of the environmental characteristic. The target value of the environmental characteristic is the ideal value that the environmental characteristic is expected to achieve by control.
[0105] The calculation formula of the control parameters is as follows:
[0106] ,
[0107] ,
[0108] Where z is the control parameter, T is the current value of the air temperature, and T0 is the target value of the air temperature. is the standard deviation of the historical data of air temperature, L is the current value of solar radiation intensity, L0 is the target value of solar radiation intensity, is the standard deviation of the historical data of solar radiation intensity, S is the current value of air humidity, S0 is the target value of air humidity, is the standard deviation of the historical data of air humidity, , , is the coefficient.
[0109] represents the interaction strength between air temperature and solar radiation intensity, represents the interaction strength between air temperature and air humidity, Indicates the strength of the interaction between air humidity and solar radiation intensity.
[0110] coefficient The Kendall rank correlation coefficient between air temperature and solar radiation intensity is calculated based on the historical data of air temperature and solar radiation intensity, and the Kendall rank correlation coefficient is used as the coefficient The calculation of Kendall's rank correlation coefficient is based on the consistency and inconsistency of data pairs, so it is particularly suitable for small samples or situations with complex data relationships, such as air temperature and solar radiation intensity.
[0111] ,
[0112] Where n represents the number of historical samples of environmental characteristics, n c represents the number of sample consistent pairs (pairs of observations whose rankings of two factors are consistent in direction), n d represents the number of sample inconsistent pairs (pairs of observations where the rankings of the two factors disagree in direction).
[0113] coefficient It is obtained by the following method: the Spearman rank correlation coefficient between air temperature and air humidity is calculated based on the historical data of air temperature and air humidity, and the Spearman rank correlation coefficient is used as the coefficient The Spearman rank correlation coefficient is suitable for analyzing the monotonic relationship between air temperature and air humidity, not just a linear relationship.
[0114] ,
[0115] Among them, d i is the difference in ranking between the two variables of the ith sample (for example, if the first and third place differ by two places, the ranking difference is 2).
[0116] coefficient It is obtained by the following method: the Pearson correlation coefficient between air humidity and solar radiation intensity is calculated based on the historical data of air humidity and solar radiation intensity, and the Pearson correlation coefficient is used as the coefficient Air humidity and solar radiation intensity may have a more direct linear relationship driven by physical or chemical reactions, and the Pearson correlation coefficient can be used to calculate the relationship between the two.
[0117] ,
[0118] Among them, x i is the value of the air humidity of the i-th sample, y i is the value of solar radiation intensity of the i-th sample, is the sample mean of air humidity, Sample mean of solar radiation intensity.
[0119] S32: Calculate the membership value of the control parameter to each fuzzy subset on the control parameter domain, and use the fuzzy subset with a membership value greater than 0 as the input fuzzy subset.
[0120] There are three fuzzy subsets on the control parameter domain, namely {Low K}, {Medium K}, and {High K}.
[0121] The membership function corresponding to {Low K} is:
[0122] ,
[0123] The membership function corresponding to {Medium K} is:
[0124] ,
[0125] The membership function corresponding to {High K} is:
[0126] ,
[0127] where z is the control parameter.
[0128] S33: According to the current season and the input fuzzy subset, find the membership function of the corresponding greenhouse equipment working state fuzzy subset from the preset fuzzy control rule table.
[0129] The greenhouse equipment includes a fan, a wet curtain, an external sunshade curtain, and an internal sunshade curtain. The working states of the greenhouse equipment include the opening degree of the fan, the opening time of the fan, the opening degree of the wet curtain, the opening time of the wet curtain, the opening degree of the external sunshade curtain, and the opening degree of the internal sunshade curtain.
[0130] There are three fuzzy subsets on the domain of the fan opening degree, namely {Off F}, {Medium Speed F}, and {High Speed F}.
[0131] The membership function corresponding to {Off F} is:
[0132] ,
[0133] The membership function corresponding to {Medium Speed F} is:
[0134] ,
[0135] The membership function corresponding to {High Speed F} is:
[0136] ,
[0137] where x1 is the fan opening degree, 0 ≤ x1 ≤ 100, and the fan opening degree x1 represents that the fan running wind speed is x1% of the maximum wind speed.
[0138] There are three fuzzy subsets on the domain of the fan start-up time, namely {Short F}, {Medium F}, and {Long F}.
[0139] The membership function corresponding to {Short F} is:
[0140] ,
[0141] The membership function corresponding to {Medium F} is:
[0142] ,
[0143] The membership function corresponding to {Long F} is:
[0144] ,
[0145] Among them, x2 is the fan start-up time, 0 ≤ x2 ≤ 60, and the fan start-up time x2 means the fan needs to run for x2 minutes.
[0146] There are three fuzzy subsets on the domain of the wet curtain opening degree, namely {Closed S}, {Partially Open S}, and {Fully Open S}.
[0147] The membership function corresponding to {Closed S} is:
[0148] ,
[0149] The membership function corresponding to {Partially Open S} is:
[0150] ,
[0151] The membership function corresponding to {Fully Open S} is:
[0152] ,
[0153] Among them, x3 is the wet curtain opening degree, 0 ≤ x3 ≤ 100, and the wet curtain opening degree x3 means the wet curtain opening is x3% of the maximum opening.
[0154] There are three fuzzy subsets on the domain of the wet curtain start-up time, namely {Short S}, {Medium S}, and {Long S}.
[0155] The membership function corresponding to {Short S} is:
[0156] ,
[0157] The membership function corresponding to {Medium S} is:
[0158] ,
[0159] The membership function corresponding to {long S} is:
[0160] ,
[0161] where x4 is the wet curtain opening time, 0 ≤ x4 ≤ 60, and the wet curtain opening time x4 means the wet curtain needs to run for x4 minutes.
[0162] There are three fuzzy subsets on the domain of the external sunshade curtain opening degree, namely {closed W}, {partially opened W}, and {fully opened W},
[0163] The membership function corresponding to {closed W} is:
[0164] ,
[0165] The membership function corresponding to {partially opened W} is:
[0166] ,
[0167] The membership function corresponding to {fully opened W} is:
[0168] ,
[0169] where x5 is the external sunshade curtain opening degree, 0 ≤ x5 ≤ 100, and the external sunshade curtain opening degree x5 means the opening of the external sunshade curtain is x5% of the maximum opening.
[0170] There are three fuzzy subsets on the domain of the internal sunshade curtain opening degree, namely {closed N}, {partially opened N}, and {fully opened N},
[0171] The membership function corresponding to {closed N} is:
[0172] ,
[0173] The membership function corresponding to {partially opened N} is:
[0174] ,
[0175] The membership function corresponding to {fully opened N} is:
[0176] ,
[0177] Among them, x6 is the opening degree of the inner sunshade curtain, 0≤x6≤100, and the opening degree of the inner sunshade curtain x6 means that the opening of the inner sunshade curtain is x6% of the maximum opening.
[0178] The fuzzy control rule table is established based on expert experience. The fuzzy control rule table includes the fan opening degree and fan opening time control rule table, the wet curtain opening degree and wet curtain opening time control rule table, the external sunshade curtain opening degree control rule table, and the internal sunshade curtain opening degree control rule table.
[0179] The fan opening degree and fan opening time control rules are shown in Table 1 below:
[0180] Table 1
[0181]
[0182] The wet curtain opening degree and wet curtain opening time control rules are shown in Table 2 below:
[0183] Table 2
[0184]
[0185] The table of control rules for the opening degree of the external sunshade curtain is shown in Table 3 below:
[0186] Table 3
[0187]
[0188] The internal sunshade curtain opening degree control rule table is shown in Table 4 below:
[0189] Table 4
[0190]
[0191] S34: Using the membership value of the control parameter to the input fuzzy subset as the activation intensity to cut the membership function of the corresponding greenhouse equipment working state fuzzy subset.
[0192] The method of using the membership value of the control parameter to a certain input fuzzy subset as the activation strength to cut the membership function of the corresponding greenhouse equipment working state fuzzy subset in step S34 includes the following steps:
[0193] N1: The membership value of the control parameter to the input fuzzy subset is taken as the activation strength, denoted as a;
[0194] N2: taking the membership function corresponding to the input fuzzy subset found in the fuzzy control rule table in step S33 as the membership function to be cut;
[0195] N3: Use the activation intensity to cut the membership function to be cut so that the value of the membership function to be cut is in the interval [0, a].
[0196] For example: Assuming z = 0.3, the membership value of the control parameter to the fuzzy subset {low K} is 0.5, which controls the membership value of the parameter to the fuzzy subset {in K} is 0.5, which controls the membership value of the parameter to the fuzzy subset {high K} If is 0, the fuzzy subsets {low K} and {medium K} are used as the input control parameter fuzzy subsets. Assuming the current season is spring, the corresponding membership function of the fan opening degree is found from the fuzzy control rule table. , , membership function of fan opening time , , membership function of wet curtain opening degree , , membership function of wet curtain opening time , , membership function of the degree of opening of the external sunshade , , membership function of the opening degree of the inner sunshade , ;
[0197] because is 0.5, so 0.5 is used as the activation strength for the membership function , , , , , Cut so that the values of these membership functions are within the interval [0,0.5];
[0198] because is 0.5, so 0.5 is used as the activation strength for the membership function , , , , , Cutting is performed so that the values of these membership functions are in the interval [0,0.5].
[0199] S35: Aggregate the membership functions of the cut greenhouse equipment working state fuzzy subsets belonging to the same greenhouse equipment working state domain to obtain a corresponding aggregated fuzzy set.
[0200] The membership function of the fuzzy subset of greenhouse equipment working status that belongs to the same greenhouse equipment working status domain after cutting is recorded as , xk is the value of the working state of the greenhouse equipment, m is the number of membership functions, then the aggregate fuzzy set The calculation formula is as follows:
[0201] .
[0202] The membership functions of the fuzzy subsets of greenhouse equipment working status that belong to the same greenhouse equipment working status domain are merged after cutting, and the value of the overlapping part takes the maximum value of each membership function, that is, the maximum value of the membership in all output sets is taken to merge into a comprehensive set to obtain an aggregated fuzzy set.
[0203] For example: According to spring, {low K}, {medium K}, find the corresponding membership function of the fan opening degree from the fuzzy control rule table , ,because is 0.5, is 0.5, and 0.5 is used as the activation strength for the membership function Cut, use 0.5 as the activation strength for the membership function Cutting, the cut , Aggregate and get the aggregated fuzzy set .
[0204] S36: Defuzzify the aggregated fuzzy set using the centroid method to obtain the precise control value corresponding to the working state of the greenhouse equipment.
[0205] Assume that an aggregate fuzzy set is , then the defuzzification formula is as follows:
[0206] ,
[0207] Among them, J k To precisely control the value.
[0208] For example: using the centroid method to aggregate fuzzy sets Defuzzification is performed to obtain the precise control value J1 of the fan opening degree. , assuming that J1=30 means the wind speed of the fan is 30% of the maximum wind speed.
[0209] S37: Control the operation of the corresponding greenhouse equipment according to the precise control value of the working state of the greenhouse equipment.
[0210] In this scheme, the function of the fan is to lower the temperature when the air temperature inside the greenhouse is slightly high; the function of the wet curtain is to lower the temperature inside the greenhouse and increase the humidity when the air temperature inside the greenhouse is too high or the solar radiation is too strong; the function of the external sunshade is to reduce the intensity of solar radiation inside the greenhouse and provide insulation when the air temperature is low; the function of the internal sunshade is to open when the solar radiation is too high.
[0211] The sampling frequency of the air temperature, air humidity and solar radiation intensity in the greenhouse is once an hour. The control cycle of this method is 1 hour, that is, after the precise control value of the working state of each greenhouse equipment is calculated by this method, the corresponding greenhouse equipment is controlled to work for 1 hour according to the precise control value, and then the precise control value of the working state of each greenhouse equipment is recalculated by this method, and the corresponding greenhouse equipment is controlled to work for 1 hour according to the precise control value, and this cycle is repeated.
[0212] The control parameter domain is [0,1], the fan opening time domain and the wet curtain opening time domain are both [0,60], the fan opening degree domain, the wet curtain opening degree domain, the external sunshade curtain opening degree domain, and the internal sunshade curtain opening degree domain are all [0,100].
[0213] The control parameters are calculated based on the current values, historical data, and target values of environmental characteristics. Environmental characteristics include air temperature, air humidity, and solar radiation intensity. Therefore, the membership functions of {low K}, {medium K}, and {high K} take into account the interaction effects among temperature, solar radiation, and humidity, which can help to more accurately simulate the combined effects of these three factors and thus more accurately describe the system status.
[0214] The control parameters of this method better characterize the correlation between greenhouse variables. For example, the term γ1(T-T0)(L-L0) in the formula indicates the interaction between air temperature and air humidity. Such an interaction term can reflect the complex relationship between multiple variables in a real system. Assuming that in a certain environment, there is a negative correlation between air temperature and air humidity, that is, when the air temperature increases, the air humidity decreases. This negative correlation can be captured by the interaction term γ1, thereby more accurately describing the system state.
[0215] In a greenhouse environment, the air temperature, air humidity, and solar radiation intensity data measured by sensors are usually affected by various factors, resulting in certain measurement errors. The calculation formula of the control parameter can effectively represent and process these errors through its standard deviation parameter. For example, if the sensor data has a large fluctuation, the standard deviation parameter can be adjusted to be larger, so that the membership function of the control parameter has a better tolerance for these errors, thereby avoiding frequent control adjustments. By combining historical data, the calculation formula of the control parameter can dynamically adjust the standard deviation parameter according to the fluctuation pattern and trend of past environmental characteristics, thereby further improving the tolerance of the membership function of the control parameter to measurement errors and reducing unnecessary control adjustments. However, the traditional greenhouse environment fuzzy control technology relies on the currently collected data and is difficult to make full use of historical data, which limits its flexibility and robustness in dealing with complex environments. In addition, when the environmental characteristic variables change, the output membership value of the control parameter membership function changes smoothly, avoiding mutations and discontinuities. The membership function of the control parameter has a variety of parameters (standard deviation, cross-term coefficient) that can be adjusted, which can represent more diverse membership shapes and is suitable for modeling complex systems.
[0216] In general, the membership function of the control parameters can more accurately describe the complex relationship and uncertainty between multiple variables in the system, provide smooth membership changes, and is suitable for the modeling and control of complex systems. It can also make full use of historical data for regulation, thereby improving control accuracy and effectively improving the control effect, and controlling the air temperature, air humidity, and solar radiation intensity in the greenhouse within a reasonable range.
Claims
1. A greenhouse environment intelligent control method, characterized in that: The following steps are involved: S1: Collecting current values and historical data of environmental characteristics inside the greenhouse, wherein the environmental characteristics include air temperature, air humidity and solar radiation intensity; S2: preprocessing the collected historical data of environmental characteristics; S3: Calculate control parameters according to the current value, historical data and target value of environmental characteristics, and control the operation of corresponding greenhouse equipment according to the control parameters; The step S3 comprises the following steps: S31: Calculate control parameters based on current values, historical data, and target values of environmental characteristics; S32: Calculate the membership value of the control parameter to each fuzzy subset on the control parameter domain, and use the fuzzy subset with a membership value greater than 0 as the input fuzzy subset; S33: according to the current season and the input fuzzy subset, the membership function of the corresponding greenhouse equipment working state fuzzy subset is found from the preset fuzzy control rule table; S34: using the membership value of the control parameter to the input fuzzy subset as the activation intensity to cut the membership function of the corresponding greenhouse equipment working state fuzzy subset; S35: Aggregating the membership functions of the cut greenhouse equipment working state fuzzy subsets belonging to the same greenhouse equipment working state domain to obtain a corresponding aggregated fuzzy set; S36: Defuzzify the aggregated fuzzy set to obtain an accurate control value corresponding to the working state of the greenhouse equipment; S37: Controlling the operation of the corresponding greenhouse equipment according to the precise control value of the working state of the greenhouse equipment; The calculation formula of the control parameter is as follows: , , Among them, z is the control parameter, T is the current value of air temperature, T0 is the target value of air temperature, is the standard deviation of the historical data of air temperature, L is the current value of solar radiation intensity, L0 is the target value of solar radiation intensity, is the standard deviation of the historical data of solar radiation intensity, S is the current value of air humidity, S0 is the target value of air humidity, is the standard deviation of the historical data of air humidity, , , is the coefficient.
2. A greenhouse environment intelligent control method according to claim 1, characterized in that: The greenhouse equipment includes a fan, a wet curtain, an external sunshade curtain, and an internal sunshade curtain. The working status of the greenhouse equipment includes the fan opening degree, the fan opening time, the wet curtain opening degree, the wet curtain opening time, the external sunshade curtain opening degree, and the internal sunshade curtain opening degree.
3. A greenhouse environment intelligent control method according to claim 1, characterized in that: The coefficient The Kendall rank correlation coefficient between air temperature and solar radiation intensity is calculated based on the historical data of air temperature and solar radiation intensity, and the Kendall rank correlation coefficient is used as the coefficient .
4. A greenhouse environment intelligent control method according to claim 1, characterized in that: The coefficient It is obtained by the following method: the Spearman rank correlation coefficient between air temperature and air humidity is calculated based on the historical data of air temperature and air humidity, and the Spearman rank correlation coefficient is used as the coefficient .
5. The greenhouse environment intelligent control method according to claim 1, characterized in that: The coefficient It is obtained by the following method: the Pearson correlation coefficient between air humidity and solar radiation intensity is calculated based on the historical data of air humidity and solar radiation intensity, and the Pearson correlation coefficient is used as the coefficient .
6. A greenhouse environment intelligent control method according to claim 1, characterized in that: There are three fuzzy subsets on the control parameter domain, namely {low K}, {medium K}, and {high K}. Membership function corresponding to {low K} for: , Membership function corresponding to {Zhong K} is as follows: , Membership function corresponding to {high K} for: , Among them, z is the control parameter.
7. The greenhouse environment intelligent control method according to claim 1, characterized in that: The method of using the membership value of the control parameter to a certain input fuzzy subset as the activation strength to cut the membership function of the corresponding greenhouse equipment working state fuzzy subset in step S34 includes the following steps: N1: The membership value of the control parameter to the input fuzzy subset is taken as the activation strength, denoted as a; N2: taking the membership function corresponding to the input fuzzy subset found in the fuzzy control rule table in step S33 as the membership function to be cut; N3: Use the activation intensity to cut the membership function to be cut so that the value of the membership function to be cut is in the interval [0, a].
8. The greenhouse environment intelligent control method according to claim 1, characterized in that: The preprocessing method is data cleaning.
Citation Information
Patent Citations
System and method for fuzzy-based control and measurement of temperature and humidity for greenhouse
AU2021100311A4
Greenhouse crop growth environment intelligent adjusting method and system
CN116830943A
