Heat storage capacity intelligent control method for assembly type sunlight greenhouse

Abstract

The invention relates to the field of sunlight greenhouse control, in particular to a heat storage capacity intelligent control method for an assembly type sunlight greenhouse. The method comprises the following steps: acquiring a multi-dimensional environmental parameter vector sequence and an initial PID parameter value of a crop; obtaining an environment parameter fluctuation factor, a crop growth stage factor and a long-term environment trend factor according to the multi-dimensional environment parameter vector sequence, and obtaining a target clustering number and a clustering cluster; obtaining a target clustering cluster according to the distance between the new multi-dimensional environment parameter vector and the cluster center of the clustering cluster; and calculating a PID parameter adjustment factor according to the distance between the new multi-dimensional environment parameter vector and the cluster center of the target cluster and the internal and external temperature difference, obtaining a final PID parameter according to the PID parameter adjustment factor and the initial PID parameter value, and performing real-time control on the heat storage amount of the greenhouse environment where the crops are located. According to the method, selection of the K value in clustering is optimized, so that the K value can adaptively respond to changes of environment and crop growth requirements.

Description

Technical Field

[0001] The invention relates to the field of solar greenhouse control, and in particular to a heat storage intelligent control method for an assembled solar greenhouse. Background Art

[0002] Prefabricated solar greenhouses are widely used in modern agricultural production, but their environmental control, especially the intelligent management of heat storage, still faces many challenges. Traditional greenhouse environmental control methods mainly rely on preset fixed control strategies, which are difficult to adapt to complex and changing environments and crop growth requirements. Therefore, in recent years, machine learning technology, especially K-means clustering algorithm, has been introduced into the field of greenhouse environmental control.

[0003] In the existing greenhouse environment control methods, clustering algorithms usually use a fixed number of clusters or select the optimal K value through the elbow rule. However, the fixed number of clusters cannot adapt to the dynamic changes of the greenhouse environment, resulting in insufficient classification when the environment changes rapidly and inability to flexibly respond to environmental state changes, which leads to inaccurate environmental state classification, thereby affecting the accuracy and efficiency of greenhouse environment PID control, resulting in poor crop growth conditions and energy waste.

[0004] Therefore, how to achieve dynamic selection of clustering number so that it can adaptively respond to changes in environment and crop growth requirements has become an urgent problem to be solved. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a heat storage intelligent control method for an assembled solar greenhouse to solve the problem of how to achieve dynamic selection of the number of clusters so that it can adaptively respond to changes in the environment and crop growth requirements.

[0006] An embodiment of the present invention provides a heat storage intelligent control method for an assembled solar greenhouse, the method comprising the following steps: Obtain a multidimensional environmental parameter vector of the crop at each sampling moment, and obtain a multidimensional environmental parameter vector sequence within the current preset period, wherein each of the multidimensional environmental parameter vectors includes temperature, humidity, light intensity, concentration and internal and external temperature difference, obtaining the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence; According to the multidimensional historical environmental parameter vector sequence before the multidimensional environmental parameter vector sequence, the environmental parameter fluctuation factor, the crop growth stage factor and the long-term environmental trend factor within the current preset time period are obtained, and according to the environmental parameter fluctuation factor, the crop growth stage factor and the long-term environmental trend factor within the current preset time period, the target cluster quantity is obtained; Clustering the multidimensional environmental parameter vector sequence to obtain K clusters equal to the number of target clusters, obtaining a new multidimensional environmental parameter vector of the crop, and obtaining the target cluster to which the new multidimensional environmental parameter vector belongs based on the distance between the new multidimensional environmental parameter vector and the cluster center of each cluster; According to the distance between the new multidimensional environmental parameter vector and the cluster center of the target cluster, and the internal and external temperature difference in the new multidimensional environmental parameter vector, a PID parameter adjustment factor is calculated, and according to the PID parameter adjustment factor and the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, the final PID parameter of the new multidimensional environmental parameter vector is obtained, and the heat storage of the greenhouse environment where the crops are located is controlled in real time according to the final PID parameter.

[0007] Preferably, the method for obtaining the environmental parameter fluctuation factor comprises: According to the multidimensional environmental parameter vector sequence, the temperature standard deviation, humidity standard deviation, light intensity standard deviation and concentration standard deviation; The temperature standard deviation, humidity standard deviation, light intensity standard deviation and The concentration standard deviation, the average value of the temperature standard deviation, the average value of the humidity standard deviation, the average value of the light intensity standard deviation and mean of the standard deviations of the concentrations; Calculate a first ratio between the temperature standard deviation of the multidimensional environmental parameter vector sequence and the average value of the temperature standard deviation, calculate a second ratio between the humidity standard deviation of the multidimensional environmental parameter vector sequence and the average value of the humidity standard deviation, calculate a third ratio between the illumination intensity standard deviation of the multidimensional environmental parameter vector sequence and the average value of the illumination intensity standard deviation, calculate the illumination intensity of the multidimensional environmental parameter vector sequence. The standard deviation of concentration is A fourth ratio between the average value of the concentration and the standard deviation, and a sum of squares of the first ratio, the second ratio, the third ratio and the fourth ratio are calculated as an environmental parameter fluctuation factor of the crop.

[0008] Preferably, the method for obtaining the crop growth stage factor comprises: For any multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, calculate a first difference between the temperature of the multidimensional environmental parameter vector and a preset base temperature, obtain a fourth ratio between the first difference and the number of multidimensional environmental parameter vectors in the multidimensional environmental parameter vector sequence, obtain a maximum value between the fourth ratio and a preset value, accumulate the maximum values ​​corresponding to each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, and use the corresponding accumulated value as the temperature index of the multidimensional environmental parameter vector sequence; Obtaining the temperature index of each of the multidimensional historical environmental parameter vector sequences before the multidimensional environmental parameter vector sequence, obtaining the maximum temperature index among the temperature indexes of the multidimensional environmental parameter vector sequence and each of the multidimensional historical environmental parameter vector sequences before the multidimensional environmental parameter vector sequence, and obtaining the accumulated value of the temperature indexes between the multidimensional environmental parameter vector sequence and each of the multidimensional historical environmental parameter vector sequences before the multidimensional environmental parameter vector sequence; Calculate a fifth ratio of the accumulated value of the temperature index to the maximum temperature index, perform inverse normalization processing on the product of the fifth ratio and a preset first environmental sensitivity parameter, use the corresponding result as a first inverse normalized value, and calculate a difference between a constant 1 and the first inverse normalized value as a temperature growth trend indicator; Obtaining a sine value of the product of the fifth ratio and pi, calculating a first product of the sine value and a preset second environmental sensitivity parameter, and calculating a sum of a constant 1 and the first product as a temperature adjustment coefficient; The multiplication result among the preset basic value, the temperature growth trend index and the temperature adjustment coefficient is calculated, and the sum of the multiplication result and the preset basic value is used as the crop growth stage factor.

[0009] Preferably, the method for obtaining the long-term environmental trend factor includes: According to a preset number of multidimensional historical environmental parameter vector sequences and the humidity, light intensity average and concentration, obtain the average humidity, average light intensity and concentration average, calculate the absolute value of the first difference between the humidity average and the preset optimal humidity, calculate the absolute value of the second difference between the illumination average and the preset optimal illumination, calculate the Concentration average and preset optimum The third difference absolute value of the concentration is obtained, and the average of the first difference absolute value, the second difference absolute value, and the third difference absolute value is obtained. Hyperbolic tangent processing is performed on the product of the average and a preset third environmental sensitivity parameter to obtain a corresponding hyperbolic tangent processing result, and the sum of the hyperbolic tangent processing result and a constant 1 is used as the long-term environmental trend factor.

[0010] Preferably, obtaining the target cluster quantity according to the environmental parameter fluctuation factor, the crop growth stage factor and the long-term environmental trend factor within the current preset time period includes: An initial number of clusters is obtained according to the environmental parameter fluctuation factor, and the initial number of clusters is optimized according to the crop growth stage factor and the long-term environmental trend factor to obtain a target number of clusters.

[0011] Preferably, obtaining the initial number of clusters according to the environmental parameter fluctuation factor includes: The difference between the preset maximum number of clusters and the preset minimum number of clusters is calculated as the second difference, the product of the environmental parameter fluctuation factor and the preset fourth environmental sensitivity parameter is inversely normalized to obtain a second inversely normalized value, the third difference between the constant 1 and the second inversely normalized value is calculated, the product value of the second difference and the third difference is obtained, the sum of the preset minimum number of clusters and the product value is rounded off to obtain the initial number of clusters of the crop.

[0012] Preferably, the step of optimizing the initial number of clusters according to the crop growth stage factor and the long-term environmental trend factor to obtain the target number of clusters includes: Obtain a second product of the crop growth stage factor and a preset growth stage factor weight coefficient, calculate the addition result of the second product and a constant 1, round off the addition result, the product of the initial cluster number and the long-term environmental trend factor, and use the corresponding result as the target cluster number.

[0013] Preferably, the calculating of the PID parameter adjustment factor according to the distance between the new multidimensional environmental parameter vector and the cluster center of the target cluster, and the internal and external temperature difference in the new multidimensional environmental parameter vector, comprises: Obtaining the distance between the new multidimensional environmental parameter vector and the cluster center of the target cluster as the target distance, performing inverse normalization processing on the product of a preset first parameter and the target distance to obtain a third inverse normalized value, obtaining a first subtraction result of a constant 1 and the third inverse normalized value, calculating the product of the first subtraction result and a preset second parameter as a third product, and calculating a first addition result of the third product and the constant 1; Performing inverse normalization processing on the product of the preset third parameter and the absolute value of the internal and external temperature difference of the new multidimensional environmental parameter vector to obtain a fourth inverse normalized value, obtaining a second subtraction result of a constant 1 and the fourth inverse normalized value, calculating the product of the second subtraction result and the preset fourth parameter as a fourth product, and calculating a second addition result of the fourth product and the constant 1; A product of the first addition result and the second addition result is obtained as a PID parameter adjustment factor.

[0014] Preferably, obtaining the final PID parameter of the new multidimensional environmental parameter vector according to the PID parameter adjustment factor and the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence comprises: According to the initial PID parameter value of each multidimensional environmental parameter vector in the target cluster to which the new multidimensional environmental parameter vector belongs, the mean of the initial PID parameter values ​​is obtained; and the product of the mean of the initial PID parameter values ​​and the PID parameter adjustment factor is calculated as the final PID parameter of the new multidimensional environmental parameter vector.

[0015] Preferably, obtaining the target cluster to which the new multidimensional environmental parameter vector belongs according to the distance between the new multidimensional environmental parameter vector and the cluster center of each cluster includes: The minimum distance between the new multidimensional environmental parameter vector and the cluster center of each cluster is selected, and the cluster corresponding to the minimum distance is used as the target cluster to which the new multidimensional environmental parameter vector belongs.

[0016] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: The present invention obtains a multidimensional environmental parameter vector of the crop at each sampling moment, and obtains a multidimensional environmental parameter vector sequence within the current preset period, wherein each of the multidimensional environmental parameter vectors includes temperature, humidity, light intensity, concentration and internal and external temperature difference, obtain the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence; according to the multidimensional historical environmental parameter vector sequence before the multidimensional environmental parameter vector sequence, obtain the environmental parameter fluctuation factor, crop growth stage factor and long-term environmental trend factor in the current preset time period, and according to the environmental parameter fluctuation factor, crop growth stage factor and long-term environmental trend factor in the current preset time period, obtain the target cluster number; cluster the multidimensional environmental parameter vector sequence to obtain K cluster clusters equal to the target cluster number, obtain a new multidimensional environmental parameter vector of the crop, based on According to the distance between the new multidimensional environmental parameter vector and the cluster center of each cluster cluster, the target cluster cluster to which the new multidimensional environmental parameter vector belongs is obtained; according to the distance between the new multidimensional environmental parameter vector and the cluster center of the target cluster cluster, and the internal and external temperature difference in the new multidimensional environmental parameter vector, the PID parameter adjustment factor is calculated, and according to the PID parameter adjustment factor and the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, the final PID parameter of the new multidimensional environmental parameter vector is obtained, and the heat storage of the greenhouse environment where the crop is located is controlled in real time according to the final PID parameter. The present invention optimizes the selection of K value in clustering by introducing growth stage factors and long-term environmental trend factors, and can automatically increase the number of clusters at the key growth stage of crops, providing a more refined environmental state classification, and this refined classification helps to finely regulate PID parameters to achieve faster and more accurate environmental parameter control, meeting the specific needs of crops at different growth stages. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0018] Figure 1 It is a method flow chart of a heat storage intelligent control method for an assembled solar greenhouse provided in Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0019] Embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, but should not be understood as limiting the present disclosure.

[0020] It should be noted that the terms "first", "second", etc. in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0021] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0022] See also Figure 1 , is a method flow chart of a heat storage intelligent control method for an assembled solar greenhouse provided in Embodiment 1 of the present invention, such as Figure 1 As shown, the method may include: Step S101, obtaining a multidimensional environmental parameter vector of the crop at each sampling moment, and obtaining a multidimensional environmental parameter vector sequence within the current preset period, wherein each of the multidimensional environmental parameter vectors includes temperature, humidity, light intensity, concentration and the internal and external temperature difference, and obtain the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence.

[0023] Collect environmental parameter information of assembled solar greenhouses, collect temperature, humidity, light intensity, The average value of the concentration and the temperature difference between inside and outside is recorded to obtain the multidimensional environmental parameter vector X. , 288 multidimensional environmental parameter vectors are collected every 24 hours to obtain a multidimensional environmental parameter vector sequence within 24 hours, where i represents the sequence number of the crop multidimensional environmental parameter vector sequence, j represents the sequence number of the multidimensional environmental parameter vector, is the temperature value of the jth multidimensional environmental parameter vector in the i-th multidimensional environmental parameter vector sequence (that is, the average temperature within 5 minutes), is the humidity value of the jth multidimensional environmental parameter vector in the i-th multidimensional environmental parameter vector sequence (that is, the average humidity value within 5 minutes), is the light intensity value of the jth multidimensional environmental parameter vector in the i-th multidimensional environmental parameter vector sequence (that is, the average light intensity within 5 minutes), is the i-th multidimensional environmental parameter vector sequence and the j-th multidimensional environmental parameter vector Concentration value (that is, within 5 minutes The average concentration is the internal and external temperature difference of the jth multidimensional environmental parameter vector in the i-th multidimensional environmental parameter vector sequence (that is, the difference between the average internal and external temperatures within 5 minutes). In this embodiment, 5 minutes is used as each sampling moment and 24 hours is used as the preset time period. There is no limitation here and it can be set according to the specific implementation scenario.

[0024] According to the records of the PID controller, the initial PID parameter value of each multidimensional environmental parameter vector in each multidimensional environmental parameter vector sequence is collected, and the PID control parameter (initial PID parameter value) is expressed as Y, , where P, I, and D are the proportional, integral, and differential parameters of the PID controller, respectively. , , They are the proportional, integral and differential parameters corresponding to the jth multidimensional environmental parameter vector in the i-th multidimensional environmental parameter vector sequence, and the PID control parameters are used as the control data for calculating the intelligent control of the heat storage capacity.

[0025] At this point, a multidimensional environmental parameter vector sequence within the current 24 hours and the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence can be obtained.

[0026] Step S102, based on the multidimensional historical environmental parameter vector sequence before the multidimensional environmental parameter vector sequence, obtain the environmental parameter fluctuation factor, crop growth stage factor and long-term environmental trend factor within the current preset time period, and obtain the target cluster number based on the environmental parameter fluctuation factor, crop growth stage factor and long-term environmental trend factor within the current preset time period.

[0027] The challenge faced in the intelligent control of heat storage in prefabricated solar greenhouses is how to accurately capture the short-term dynamic changes in the greenhouse environment. The greenhouse environment is affected by many factors such as sunlight intensity, external temperature, humidity changes, etc. These factors will change significantly in a short period of time as the environment changes. For example, on a clear summer day, the temperature and light intensity in the greenhouse at sunrise may rise sharply within an hour; while in cloudy weather, this change may be relatively slow. The traditional fixed cluster number method cannot flexibly respond to such dynamic changes, resulting in insufficient classification when the environment changes rapidly, and cannot capture the change of environmental state or the classification is too detailed when the environment is relatively stable, resulting in a waste of computing resources.

[0028] To solve this problem, we need to obtain a preliminary The core idea of ​​this method is to dynamically determine the number of initial clusters by analyzing the short-term fluctuations of key environmental parameters in the greenhouse, and select temperature, humidity, light intensity and Concentration and internal and external temperature difference are key parameters that directly affect plant growth and can better reflect the overall state of the greenhouse environment. Specifically, it is necessary to first obtain the multidimensional historical environmental parameter vector sequence before the current 24 hours according to the method for obtaining the multidimensional environmental parameter vector sequence within the current 24 hours, and then obtain the environmental parameter fluctuation factor W within the current 24 hours according to the multidimensional historical environmental parameter vector sequence before the multidimensional environmental parameter vector sequence within the current 24 hours, taking into account the temperature, humidity, light intensity and The concentration change within the current 24 hours is compared with the data from the previous few days to obtain a standardized and comparable environmental parameter fluctuation factor. The initial number of clusters within the current 24 hours is calculated by combining the obtained environmental parameter fluctuation factor with the preset maximum and minimum number of clusters.

[0029] The method for obtaining the environmental parameter fluctuation factor W of the multidimensional environmental parameter vector sequence corresponding to the current 24 hours is as follows: According to the multidimensional environmental parameter vector sequence, the temperature standard deviation, humidity standard deviation, light intensity standard deviation and concentration standard deviation; temperature standard deviation, humidity standard deviation, light intensity standard deviation and corresponding to a preset number of multidimensional historical environmental parameter vector sequences and the multidimensional environmental parameter vector sequences respectively; The concentration standard deviation, the average value of the temperature standard deviation, the average value of the humidity standard deviation, the average value of the light intensity standard deviation and The mean of the concentration standard deviations.

[0030] Calculate a first ratio between the temperature standard deviation of the multidimensional environmental parameter vector sequence and the average value of the temperature standard deviation, calculate a second ratio between the humidity standard deviation of the multidimensional environmental parameter vector sequence and the average value of the humidity standard deviation, calculate a third ratio between the illumination intensity standard deviation of the multidimensional environmental parameter vector sequence and the average value of the illumination intensity standard deviation, calculate the illumination intensity of the multidimensional environmental parameter vector sequence. The standard deviation of concentration is A fourth ratio between the average value of the concentration and the standard deviation, and a sum of squares of the first ratio, the second ratio, the third ratio and the fourth ratio are calculated as an environmental parameter fluctuation factor of the crop.

[0031] In one embodiment, a multidimensional environmental parameter vector sequence within the current 24 hours and a multidimensional historical environmental parameter vector sequence 2 days before the current 24 hours are obtained. There is no limitation here. The environmental parameter fluctuation factor W of the multidimensional environmental parameter vector sequence within the current 24 hours can be calculated according to the specific implementation scenario setting: Where: W is the environmental parameter fluctuation factor of the multidimensional environmental parameter vector sequence within the current 24 hours; , , , Respectively represent the temperature standard deviation, humidity standard deviation, light intensity standard deviation and concentration standard deviation; , , , They represent the average value of the temperature standard deviation, the average value of the humidity standard deviation, the average value of the light intensity standard deviation and the average value of the humidity standard deviation between the multidimensional environmental parameter vector sequence within the current 24 hours and the multidimensional historical environmental parameter vector sequence 2 days before the current 24 hours. mean of the standard deviations of the concentrations; Furthermore, the current The method for the initial number of clusters within an hour is as follows: The difference between the preset maximum number of clusters and the preset minimum number of clusters is calculated as the second difference, the product of the environmental parameter fluctuation factor and the preset fourth environmental sensitivity parameter is inversely normalized to obtain a second inversely normalized value, the third difference between the constant 1 and the second inversely normalized value is calculated, the product value of the second difference and the third difference is obtained, the sum of the preset minimum number of clusters and the product value is rounded off to obtain the initial number of clusters of the crop.

[0032] The calculation formula for the initial number of clusters is: in, is the initial number of clusters, is the preset minimum number of clusters; is the preset maximum number of clusters; is to preset a fourth environmental sensitivity parameter; is the rounding function; W is the environmental parameter fluctuation factor of the multidimensional environmental parameter vector sequence within the current 24 hours; is the inverse normalization function.

[0033] It should be noted that the preset minimum number of clusters and preset maximum number of clusters In order to set according to the changing characteristics of the external environment, in this embodiment, , , there is no restriction here, and it can be set according to the specific implementation scenario; the fourth environmental sensitivity parameter is preset It is an adjustable sensitivity parameter used to control the influence of environmental parameter fluctuation factors on the initial number of clusters; The function is used to round the result to the nearest integer; the environmental parameter fluctuation factor W of the current 24-hour multidimensional environmental parameter vector sequence reflects the severity of environmental changes in a short period of time. The greater the severity of environmental changes in a short period of time, the larger the corresponding initial number of clusters.

[0034] For example: Suppose one summer day observes Temperature, humidity, light intensity and The standard deviation of the concentration relative to the nearest The average standard deviation of the day is large, and the calculated environmental parameter fluctuation factor W is: ,set up , , There is no restriction here, and it can be set according to the specific implementation scenario. Approx. , which means that in summer when the environment changes greatly, In contrast, in a relatively stable environment, the environmental parameters The hourly standard deviation is close to The average standard deviation of the day, the calculated W is only ,at this time Maybe only or ,This dynamic adjustment ensures that the environmental state can be classified more ,detailedly when the environment changes dramatically, while fewer clusters are used when the environment is ,relatively stable, which not only ensures the accuracy of classification ,but also avoids unnecessary waste of computing resources.

[0035] The initial number of clusters can be obtained by obtaining environmental parameter fluctuation factors based on short-term dynamic changes in the greenhouse environment. However, there are limitations in relying solely on short-term environmental fluctuations to determine the target number of clusters, because greenhouse environmental control is highly affected by changes in the external environment, and crops have significantly different requirements for the environment at different growth stages. For example, tomatoes have stricter requirements for temperature and humidity during the flowering and fruiting period than during the vegetative growth period; and with the change of seasons, the overall environmental trend outside the greenhouse will also change. From summer to early autumn, the temperature difference between morning and evening becomes larger, the temperature rise in the morning becomes slower, and there are more rainy days. Therefore, the initial number of clusters obtained only through short-term environmental fluctuations ignores the long-term changing factors of these external environments and plant growth requirements, affecting the accuracy of environmental control and the growth quality of crops. To solve the above problems, it is necessary to obtain the crop growth stage factor and long-term environmental trend factor within the current 24 hours based on the multidimensional historical environmental parameter vector sequence before the multidimensional environmental parameter vector sequence within the current 24 hours, which are used to adjust the initial number of clusters. Optimize and adjust to obtain a more accurate number of clusters, which is also the target number of clusters.

[0036] First, obtain the crop growth stage factor of the multidimensional environmental parameter vector sequence within the current 24 hours: (1) For any multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, calculate a first difference between the temperature of the multidimensional environmental parameter vector and a preset base temperature, obtain a fourth ratio between the first difference and the number of multidimensional environmental parameter vectors in the multidimensional environmental parameter vector sequence, obtain a maximum value between the fourth ratio and a preset value, accumulate the maximum values ​​corresponding to each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, and use the corresponding accumulated value as the temperature index of the multidimensional environmental parameter vector sequence.

[0037] In one implementation, taking the i-th multidimensional environmental parameter vector sequence as an example, the temperature index of the i-th multidimensional environmental parameter vector sequence is calculated: in, is the temperature index of the i-th multidimensional environmental parameter vector sequence, is the temperature value of the jth multidimensional environmental parameter vector in the i-th multidimensional environmental parameter vector sequence; The base temperature of the crop.

[0038] It should be noted that To ensure that the temperature index only considers positive temperature differences, if Lower than Pick Because when the temperature is lower than the base temperature, plant growth almost stagnates and is not included in the effective accumulated temperature.

[0039] (2) Obtaining the temperature index of each of the multidimensional historical environmental parameter vector sequences before the multidimensional environmental parameter vector sequence, obtaining the maximum temperature index among the temperature indexes of the multidimensional environmental parameter vector sequence and each of the multidimensional historical environmental parameter vector sequences before the multidimensional environmental parameter vector sequence, and obtaining the accumulated value of the temperature indexes between the multidimensional environmental parameter vector sequence and each of the multidimensional historical environmental parameter vector sequences before the multidimensional environmental parameter vector sequence.

[0040] In one embodiment, taking the i-th multidimensional environmental parameter vector sequence as an example, the temperature index cumulative value between the i-th multidimensional environmental parameter vector sequence and each multidimensional historical environmental parameter vector sequence before the i-th multidimensional environmental parameter vector sequence is calculated, and the calculation expression of the temperature index cumulative value is: in, is the accumulated value of the temperature index between the i-th multidimensional environmental parameter vector sequence and each multidimensional historical environmental parameter vector sequence before the i-th multidimensional environmental parameter vector sequence, is the temperature index of the i-th multidimensional environmental parameter vector sequence; n is the total number of sequences up to the i-th multidimensional environmental parameter vector sequence; i represents the sequence number of the crop multidimensional environmental parameter vector sequence.

[0041] (3) Calculate the fifth ratio of the accumulated value of the temperature index to the maximum temperature index, perform inverse normalization on the product of the fifth ratio and the preset first environmental sensitivity parameter, use the corresponding result as the first inverse normalized value, calculate the difference between constant 1 and the first inverse normalized value as the temperature growth trend indicator; obtain the sine value of the product of the fifth ratio and pi, calculate the first product of the sine value and the preset second environmental sensitivity parameter, and calculate the sum of constant 1 and the first product as the temperature adjustment coefficient; calculate the multiplication result of the preset basic value, the temperature growth trend indicator, and the temperature adjustment coefficient, and use the sum of the multiplication result and the preset basic value as the crop growth stage factor.

[0042] In one embodiment, the crop growth stage factor of the multidimensional environmental parameter vector sequence within the current 24 hours is calculated: Among them, G is the crop growth stage factor of the multidimensional environmental parameter vector sequence within the current 24 hours; 0.5 is the basic value of the crop growth cycle; n is the total number of sequences up to the i-th multidimensional environmental parameter vector sequence; is the accumulated value of the temperature index between the i-th multidimensional environmental parameter vector sequence and each multidimensional historical environmental parameter vector sequence before the i-th multidimensional environmental parameter vector sequence; is the maximum temperature index in each multidimensional historical environmental parameter vector sequence before the current 24 hours and the multidimensional environmental parameter vector sequence within the current 24 hours; a is the preset first environmental sensitivity parameter; b is the preset second environmental sensitivity parameter; is the inverse normalization function.

[0043] It should be noted that the basic value of 0.5 ensures the minimum value of the factor G during the crop growth stage. It can be set according to specific crops. For example, the basic value of tomatoes with a growth cycle of 100 days is set to 0.5. Even in the early growth stage, a certain degree of environmental control accuracy is required. The reason for not using a higher value is that the plants are protected by the soil in the early growth stage, and slight changes in the external environment or small temperature adjustments in the solar greenhouse have little effect on them. There is no restriction here and it can be set according to the specific implementation scenario; Temperature growth trend indicator Describes the overall growth trend of the factor G value during the crop growth stage. As the plant begins to grow, the G value gradually increases. The preset first environmental sensitivity parameter a controls the growth rate. A larger a value will make G grow faster in the early growth stage, reflecting the process of gradually increasing requirements for environmental control as the plant grows; Temperature adjustment coefficient Control reaches the peak in the middle of growth. Taking tomatoes as an example, their flowering and fruiting period is in the middle of growth, which is a more important stage in cultivation. This coefficient makes the environmental control accuracy be raised to the highest level in the critical growth stage.

[0044] Then, the long-term environmental trend factors of the multidimensional environmental parameter vector sequence within the current 24 hours are obtained, including: According to a preset number of multidimensional historical environmental parameter vector sequences and the humidity, light intensity average and concentration, obtain the average humidity, average light intensity and concentration average, calculate the absolute value of the first difference between the humidity average and the preset optimal humidity, calculate the absolute value of the second difference between the illumination average and the preset optimal illumination, calculate the Concentration average and preset optimum The third difference absolute value of the concentration is obtained, and the average of the first difference absolute value, the second difference absolute value, and the third difference absolute value is obtained. Hyperbolic tangent processing is performed on the product of the average and a preset third environmental sensitivity parameter to obtain a corresponding hyperbolic tangent processing result, and the sum of the hyperbolic tangent processing result and a constant 1 is used as the long-term environmental trend factor.

[0045] In one embodiment, the long-term environmental trend factor of the multidimensional environmental parameter vector sequence within the current 24 hours is calculated: Among them, L is the long-term environmental trend factor of the multidimensional environmental parameter vector sequence within the current 24 hours; , , They are the average humidity, average light intensity and average humidity of the six multidimensional historical environmental parameter vector sequences before the current 24 hours and the multidimensional environmental parameter vector sequence within the current 24 hours. Concentration average value; , , They are the optimal humidity, optimal light intensity and optimal light intensity during the crop growth stage. concentration; To preset a third environmental sensitivity parameter; is the hyperbolic tangent function, and || represents the absolute value sign.

[0046] It should be noted that the third environmental sensitivity parameter is preset It is an adjustable sensitivity parameter. The initial value is 0.1, which can be adjusted according to actual needs. A larger value will be more sensitive to environmental changes; hyperbolic tangent function The humidity average value, light intensity average value and The average concentration is related to the optimal humidity, optimal light intensity and optimal Larger differences in concentration significantly increase the L value, while keeping the L value relatively stable when approaching the optimal state.

[0047] Example: Optimal humidity, optimal light intensity and optimal growth stage of tomatoes concentration , , The results of several scientific studies were referenced, including but not limited to Peet & Welles (2005), Shamshiri et al. (2018) and Kubota et al. (2018), which provided the optimal environmental parameter ranges for tomatoes at different growth stages, including but not limited to humidity, light intensity and Concentration, giving the optimal parameter values ​​for each growth stage of tomatoes: Seedling stage : Vegetative growth period : Flowering and fruiting period : Maturity : Finally, the initial number of clusters is optimized according to the crop growth stage factors and long-term environmental trend factors to obtain the target number of clusters, including: Obtain a second product of the crop growth stage factor and a preset growth stage factor weight coefficient, calculate the addition result of the second product and a constant 1, round off the addition result, the product of the initial cluster number and the long-term environmental trend factor, and use the corresponding result as the target cluster number.

[0048] In one embodiment, the calculation formula for the target cluster number is: in, is the number of target clusters; is the initial cluster number; G is the crop growth stage factor of the multidimensional environmental parameter vector sequence within the current 24 hours; L is the long-term environmental trend factor of the multidimensional environmental parameter vector sequence within the current 24 hours; is the preset growth stage factor weight coefficient; Is the rounding function.

[0049] It should be noted that the larger the crop growth stage factor G and the long-term environmental trend factor L of the multidimensional environmental parameter vector sequence within the current 24 hours, the larger the number of target clusters. The larger the value, the higher the preset growth stage factor weight coefficient. The initial value used to adjust the influence of the growth stage factor is set to 0.5 and can be adjusted according to specific needs.

[0050] For example: A greenhouse for growing tomatoes is in the flowering stage on the 40th day of tomato growth. If the initial , the calculated growth phase factor It is about 0.87, indicating that we are now in a stage where higher control accuracy is required; The average environmental parameters of the day deviate from the optimal parameters: the average humidity is lower than the optimal humidity , the average light intensity is lower than the optimal light intensity , while the average concentrations are close to the optimum concentrations; these deviations result in the calculated long-term environmental trend factors is about 1.24, indicating that the number of clusters needs to be further increased to better control these deviations; setting , then the final calculated target cluster number is about 11; this means that it is recommended to increase the number of clusters from the initial Increased to 11 to account for the current critical growth phase and unfavorable long-term environmental trends.

[0051] At this point, the target number of clusters for crops is obtained.

[0052] Step S103, clustering the multidimensional environmental parameter vector sequence to obtain K clusters equal to the number of target clusters, obtaining a new multidimensional environmental parameter vector of the crop, and obtaining the target cluster to which the new multidimensional environmental parameter vector belongs based on the distance between the new multidimensional environmental parameter vector and the cluster center of each cluster.

[0053] In one embodiment, the target number of clusters is obtained. Finally, the classic K-means algorithm is used to perform cluster analysis on the multidimensional environmental parameter vector sequence: using the temperature, humidity, light intensity, The environmental parameter vector of concentration and internal and external temperature difference, with the number of target clusters As the number of clusters K, cluster analysis is performed on the multidimensional environmental parameter vector sequence in the past 24 hours to obtain K clusters. Since the K-means algorithm is an existing method, a brief description of the steps is given: (1) Randomly select points as the initial cluster centers; (2) assign each data point to the nearest cluster center; (3) recalculate the center point of each cluster; (4) repeat steps (2) and (3) until the cluster center no longer changes significantly or reaches the preset number of iterations. After clustering is completed, each cluster center represents a specific environmental state.

[0054] Get the new multidimensional environmental parameter vector of the crop , for the kth cluster among the K clusters, calculate the distance between the new multidimensional environment parameter vector and the cluster center of the kth cluster: in, is the new multidimensional environment parameter vector The distance from the cluster center of the kth cluster; , , , , are the temperature value, humidity value, light intensity value, Concentration and internal and external temperature difference; , , , , The temperature value, humidity value, and light intensity value of the kth cluster center are respectively concentration value and internal and external temperature difference; k represents the serial number of the cluster; a is the serial number of the new multidimensional environmental parameter vector.

[0055] Furthermore, according to the distance between the new multidimensional environmental parameter vector and the cluster center of each cluster, the target cluster to which the new multidimensional environmental parameter vector belongs is obtained, including: The minimum distance between the new multidimensional environmental parameter vector and the cluster center of each cluster is selected, and the cluster corresponding to the minimum distance is used as the target cluster to which the new multidimensional environmental parameter vector belongs.

[0056] At this point, the target cluster to which the new multidimensional environmental parameter vector belongs is obtained.

[0057] Step S104, calculate the PID parameter adjustment factor according to the distance between the new multidimensional environmental parameter vector and the cluster center of the target cluster, and the internal and external temperature difference in the new multidimensional environmental parameter vector, obtain the final PID parameter of the new multidimensional environmental parameter vector according to the PID parameter adjustment factor and the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, and perform real-time control of the heat storage capacity of the greenhouse environment where the crops are located according to the final PID parameters.

[0058] After obtaining the target cluster to which the new multidimensional environmental parameter vector belongs, it is necessary to dynamically adjust the parameters of the PID parameter controller according to the distance between the new multidimensional environmental parameter vector and the target cluster to which it belongs, combined with the internal and external temperature difference.

[0059] First, according to the distance between the new multidimensional environmental parameter vector and the cluster center of the target cluster to which it belongs, and the internal and external temperature difference in the new multidimensional environmental parameter vector, the PID parameter adjustment factor is calculated. The specific method includes: Obtain the distance between the new multidimensional environmental parameter vector and the cluster center of the target cluster as the target distance, perform inverse normalization on the product of a preset first parameter and the target distance to obtain a third inverse normalized value, obtain a first subtraction result of a constant 1 and the third inverse normalized value, calculate the product of the first subtraction result and a preset second parameter as the third product, and calculate a first addition result of the third product and constant 1.

[0060] The product of the preset third parameter and the absolute value of the internal and external temperature difference of the new multidimensional environmental parameter vector is inversely normalized to obtain a fourth inversely normalized value, a second subtraction result of constant 1 and the fourth inversely normalized value is obtained, the product of the second subtraction result and the preset fourth parameter is calculated as the fourth product, and a second addition result of the fourth product and constant 1 is calculated.

[0061] A product of the first addition result and the second addition result is obtained as a PID parameter adjustment factor.

[0062] In one embodiment, the calculation formula of the PID parameter adjustment factor is: in, is the PID parameter adjustment factor; is the target distance between the new multidimensional environment parameter vector and the cluster center of the target cluster; is to preset the first parameter; To preset the second parameter; is the internal and external temperature difference of the new multidimensional environmental parameter vector; To preset the third parameter; is to preset the fourth parameter; is the inverse normalization function.

[0063] It should be noted that the first addition result For distance conversion based on exponential function, the second addition result It is a temperature difference conversion based on an exponential function. When the distance or temperature difference increases, the values ​​of these two parts will gradually transition from 0 to 1. The first parameter is preset Control the rate at which distance affects the adjustment factor; preset the second parameter It is an adjustable parameter that controls the influence of distance on the adjustment factor; the third parameter is preset Control the rate at which the temperature difference affects the adjustment factor; preset the fourth parameter is an adjustable parameter, which controls the influence of distance and temperature difference on the adjustment factor; in this embodiment, There is no restriction on the value of , which can be set according to the specific implementation scenario.

[0064] Further, according to the PID parameter adjustment factor and the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, the final PID parameter of the new multidimensional environmental parameter vector is obtained, including: According to the initial PID parameter value of each multidimensional environmental parameter vector in the target cluster to which the new multidimensional environmental parameter vector belongs, the mean of the initial PID parameter values ​​is obtained; and the product of the mean of the initial PID parameter values ​​and the PID parameter adjustment factor is calculated as the final PID parameter of the new multidimensional environmental parameter vector.

[0065] In one embodiment, the mean of the initial PID parameter values ​​in the target cluster to which the new multidimensional environmental parameter vector belongs is obtained. , calculate the final PID parameters of the new multidimensional environmental parameter vector: in, is the final PID parameter of the new multidimensional environmental parameter vector, is the mean of the initial PID parameter values; is the PID parameter adjustment factor; k represents the sequence number of the cluster.

[0066] Finally, the parameters of the PID parameter controller are dynamically adjusted according to the final PID parameters to perform real-time control of the heat storage in the greenhouse environment where the crops are located.

[0067] In summary, the embodiment of the present invention obtains a multidimensional environmental parameter vector of the crop at each sampling moment, and obtains a multidimensional environmental parameter vector sequence within the current preset period, wherein each of the multidimensional environmental parameter vectors includes temperature, humidity, light intensity, concentration and internal and external temperature difference, obtain the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence; according to the multidimensional historical environmental parameter vector sequence before the multidimensional environmental parameter vector sequence, obtain the environmental parameter fluctuation factor, crop growth stage factor and long-term environmental trend factor in the current preset time period, and according to the environmental parameter fluctuation factor, crop growth stage factor and long-term environmental trend factor in the current preset time period, obtain the target cluster number; cluster the multidimensional environmental parameter vector sequence to obtain K cluster clusters equal to the target cluster number, obtain a new multidimensional environmental parameter vector of the crop, based on According to the distance between the new multidimensional environmental parameter vector and the cluster center of each of the cluster clusters, the target cluster cluster to which the new multidimensional environmental parameter vector belongs is obtained; according to the distance between the new multidimensional environmental parameter vector and the cluster center of the target cluster cluster, and the internal and external temperature difference in the new multidimensional environmental parameter vector, the PID parameter adjustment factor is calculated, and according to the PID parameter adjustment factor and the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, the final PID parameter of the new multidimensional environmental parameter vector is obtained, and the heat storage of the greenhouse environment where the crop is located is controlled in real time according to the final PID parameter. The present invention optimizes the selection of K value in clustering by introducing growth stage factors and long-term environmental trend factors, and can automatically increase the number of clusters in the key growth stage of crops, providing a more refined classification of environmental status. This refined classification is conducive to refined regulation. Parameters enable faster and more accurate environmental parameter control to meet the specific needs of crops at different growth stages.

[0068] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A heat storage intelligent control method for an assembled solar greenhouse, characterized in that: The method comprises: Obtain a multidimensional environmental parameter vector of the crop at each sampling moment, and obtain a multidimensional environmental parameter vector sequence within the current preset period, wherein each of the multidimensional environmental parameter vectors includes temperature, humidity, light intensity, concentration and internal and external temperature difference, obtaining the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence; According to the multidimensional historical environmental parameter vector sequence before the multidimensional environmental parameter vector sequence, the environmental parameter fluctuation factor, the crop growth stage factor and the long-term environmental trend factor within the current preset time period are obtained, and according to the environmental parameter fluctuation factor, the crop growth stage factor and the long-term environmental trend factor within the current preset time period, the target cluster quantity is obtained; Clustering the multidimensional environmental parameter vector sequence to obtain K clusters equal to the number of target clusters, obtaining a new multidimensional environmental parameter vector of the crop, and obtaining the target cluster to which the new multidimensional environmental parameter vector belongs based on the distance between the new multidimensional environmental parameter vector and the cluster center of each cluster; According to the distance between the new multidimensional environmental parameter vector and the cluster center of the target cluster, and the internal and external temperature difference in the new multidimensional environmental parameter vector, a PID parameter adjustment factor is calculated, and according to the PID parameter adjustment factor and the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, the final PID parameter of the new multidimensional environmental parameter vector is obtained, and the heat storage of the greenhouse environment where the crops are located is controlled in real time according to the final PID parameter.

2. The heat storage intelligent control method for an assembled solar greenhouse according to claim 1, characterized in that: The method for obtaining the environmental parameter fluctuation factor comprises: According to the multidimensional environmental parameter vector sequence, the temperature standard deviation, humidity standard deviation, light intensity standard deviation and concentration standard deviation; The temperature standard deviation, humidity standard deviation, light intensity standard deviation and The concentration standard deviation, the average value of the temperature standard deviation, the average value of the humidity standard deviation, the average value of the light intensity standard deviation and mean of the standard deviations of the concentrations; Calculate a first ratio between the temperature standard deviation of the multidimensional environmental parameter vector sequence and the average value of the temperature standard deviation, calculate a second ratio between the humidity standard deviation of the multidimensional environmental parameter vector sequence and the average value of the humidity standard deviation, calculate a third ratio between the illumination intensity standard deviation of the multidimensional environmental parameter vector sequence and the average value of the illumination intensity standard deviation, calculate the illumination intensity of the multidimensional environmental parameter vector sequence. The standard deviation of concentration is A fourth ratio between the average value of the concentration and the standard deviation, and a sum of squares of the first ratio, the second ratio, the third ratio and the fourth ratio are calculated as an environmental parameter fluctuation factor of the crop.

3. The heat storage intelligent control method for an assembled solar greenhouse according to claim 1, characterized in that: The method for obtaining the crop growth stage factor comprises: For any multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, calculate a first difference between the temperature of the multidimensional environmental parameter vector and a preset base temperature, obtain a fourth ratio between the first difference and the number of multidimensional environmental parameter vectors in the multidimensional environmental parameter vector sequence, obtain a maximum value between the fourth ratio and a preset value, accumulate the maximum values ​​corresponding to each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence, and use the corresponding accumulated value as the temperature index of the multidimensional environmental parameter vector sequence; Obtaining the temperature index of each of the multidimensional historical environmental parameter vector sequences before the multidimensional environmental parameter vector sequence, obtaining the maximum temperature index among the temperature indexes of the multidimensional environmental parameter vector sequence and each of the multidimensional historical environmental parameter vector sequences before the multidimensional environmental parameter vector sequence, and obtaining the accumulated value of the temperature indexes between the multidimensional environmental parameter vector sequence and each of the multidimensional historical environmental parameter vector sequences before the multidimensional environmental parameter vector sequence; Calculate a fifth ratio of the accumulated value of the temperature index to the maximum temperature index, perform inverse normalization processing on the product of the fifth ratio and a preset first environmental sensitivity parameter, use the corresponding result as a first inverse normalized value, and calculate a difference between a constant 1 and the first inverse normalized value as a temperature growth trend indicator; Obtaining a sine value of the product of the fifth ratio and pi, calculating a first product of the sine value and a preset second environmental sensitivity parameter, and calculating a sum of a constant 1 and the first product as a temperature adjustment coefficient; The multiplication result among the preset basic value, the temperature growth trend index and the temperature adjustment coefficient is calculated, and the sum of the multiplication result and the preset basic value is used as the crop growth stage factor.

4. The heat storage intelligent control method for an assembled solar greenhouse according to claim 1, characterized in that: The method for obtaining the long-term environmental trend factor comprises: According to a preset number of multidimensional historical environmental parameter vector sequences and the humidity, light intensity average and concentration, obtain the average humidity, average light intensity and concentration average, calculate the absolute value of the first difference between the humidity average and the preset optimal humidity, calculate the absolute value of the second difference between the illumination average and the preset optimal illumination, calculate the Concentration average and preset optimum The third difference absolute value of the concentration is obtained, and the average of the first difference absolute value, the second difference absolute value, and the third difference absolute value is obtained. The product of the average and the preset third environmental sensitivity parameter is subjected to hyperbolic tangent processing to obtain the corresponding hyperbolic tangent processing result, and the sum of the hyperbolic tangent processing result and the constant 1 is used as the long-term environmental trend factor.

5. The heat storage intelligent control method for an assembled solar greenhouse according to claim 1, characterized in that: The step of obtaining the target cluster quantity according to the environmental parameter fluctuation factor, the crop growth stage factor and the long-term environmental trend factor within the current preset time period includes: An initial number of clusters is obtained according to the environmental parameter fluctuation factor, and the initial number of clusters is optimized according to the crop growth stage factor and the long-term environmental trend factor to obtain a target number of clusters.

6. The heat storage intelligent control method for an assembled solar greenhouse according to claim 5, characterized in that: The obtaining of the initial number of clusters according to the environmental parameter fluctuation factor comprises: The difference between the preset maximum number of clusters and the preset minimum number of clusters is calculated as the second difference, the product of the environmental parameter fluctuation factor and the preset fourth environmental sensitivity parameter is inversely normalized to obtain a second inversely normalized value, the third difference between the constant 1 and the second inversely normalized value is calculated, the product value of the second difference and the third difference is obtained, the sum of the preset minimum number of clusters and the product value is rounded off to obtain the initial number of clusters of the crop.

7. The heat storage intelligent control method for an assembled solar greenhouse according to claim 5, characterized in that: The step of optimizing the initial number of clusters according to the crop growth stage factor and the long-term environmental trend factor to obtain the target number of clusters includes: Obtain a second product of the crop growth stage factor and a preset growth stage factor weight coefficient, calculate the addition result of the second product and a constant 1, round off the addition result, the product of the initial cluster number and the long-term environmental trend factor, and use the corresponding result as the target cluster number.

8. The heat storage intelligent control method for an assembled solar greenhouse according to claim 1, characterized in that: The calculating of the PID parameter adjustment factor according to the distance between the new multidimensional environmental parameter vector and the cluster center of the target cluster, and the internal and external temperature difference in the new multidimensional environmental parameter vector, comprises: Obtaining the distance between the new multidimensional environmental parameter vector and the cluster center of the target cluster as the target distance, performing inverse normalization processing on the product of a preset first parameter and the target distance to obtain a third inverse normalized value, obtaining a first subtraction result of a constant 1 and the third inverse normalized value, calculating the product of the first subtraction result and a preset second parameter as a third product, and calculating a first addition result of the third product and the constant 1; Performing inverse normalization processing on the product of the preset third parameter and the absolute value of the internal and external temperature difference of the new multidimensional environmental parameter vector to obtain a fourth inverse normalized value, obtaining a second subtraction result of a constant 1 and the fourth inverse normalized value, calculating the product of the second subtraction result and the preset fourth parameter as a fourth product, and calculating a second addition result of the fourth product and the constant 1; A product of the first addition result and the second addition result is obtained as a PID parameter adjustment factor.

9. The heat storage intelligent control method for an assembled solar greenhouse according to claim 1, characterized in that: The step of obtaining the final PID parameter of the new multidimensional environmental parameter vector according to the PID parameter adjustment factor and the initial PID parameter value of each multidimensional environmental parameter vector in the multidimensional environmental parameter vector sequence comprises: According to the initial PID parameter value of each multidimensional environmental parameter vector in the target cluster to which the new multidimensional environmental parameter vector belongs, the mean of the initial PID parameter values ​​is obtained; and the product of the mean of the initial PID parameter values ​​and the PID parameter adjustment factor is calculated as the final PID parameter of the new multidimensional environmental parameter vector.

10. The heat storage intelligent control method for an assembled solar greenhouse according to claim 1, characterized in that: The step of obtaining the target cluster to which the new multidimensional environmental parameter vector belongs according to the distance between the new multidimensional environmental parameter vector and the cluster center of each cluster, comprises: The minimum distance between the new multidimensional environmental parameter vector and the cluster center of each cluster is selected, and the cluster corresponding to the minimum distance is used as the target cluster to which the new multidimensional environmental parameter vector belongs.

Images