A modern agricultural wisdom greenhouse ecological regulation system

Abstract

The present application relates to the technical field of data processing, and particularly relates to a modern agricultural intelligent greenhouse ecological regulation and control system, which comprises: an acquisition module, which is used for acquiring multi-dimensional ecological parameters of crops in a greenhouse from a sowing starting moment to a current moment; a determination module, which is used for determining potential influence trends of each dimension ecological parameter on crops in a time period; the determination module is also used for determining a common relationship between matching time periods of any two ecological parameters, determining a dimension reduction priority of any two ecological parameters based on dimension reduction parameters of the any two ecological parameters and the common relationship, and determining a principal component dimension according to the dimension reduction priority of the any two ecological parameters; and a regulation and control module, which is used for regulating and controlling an ecological environment of crops based on the principal component dimension. In this way, the present application improves the accuracy of ecological parameter regulation and control of the intelligent greenhouse.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a modern agricultural smart greenhouse ecological control system. Background Technology

[0002] In the process of agricultural modernization, smart greenhouses, with their intelligent environmental control capabilities, have become core equipment for breaking through the geographical limitations of traditional planting and improving crop yield and quality. They create an optimal growth environment for crops within the greenhouse by combining an intelligent control system with biosimulation technology: multiple types of sensors, including those for temperature, humidity, CO2 concentration, and light intensity, are deployed to collect key ecological parameters in real time; a backend microcomputer system analyzes and processes the collected data and, based on crop growth models, automatically controls actuators such as water curtains, fans, and shading panels to dynamically regulate temperature, humidity, gas composition, and light conditions within the greenhouse. This achieves precise control of the crop growth environment, significantly reducing the cost of manual intervention and improving planting efficiency and stability.

[0003] However, to ensure the precision of environmental regulation, continuous observation and collection of multivariate data such as temperature, humidity, CO2, and light intensity are required simultaneously. This results in high data dimensionality and a large workload for data collection, increasing the data storage and computational load on the backend system. Furthermore, to improve data processing efficiency and optimize regulatory decisions, dimensionality reduction of high-dimensional data is necessary to retain key information and eliminate redundant data. The current mainstream data dimensionality reduction technique is principal component analysis (PCA). Its core principle is to map high-dimensional data to a low-dimensional space through linear transformation, prioritizing the retention of data dimensions with the largest variance contribution rate, thereby simplifying the data. However, directly applying PCA to the dimensionality reduction of ecological parameters in smart greenhouses has significant limitations. Different types of ecological parameters in smart greenhouses have fundamentally different importance to crop growth, while PCA only filters dimensions based on the statistical variance characteristics of the data, without considering the weight of each ecological parameter's impact on crop growth. This dimensionality reduction approach, which emphasizes statistical characteristics while neglecting actual physiological impacts, may lead to the erroneous removal of key ecological parameters due to their low variance contribution, or the over-retention of non-key parameters, resulting in excessively large and abnormally frequent adjustments to some core parameters by the backend regulation system. If such improper regulation continues for an extended period, it will disrupt the optimal growth balance of crops within the greenhouse. This can lead to stunted growth in mild cases, and in severe cases, trigger pests and diseases or reduce yields, seriously limiting the precision of smart greenhouse regulation and planting efficiency. Thus, the aforementioned methods result in low accuracy in the regulation of ecological parameters within smart greenhouses. Summary of the Invention

[0004] To address the technical problem of low accuracy in the regulation of ecological parameters in smart greenhouses, the present invention aims to provide a modern agricultural smart greenhouse ecological regulation system.

[0005] To solve the above technical problems, the specific technical solution adopted is as follows:

[0006] In a first aspect, embodiments of the present invention provide a modern agricultural smart greenhouse ecological regulation system, comprising: an acquisition module for acquiring multidimensional ecological parameters of crops in the greenhouse from the start of sowing to the current time; a determination module for determining the potential impact trend of each ecological parameter on the crops over a time period based on the length of the time period between adjacent extreme points of each ecological parameter, the corresponding ecological parameter, and the ideal adaptation range of each ecological parameter; the determination module is further configured to determine the common relationship between the matching time periods of any two ecological parameters based on the difference between the potential impact trends between the matching time periods of any two ecological parameters, the potential impact trends and lengths of the current ecological parameters over each time period, and based on the dimensionality reduction parameters and common relationship of any two ecological parameters, determine the dimensionality reduction priority of any two ecological parameters, and determine the principal component dimension based on the dimensionality reduction priority of any two ecological parameters; and a regulation module for regulating the ecological environment of the crops based on the principal component dimension.

[0007] Preferably, the determining module is further configured to: determine the time period between adjacent minimum points of the ecological parameter; determine a first number of time points within the time period where the ecological parameter is below the minimum value of the ideal adaptation range, and a second number of time points within the time period where the ecological parameter exceeds the maximum value of the ideal adaptation range; determine the first mean of the ecological parameter within the time period where the maximum value of the ecological parameter exceeds the maximum value of the ideal adaptation range; and determine the potential impact trend based on the length of the time period, the first number, the second number, the maximum value within the time period, and the first mean.

[0008] Preferably, the determining module is further configured to: determine the time period between adjacent maximum points of the ecological parameter; determine a first number of time points within the time period where the ecological parameter is lower than the minimum value of the ideal adaptation range, and a second number of time points within the time period where the ecological parameter exceeds the maximum value of the ideal adaptation range; determine the minimum value within the time period and the second mean value of the ecological parameter within the time period where the ecological parameter is lower than the minimum value of the ideal adaptation range; and determine the potential impact trend based on the length of the time period, the first number, the second number, the minimum value within the time period, and the second mean value.

[0009] Preferably, the determining module is further configured to: determine the matching time period of any two ecological parameters based on the dynamic time warping algorithm, and determine the difference between the potential influence trends between the matching time periods of any two ecological parameters; and determine the common relationship between the matching time periods of any two ecological parameters based on the potential influence trends of each time period of the current ecological parameters, the length of the time period and the difference.

[0010] Preferably, the determining module is further configured to: determine the dynamic time warping distance between any two time periods of any two ecological parameters based on the dynamic time warping algorithm; and select the two time periods corresponding to the minimum value of the dynamic time warping distance as the matching time periods of any two ecological parameters.

[0011] Preferably, the determining module is further configured to: determine the average impact trend of the potential impact trends of the current ecological parameter and other ecological parameters for each time period, and the average length of the current ecological parameter and other ecological parameters for all time periods; superimpose the differences between the potential impact trends of each matched time period to obtain superimposed differences; and determine commonalities based on the average impact trend, the average length, and the superimposed differences.

[0012] Preferably, the determining module is further configured to: normalize the product of the dimensionality reduction parameters and common relationships of any two ecological parameters based on the range normalization algorithm, so as to obtain the dimensionality reduction priority of any two ecological parameters.

[0013] Preferably, the determining module is further configured to: determine the optimal dimensionality order for principal component analysis based on the dimensionality reduction priority of any two ecological parameters and the dimensionality reduction dimension for principal component analysis; and use the ecological data of each dimension in the optimal dimensionality order as the principal component dimensions.

[0014] Preferably, the regulation module is also used to: determine the ideal adaptation range corresponding to each principal component dimension; and to regulate the ecological environment of crops in relation to the principal component dimensions so that the ecological parameters of the principal component dimensions are within the ideal adaptation range.

[0015] Secondly, embodiments of the present invention provide a modern agricultural smart greenhouse ecological regulation system, comprising: a processor and a memory; wherein the memory is used to store a computer program that can run on the processor; the processor is used to execute the program stored in the memory to implement the following steps: acquiring multidimensional ecological parameters of crops in the greenhouse from the start of sowing to the current time; determining the potential impact trend of each dimension of ecological parameters on crops over a time period based on the length of the time period between adjacent extreme points of each dimension of ecological parameters, the corresponding ecological parameters, and the ideal adaptation range of each ecological parameter; determining the common relationship between the matching time periods of any two ecological parameters based on the difference between the potential impact trends between the matching time periods of any two ecological parameters, the potential impact trends and lengths of each time period of the current ecological parameter; determining the dimensionality reduction priority of any two ecological parameters based on the dimensionality reduction parameters and the common relationship of any two ecological parameters; determining the principal component dimension based on the dimensionality reduction priority of any two ecological parameters, and regulating the ecological environment of crops based on the principal component dimension.

[0016] This invention, through a module analysis of the time intervals between adjacent extreme points of ecological parameters in each dimension, the corresponding ecological parameter values, and the ideal adaptation range, accurately captures the potential impact trends of ecological parameters on crops. It then combines the common relationships between ecological parameters and the maximum projected variance to determine the priority of dimensionality reduction. Principal component dimensions are selected from both the influence relationships between ecological parameters and the data characteristics themselves, completely avoiding the omission of key parameters. This solves the problem of traditional principal component analysis relying solely on statistical variance to select dimensions, which easily leads to the accidental deletion of ecological parameters that are crucial for crop growth but have low variance contributions. Furthermore, the regulation module implements environmental regulation based on the accurately selected principal component dimensions, avoiding the problems of excessive adjustment range and abnormal frequency of core parameters caused by traditional dimensionality reduction. Through potential impact trend analysis and common relationship judgment, regulation decisions closely align with the actual growth needs of crops, preventing improper regulation from disrupting the suitable growth balance within the greenhouse, reducing the risk of slow crop growth, pest and disease occurrence, and yield reduction, significantly improving the regulation accuracy of smart greenhouses, and ensuring stable planting benefits. This improves the accuracy of ecological parameter regulation in smart greenhouses. Attached Figure Description

[0017] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic diagram of the module composition of a modern agricultural smart greenhouse ecological regulation system provided in one embodiment of the present invention;

[0019] Figure 2 This is a schematic diagram of the structure of a modern agricultural smart greenhouse ecological regulation system provided in one embodiment of the present invention. Detailed Implementation

[0020] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a modern agricultural intelligent greenhouse ecological control system proposed according to the present invention. In the following description, different "one embodiment" or "one embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0022] The specific solution of the modern agricultural smart greenhouse ecological regulation system provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Example 1:

[0024] Please see Figure 1 This document illustrates a schematic diagram of the module composition of a modern agricultural smart greenhouse ecological regulation system according to an embodiment of the present invention. The system includes: an acquisition module 101, used to acquire multidimensional ecological parameters of crops in the greenhouse from the start of sowing to the current time; a determination module 102, used to determine the potential impact trend of each ecological parameter on crops over a time period based on the length of the time period between adjacent extreme points of each ecological parameter, the corresponding ecological parameters, and the ideal adaptation range of each ecological parameter; the determination module 102 is also used to determine the common relationship between the matching time periods of any two ecological parameters based on the difference between the potential impact trends between matching time periods of any two ecological parameters, the potential impact trends and lengths of each time period of the current ecological parameter, and to determine the dimensionality reduction priority of any two ecological parameters based on the dimensionality reduction parameters and common relationship of any two ecological parameters, and to determine the principal component dimension based on the dimensionality reduction priority of any two ecological parameters; and a regulation module 103, used to regulate the ecological environment of crops based on the principal component dimension.

[0025] Specifically, in a greenhouse environment, a single greenhouse can utilize IoT technology to become a measurement and control area within a wireless sensor network. This network uses different sensor nodes and nodes with simple actuators, such as fans, low-voltage motors, and valves (actuators with low operating current), to measure substrate moisture, composition, pH value, temperature, as well as air humidity, air pressure, light intensity, and carbon dioxide concentration. Through model analysis, the greenhouse environment is automatically adjusted, and irrigation and fertilization are controlled to achieve optimal conditions for plant growth. For agricultural parks with numerous greenhouses, IoT can also achieve automatic information detection and control. Equipped with wireless sensor nodes, each node can monitor various environmental parameters. By receiving data from the wireless sensor aggregation node, storing, displaying, and managing the data, information from all base test points can be acquired, managed, and analyzed. This information is then displayed to users in each greenhouse using intuitive charts and curves. Simultaneously, various audible and visual alarms and SMS alerts are provided based on the needs of the planted crops, enabling intensive, networked, and remote management of the greenhouses. Therefore, this embodiment of the invention first uses various types of sensors installed in the greenhouse to collect and store multi-dimensional ecological data, including total light intensity, temperature, precipitation, soil pH, and nutrient content, for every hour of the historical period from sowing to the current monitoring time. This data is stored for subsequent operations. Then, the acquisition module 101 retrieves the multi-dimensional ecological data collected by various sensors from the database. Temperature differences are significant across seasons, but smaller within the same season. This embodiment uses one season as an example, meaning the analyzed data is collected within the same season, such as summer. The ecological data for each dimension is normalized to eliminate the influence of dimensions. As a specific example, the maximum-minimum normalization method can be used, which is existing technology and will not be elaborated further here. It should be noted that subsequent multi-dimensional ecological parameters are all normalized data.

[0026] More specifically, based on prior knowledge, regulating various ecological parameters within a greenhouse requires observing multivariate data, which increases the workload of data collection to some extent. However, to reduce the dimensionality of ecological parameter data, it is crucial to determine the data dimensions to be retained. Traditional Principal Component Analysis (PCA) primarily achieves this by calculating and retaining the data dimensions with the largest variance contributions. However, different data types of ecological parameters have varying importance, and analysis based solely on statistical characteristics may not yield the desired results. This could lead to excessively large adjustments to some parameters during regulation, potentially impacting crop growth if the adjustment is prolonged. Therefore, analysis based solely on statistical characteristics may not be sufficient; it is also necessary to assess the impact of different types of ecological parameters on crop growth. Based on the above logic, when various external factors are within a reasonable range and in a relative dynamic equilibrium, their effect on crop growth and quality is almost always positive. Furthermore, the impact of this positive effect on crop quality within a single day is relatively small. Therefore, using a single day to determine whether crops are in dynamic equilibrium and whether this has a positive or negative impact on quality is relatively one-sided. Analysis shows that the impact of a single ecological parameter on crops is reflected in how much the actual monitoring data under those conditions exceeds the crop's growth requirements (upper or lower limit), and the duration of this situation. Therefore, for any ecological parameter, its historically collected data is arranged chronologically on a coordinate axis. Based on prior knowledge of the theoretically most suitable production environment for any given crop and ecological parameter, the ideal adaptation range corresponding to the current ecological parameter is obtained.

[0027] More specifically, according to another embodiment of the present invention, the determining module 102 is further configured to: determine the time period between adjacent minimum points of the ecological parameter; determine a first number of time points within the time period where the ecological parameter is lower than the minimum value of the ideal adaptation range, and a second number of time points within the time period where the ecological parameter exceeds the maximum value of the ideal adaptation range; determine the maximum value of the ecological parameter within the time period and the first average value of the ecological parameter exceeding the maximum value of the ideal adaptation range within the time period; and determine a potential impact trend based on the length of the time period, the first number, the second number, the maximum value within the time period, and the first average value. Alternatively, according to another embodiment of the present invention, the determining module 102 is further configured to: determine the time period between adjacent maximum points of the ecological parameter; determine a first number of time points within the time period where the ecological parameter is lower than the minimum value of the ideal adaptation range, and a second number of time points within the time period where the ecological parameter exceeds the maximum value of the ideal adaptation range; determine the minimum value of the ecological parameter within the time period and the second average value of the ecological parameter lower than the minimum value of the ideal adaptation range within the time period; and determine a potential impact trend based on the length of the time period, the first number, the second number, the minimum value within the time period, and the second average value. In this embodiment, the method for obtaining the maximum and minimum values ​​of ecological data within a time period is as follows: Ecological parameters within the time period are arranged in chronological order to obtain an ecological parameter data sequence. Curve fitting is then performed on the data in the ecological parameter data sequence, with the horizontal axis of the fitted curve representing time and the vertical axis representing the ecological parameters. The maximum and minimum values ​​of the fitted curve are then obtained and used as the maximum and minimum values ​​of the ecological data within the time period, respectively. The curve fitting and the methods for obtaining the maximum and minimum values ​​on the curve are existing technologies and will not be elaborated upon further here.

[0028] Specifically, in this embodiment of the invention, for any ecological parameter in the historical data collected above, all maximum or minimum points are obtained. A time period is defined as the interval between any pair of adjacent maximum points or between any pair of adjacent minimum points, and these time periods respectively represent the prominence of the corresponding minimum or maximum value. The length of each time period is then obtained. , where i represents the local time interval corresponding to any maximum or minimum point. The length of the time interval can be the number of time points within that time interval.

[0029] Specifically, for the time interval between two adjacent pairs of minima, since there may be data points below the minimum value within the ideal adaptation range during the time interval between the two minima, the first number of time points below the minimum value within each time interval is obtained. Simultaneously, obtain the second number of time points where the current time period exceeds the maximum value of the ideal adaptation range. Then obtain the maximum value within that time period. The first mean of the ecological parameters that exceed the maximum value of the ideal adaptation range during this time period. Therefore, embodiments of the present invention use the following formula to calculate the potential impact trend of any ecological parameter on the current type of crop in the current time period:

[0030]

[0031] In the above formula, This represents the potential impact trend of any ecological parameter on the current type of crop during the i-th time period. This represents the length of the i-th time interval. This represents the first number of time points within the i-th time period that are below the minimum value of the ideal adaptation range. The second number of time points within the i-th time period that exceed the maximum value of the ideal adaptation range. This represents the maximum value within the i-th time interval. This represents the first mean of the ecological parameters whose maximum value exceeds the ideal adaptation range during the i-th time period. This represents the total number of data points within any given time period, excluding the minimum value that exceeds the ideal adaptation range. This represents the percentage of time points within the i-th time period that exceed the maximum value of the ideal adaptation range. A larger value indicates a longer potential negative impact on the current ecological parameters and a greater potential trend. Simultaneously, the larger the mean and maximum values ​​of the data points exceeding the range, the greater the extent of the exceedance and the stronger the impact trend. Therefore, multiplying these two values ​​together yields... The larger this value, the greater the potential impact trend.

[0032] Specifically, for the time interval between adjacent maxima, the first number of time points within any time segment that are below the minimum value of the ideal adaptation range is obtained. Simultaneously, obtain the second number of time points where the current time period exceeds the maximum value of the ideal adaptation range. Then obtain the minimum value within that time period. The second mean of ecological parameters that are below the minimum value of the ideal adaptation range within a given time period. Therefore, embodiments of the present invention use the following formula to calculate the potential impact trend of any ecological parameter on the current type of crop in the current time period:

[0033]

[0034] In the above formula, This represents the potential impact trend of any ecological parameter on the current type of crop during the i-th time period. This represents the length of the i-th time interval. This represents the first number of time points within the i-th time period that are below the minimum value of the ideal adaptation range. The second number of time points within the i-th time period that exceed the maximum value of the ideal adaptation range. This represents the minimum value within the i-th time interval. This represents the second mean of the ecological parameters that are below the minimum value of the ideal adaptation range during the i-th time period. This indicates the absolute value sign.

[0035] Thus, by iterating through all the maximum and minimum points, each data point can correspond to a potential trend. It is worth noting that for data points within the normal range, because... The value is 0, therefore the potential impact trend is 0.

[0036] More specifically, as the PCA algorithm acquires principal components, the proportion of principal components formed by any dimension gradually decreases with each dimensionality reduction process. Therefore, the earlier the dimensionality reduction order, the greater the proportion of principal components it occupies, which also means that more data features can be retained. At the same time, the proportion of principal components is related to their maximum projection variance, and this variance will also change depending on the dimensionality reduction order. Analysis reveals that various ecological parameters within the greenhouse exhibit a certain periodicity, fluctuating within a relatively suitable range. For instance, when the greenhouse temperature is low, adjustments can be made at a certain point to increase it to a peak value. Then, temperature adjustments are stopped, and the temperature gradually decreases due to crop growth and airflow within the greenhouse. When the temperature reaches the minimum of the ideal adaptive range, it is readjusted, resulting in a periodic change. Similarly, when the temperature is high, changes in airflow can lower the temperature. Once the minimum temperature range is reached, adjustments are stopped, and the temperature naturally rises, exhibiting a periodic change. For a specific ecological parameter, artificial adjustments occur within a certain period, after which the parameter begins to change naturally. The initial ecological parameter value at the time of change is determined by the natural rate of change of the humidity, which is influenced by the values ​​of other types of ecological parameters. For example, the higher the temperature inside the greenhouse, the faster the humidity decreases. Therefore, it can be concluded that the shorter the period of any ecological parameter among all types of data, and the smaller the difference in the potential influence trend of other types of data, the higher the similarity between the external conditions, the more similar the numerical change patterns calculated above, the greater the reference weight for judging common characteristics, and the higher its priority; conversely, the lower the priority. At the same time, the greater the actual potential influence trend on crops for each time period, the higher the priority of that type of data itself. Therefore, the dimensionality reduction priority of any combination of ecological parameters can be re-evaluated based on the influence of other types of ecological parameters on a certain ecological parameter.

[0037] More specifically, according to another embodiment of the present invention, the determining module is further configured to: determine the matching time period of any two ecological parameters based on the dynamic time warping algorithm, and determine the difference between the potential influence trends between the matching time periods of any two ecological parameters; and determine the common relationship between the matching time periods of any two ecological parameters based on the potential influence trends of each time period of the current ecological parameters, the length of the time period and the difference.

[0038] Specifically, this embodiment of the invention takes the j-th ecological parameter and other r-th ecological parameters as examples. Since different types of ecological parameters exhibit varying degrees of periodicity, and the time length between any two maxima or minima differs, direct segment-by-segment comparisons are not possible at the sampling time points. Therefore, this embodiment utilizes Dynamic Time Warping (DTW) to match the j-th ecological parameter data with various time periods of other r-th ecological parameter data. At this point, for each time period of the j-th ecological parameter, a matching time period can be found for the r-th ecological parameter. Furthermore, the difference in potential influence trends between the matching time periods of any two ecological parameters can also be obtained. The specific method for obtaining the difference between the potential impact trends of any two ecological parameters across matching time periods is as follows: calculate the absolute value of the difference between the potential impact trends of any two ecological parameters across matching time periods, and use this absolute value as the difference between the potential impact trends.

[0039] More specifically, according to another embodiment of the present invention, the determining module 102 is further configured to: determine the dynamic time warping distance between any two time periods of any two ecological parameters based on the dynamic time warping algorithm; and select the two time periods corresponding to the minimum value of the dynamic time warping distance as the matching time periods of any two ecological parameters.

[0040] More specifically, according to another embodiment of the present invention, the determining module 102 is further configured to: determine the average influence trend of the potential influence trends of the current ecological parameter and other ecological parameters in each time period, and the length mean of all time periods of the current ecological parameter and other ecological parameters; superimpose the differences between the potential influence trends between each matched time period to obtain superimposed differences; and determine commonalities based on the average influence trend, the length mean, and the superimposed differences.

[0041] Specifically, since the potential impact trend reflects the data performance of the current type of ecological parameter within a local time period, the potential impact trend of the j-th ecological parameter in the i-th time period is obtained. By traversing all historical time periods, the average of the potential impact trends of the j-th ecological parameter and the other r-th ecological parameter is obtained. Simultaneously, obtain the average length of all time periods for the j-th ecological parameter and other ecological parameters. Based on the above logic, we can conclude that... Therefore, in this embodiment of the invention, the common relationship between the j-th ecological parameter and other r-th ecological parameters is calculated using the following formula:

[0042]

[0043] In the above formula, This represents the common relationship between the j-th ecological parameter and the other r-th ecological parameter. This represents the mean of the potential influence trends of the j-th ecological parameter and the other r-th ecological parameter. This represents the average length of all time periods for the j-th ecological parameter and the other r-th ecological parameter. This represents the difference in the potential impact trends between the j-th ecological parameter and the i-th matching time period of the other r-th ecological parameter. This represents the number of time periods where the j-th ecological parameter matches with other r-th ecological parameters. This is determined by the potential difference in influence trends between two matching time periods. By averaging and combining this with the logical inverse proportional relationship, we obtain... ,in, This indicates the superposition of differences; by combining the above two equations, we obtain... The larger this value, the more obvious the common relationship between the j-th ecological parameter and the other r-th ecological parameter. It should be noted that 0.001 is introduced into the denominator to prevent the denominator from being 0.

[0044] The mean of the potential impact trends of the j-th ecological parameter and the other r-th ecological parameter is the average of all potential impact trends of the j-th ecological parameter and the other r-th ecological parameter. The mean of the lengths of all time periods of the j-th ecological parameter and the other r-th ecological parameter is the average of the lengths of all time periods of the j-th ecological parameter and the other r-th ecological parameter.

[0045] More specifically, for any ecological parameter in any dimension, the data is centered and then divided by its standard deviation to ensure that the mean is zero and the standard deviation is one. Furthermore, ecological parameters from multiple dimensions are paired to obtain several possible combinations. For any given combination, the coordinate system formed by the two combinations can be obtained, from which the dimensionality reduction parameter can be calculated. This dimensionality reduction parameter can be the maximum projected variance between the j-th ecological parameter and other r-th ecological parameters. In calculating the maximum projection variance, a two-dimensional data matrix of the j-th ecological parameter and the other r-th ecological parameter is first constructed. Then, the two-dimensional data matrix is ​​decentered, and the covariance matrix is ​​calculated. Finally, the two eigenvalues ​​and corresponding eigenvectors of this covariance matrix are calculated. The largest of the two eigenvalues ​​is the maximum projection variance of the coordinate system formed by the j-th ecological parameter and the other r-th ecological parameter, and the corresponding projection direction is the optimal projection direction. Next, combining the common relationships between the ecological parameters of each dimension calculated above, the corrected principal component feature parameters are obtained. These parameters also indicate the dimensionality reduction priority of any combination of dimensions. Therefore, according to another embodiment of the present invention, the determining module 102 is further used to: normalize the product of the dimensionality reduction parameters and common relationships of any two ecological parameters based on the range normalization algorithm to obtain the dimensionality reduction priority of any two ecological parameters.

[0046] Specifically, the embodiments of the present invention use the following formula to calculate the dimensionality reduction priority:

[0047]

[0048] In the above formula, This indicates the dimensionality reduction priority of the j-th ecological parameter compared to other r-th ecological parameters. This represents the maximum projection variance. This represents the common relationship between the j-th ecological parameter and the r-th ecological parameter. f() represents the range normalization algorithm, which normalizes the product of the maximum projection variance and the common relationship to obtain the dimensionality reduction priority of ecological parameters of any combination dimension; the larger the value, the more obvious the performance of the principal component within the dimension of the current combination of ecological parameters, and the greater the demand for the current one that is at the front of the dimensionality reduction order.

[0049] More specifically, this embodiment of the invention, combined with the principle of the PCA algorithm, requires sequential dimensionality reduction of multiple ecological parameters until the final contribution rate exceeds a threshold. The dimensions are arranged in different orders, noting that the initial two dimensions are not prioritized, resulting in several possible arrangements. According to the dimensionality reduction priority calculation method described above, for each arrangement, as dimensionality reduction progresses, several dimensions are essentially reduced to a single dimension and combined with the next dimension in the arrangement. Therefore, the importance and priority of these combinations are continuously calculated in the above calculation steps. Thus, for any arrangement, a dimensionality reduction priority can be obtained for each dimensionality reduction step. Therefore, based on the above logic, to obtain the arrangement method that best reflects the influence of dimensional data on the actual growth state of crops, the principal components that most significantly affect output are obtained. The method for determining the optimal order is to obtain the maximum projection variance accumulated during each dimensionality reduction process. The larger the accumulated value and the fewer dimensions required, the better the representation of the influence relationship. This yields the optimal arrangement order and its corresponding maximum projection variance. Therefore, according to another embodiment of the present invention, the determining module 102 is further configured to: determine the optimal dimensionality arrangement order of principal component analysis based on the dimensionality reduction priority of any two ecological parameters and the dimensionality reduction dimension of the principal component analysis; and use the ecological data of each dimension in the optimal dimensionality arrangement order as the principal component dimension.

[0050] Specifically, in this embodiment of the invention, principal components are selected for PCA dimensionality reduction by manually setting thresholds. In this embodiment, the most suitable dimensionality combination order is obtained according to the above steps, and PCA dimensionality reduction is performed on the ecological parameters. Specifically, the top 70% of the feature contribution rates are selected based on empirical values. Therefore, the formula for determining the optimal dimensional arrangement order is:

[0051]

[0052] In the above formula, The probability of the optimal dimensional arrangement of multi-dimensional ecological parameters. This represents the dimensionality reduction priority of the p-th dimensionality reduction in the optimal dimensionality arrangement. g represents the total number of dimensions. Since the ecological parameters of each dimension are paired, the total number of dimensionality reductions, which corresponds to the number of occurrences of a dimensionality reduction priority, is thus... y represents the total number of dimensions reduced. This represents the number of feature contribution rates retained. Specifically, by accumulating the maximum projection variance across all dimensionality reduction processes, the accumulated value satisfies the condition when iterating to the (y-1)th dimension. ,and The smaller the better, and the number of dimensions required. The fewer the number, the better the performance in influencing the relationship. The higher the priority of the summation of all ecological parameters, the better. ,Pick The order of the dimensions corresponding to the maximum value is the optimal order of dimensions, and the z principal component dimensions with the most influence in the optimal order of dimensions are obtained.

[0053] in, The method for obtaining the dimensionality priority is as follows: A dimensionality reduction priority matrix is ​​constructed using the dimensionality reduction priority of all pairwise ecological parameters; using the dimensionality reduction priority matrix as input, a greedy merging algorithm is used to iteratively select the pair of dimensions with the highest priority for merging, recording the merging sequence to obtain a basic permutation order; at each step of the greedy merging, the first K=5 (empirical values) optimal merging choices are retained, and multiple merging paths are expanded in parallel to generate multiple permutations. For the above basic permutation order, random swapping or reversal operations are repeatedly performed to generate several neighborhood permutations. Thus, all permutation orders obtained in the above steps constitute a candidate set. For any permutation order in the above candidate set, its i-th... The next dimensionality reduction operation corresponds to merging two nodes. If both nodes are of the original dimension, then the corresponding value in the dimensionality reduction priority matrix is ​​taken as the dimensionality reduction priority. If any node consists of a composite dimension composed of two or more original ecological parameter dimensions, then the arithmetic mean of all corresponding values ​​in the dimensionality reduction priority matrix between all original dimensions contained in the composite dimension and all original dimensions contained in another node is taken, thereby obtaining the priority for this dimensionality reduction. .

[0054] More specifically, based on the z principal component dimensions obtained from the above operations, these represent the z ecological parameters that have the greatest impact on crop growth in the greenhouse over time. At this point, the z ecological parameters with the greatest impact are used as the primary adjustment targets. Based on the ideal adaptation range of these ecological parameters at the current crop growth stage, the ecological environment is adjusted. Therefore, according to another embodiment of the present invention, the regulation module 103 is further used to: determine the ideal adaptation range corresponding to each principal component dimension; and to regulate the ecological environment of the crop in relation to the principal component dimensions, so that the ecological parameters of the principal component dimensions are within the ideal adaptation range.

[0055] Specifically, this embodiment of the invention uses the z most influential ecological parameters as the main adjustment targets. Based on the ideal adaptation range of these ecological parameters at the current crop growth stage, if a certain ecological parameter in the principal component dimension is not within its corresponding ideal adaptation range, it can be brought within the ideal adaptation range by increasing the ventilation system power, increasing the total irrigation water, increasing the pesticide spraying amount, etc. Ecological parameters outside the principal component dimension are not processed.

[0056] This invention provides a modern agricultural smart greenhouse ecological control system, comprising: a processor and a memory; wherein the memory is used to store a computer program that can run on the processor; the processor is used to execute the program stored in the memory to implement the steps of the modern agricultural smart greenhouse ecological control system as mentioned in the first aspect.

[0057] This invention, through a module analysis of the time intervals between adjacent extreme points of ecological parameters in each dimension, the corresponding ecological parameter values, and the ideal adaptation range, accurately captures the potential impact trends of ecological parameters on crops. It then combines the common relationships between ecological parameters and the maximum projected variance to determine the priority of dimensionality reduction. Principal component dimensions are selected from both the influence relationships between ecological parameters and the data characteristics themselves, completely avoiding the omission of key parameters. This solves the problem of traditional principal component analysis relying solely on statistical variance to select dimensions, which easily leads to the accidental deletion of ecological parameters that are crucial for crop growth but have low variance contributions. Furthermore, the regulation module implements environmental regulation based on the accurately selected principal component dimensions, avoiding the problems of excessive adjustment range and abnormal frequency of core parameters caused by traditional dimensionality reduction. Through potential impact trend analysis and common relationship judgment, regulation decisions closely align with the actual growth needs of crops, preventing improper regulation from disrupting the suitable growth balance within the greenhouse, reducing the risk of slow crop growth, pest and disease occurrence, and yield reduction, significantly improving the regulation accuracy of smart greenhouses, and ensuring stable planting benefits. This improves the accuracy of ecological parameter regulation in smart greenhouses.

[0058] Furthermore, this embodiment of the invention analyzes the local excess or missing values ​​of external conditions in historical data to determine local anomalies and obtain the potential impact trend of any extreme point. This operation avoids the inaccuracy of using historical data from a single day to determine whether crops are in dynamic equilibrium and the impact on quality, thus improving the representativeness and accuracy of different agricultural products. It also calculates the priority of each dimension to obtain the optimal dimension order, thereby obtaining the principal component dimensions most relevant to the crop's growth state after dimensionality reduction. Adjustments are made based on the ecological parameter data of these dimensions. This operation avoids the gradual decrease in the proportion of principal components formed by any dimension during each dimensionality reduction process, which could lead to misjudgments of the main influencing ecological parameters, thus improving the accuracy of ecological parameter regulation and the management efficiency of greenhouses.

[0059] Example 2:

[0060] Corresponding to the modern agricultural smart greenhouse ecological control system provided in the above embodiments, based on the same technical concept, this invention also provides a modern agricultural smart greenhouse ecological control system, which is used to execute the aforementioned modern agricultural smart greenhouse ecological control system. Figure 2 To illustrate the structural diagram of another modern agricultural smart greenhouse ecological regulation system according to various embodiments of the present invention, as shown below. Figure 2As shown at the hardware level, the modern agricultural smart greenhouse ecological control system includes a processor, and optionally, an internal bus, a network interface, and memory. The memory may include main memory, such as high-speed random-access memory (RAM), or it may also include non-volatile memory, such as at least one disk drive. Of course, this modern agricultural smart greenhouse ecological control system may also include other hardware required for its operation.

[0061] The processor, network interface, and memory can be interconnected via an internal bus, which can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, this diagram uses only a single bidirectional arrow, but it does not imply that there is only one bus or one type of bus.

[0062] Memory is used to store programs. Specifically, programs can include program code, which includes computer operation commands. Memory can include main memory and non-volatile memory, and it provides instructions and data to the processor.

[0063] The processor reads the corresponding computer program from non-volatile memory into main memory and then executes it, forming a device at the logical level that is assigned to a specific user. The processor executes the program stored in memory and specifically performs the following: Figure 1 The methods disclosed in the embodiments shown achieve the functions and beneficial effects of the methods in the preceding method embodiments, and will not be repeated here.

[0064] It should be noted that the modern agricultural smart greenhouse ecological regulation system provided in this embodiment of the invention is based on the same application concept as the modern agricultural smart greenhouse ecological regulation system provided in this embodiment of the invention. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned modern agricultural smart greenhouse ecological regulation system, and has the same or similar beneficial effects. Repeated parts will not be repeated.

[0065] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0066] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A modern agricultural wisdom greenhouse ecological regulation system, characterized in that, The modern agricultural wisdom greenhouse ecological regulation system comprises: An acquisition module is configured to acquire multi-dimensional ecological parameters of crops in a greenhouse from a sowing start time to a current time; A determination module is configured to determine potential influence trends of each dimension of the ecological parameters on the crops in a time period between adjacent extreme value points of the ecological parameters according to lengths of the time period, corresponding ecological parameters, and ideal adaptation ranges of the ecological parameters in each dimension; The determination module is further configured to determine a common relationship between matching time periods of any two ecological parameters according to differences between the potential influence trends in the matching time periods of the any two ecological parameters, the potential influence trends of each time period of current ecological parameters, and lengths of the time periods, determine a dimension reduction priority of the any two ecological parameters based on dimension reduction parameters of the any two ecological parameters and the common relationship, and determine a principal component dimension according to the dimension reduction priority of the any two ecological parameters; A regulation module is configured to regulate an ecological environment of the crops based on the principal component dimension; The determination module is further configured to: determine a time period between adjacent minimum value points of the ecological parameters; determine a first number of time points at which the ecological parameters are lower than a minimum value of the ideal adaptation range in the time period, and a second number of time points at which the ecological parameters exceed a maximum value of the ideal adaptation range in the time period; determine a maximum value in the time period and a first average of the ecological parameters that exceed the maximum value of the ideal adaptation range in the time period; determine the potential influence trends according to lengths of the time period, the first number, the second number, the maximum value in the time period, and the first average; and a calculation formula of the potential influence trends of any one ecological parameter on a current type of crops in a current time period is: in the above formulae, a potential impact trend of the i-th time period on the current type of crop, representing any one of the ecological parameters, a length of the i-th time period, a first number of time points in the i-th time period below the minimum value of the ideal adaptation range, a second number of time points in the i-th time period above the maximum value of the ideal adaptation range, a maximum value in the i-th time period, a first mean value of the ecological parameter in the i-th time period in which the ecological parameter exceeds the maximum value of the ideal adaptation range; or, determine a time period between adjacent maximum value points of the ecological parameters; determine a first number of time points at which the ecological parameters are lower than a minimum value of the ideal adaptation range in the time period, and a second number of time points at which the ecological parameters exceed a maximum value of the ideal adaptation range in the time period; determine a minimum value in the time period and a second average of the ecological parameters that are lower than the minimum value of the ideal adaptation range in the time period; determine the potential influence trends according to lengths of the time period, the first number, the second number, the minimum value in the time period, and the second average; and a calculation formula of the potential influence trends of any one ecological parameter on a current type of crops in a current time period is: In the above formulae, denotes the minimum value of the ecological parameter in the i-th time interval, denotes the second mean value of the ecological parameter which is lower than the minimum value of the ideal adaptation range of the ecological parameter in the i-th time interval; The determination module is further configured to: determine matching time periods of any two ecological parameters based on a dynamic time warping algorithm, and determine differences between the potential influence trends in the matching time periods of the any two ecological parameters; determine a common relationship between the matching time periods of the any two ecological parameters according to the potential influence trends of each time period of current ecological parameters, lengths of the time periods, and the differences; and or, determining an average impact trend of potential impact trends of each time period of the current ecological parameter and other ecological parameters except the current ecological parameter, and a length average of all time periods of the current ecological parameter and other ecological parameters except the current ecological parameter; superimposing differences between potential impact trends between each of the matching time periods to obtain superimposed differences; determining the common relationship according to the average impact trend, the length average and the superimposed differences.

2. The modern agricultural wisdom greenhouse ecological regulation and control system according to claim 1, characterized in that, The determining module is further configured to: determine a dynamic time warping distance between any two time periods of any two ecological parameters based on the dynamic time warping algorithm; select two time periods corresponding to a minimum value of the dynamic time warping distance as matching time periods of the any two ecological parameters. 3.The modern agricultural wisdom greenhouse ecological regulation and control system according to claim 1, characterized in that, The determining module is further configured to: normalize a product of a dimension reduction parameter of any two ecological parameters and the common relationship based on a range normalization algorithm to obtain a dimension reduction priority of the any two ecological parameters.

4. The modern agricultural wisdom greenhouse ecological regulation and control system according to claim 1, characterized in that, The determining module is further configured to: determine an optimal dimension arrangement order of principal component analysis according to the dimension reduction priority of the any two ecological parameters and a dimension reduction dimension of the principal component analysis; use ecological data of each dimension in the optimal dimension arrangement order as a principal component dimension.

5. The modern agricultural wisdom greenhouse ecological regulation and control system according to claim 1, characterized in that, The regulating module is further configured to: determine an ideal adaptation range corresponding to each of the principal component dimensions; regulate an ecological environment of the crops related to the principal component dimensions so that an ecological parameter of the principal component dimension is within the ideal adaptation range.

6. A modern agricultural wisdom greenhouse ecological regulation system, comprising: A processor and a memory; the memory is configured to store a computer program executable on the processor; the processor is configured to execute the program stored on the memory to implement the following steps: obtain multi-dimensional ecological parameters of crops in a greenhouse from a sowing starting time to a current time; determine a potential impact trend of each dimension of the ecological parameters on the crops in the time period between adjacent extreme value points of the ecological parameters of each dimension according to a length of the time period and corresponding ecological parameters and an ideal adaptation range of the ecological parameters of each dimension; determine a common relationship between matching time periods of any two ecological parameters according to differences between potential impact trends between the matching time periods of the any two ecological parameters, potential impact trends of each time period of the current ecological parameter and lengths of the time periods; determine a dimension reduction priority of the any two ecological parameters based on a dimension reduction parameter of the any two ecological parameters and the common relationship; determine a principal component dimension according to the dimension reduction priority of the any two ecological parameters, and regulate an ecological environment of the crops based on the principal component dimension; The potential impact trend acquisition includes: determining a time period between adjacent minimum value points of the ecological parameters; determining a first number of time points at which the ecological parameter is lower than a minimum value of the ideal adaptation range in the time period, and a second number of time points at which the ecological parameter exceeds a maximum value of the ideal adaptation range in the time period; determining a maximum value in the time period and a first average of the ecological parameter exceeding the maximum value of the ideal adaptation range in the time period. determining the potential influence trend according to the length of the time period, the first quantity, the second quantity, the maximum value in the time period and the first average value; the formula for calculating the potential influence trend of any one ecological parameter on the current type of crop in the current time period is: in the above formulae, a potential impact trend of the i-th time period on the current type of crop, representing any one of the ecological parameters, a length of the i-th time period, a first number of time points in the i-th time period below the minimum value of the ideal adaptation range, a second number of time points in the i-th time period above the maximum value of the ideal adaptation range, a maximum value in the i-th time period, a first mean value of the ecological parameter in the i-th time period in which the ecological parameter exceeds the maximum value of the ideal adaptation range; or, determining the time period between adjacent maximum points of the ecological parameter; determining the first quantity of time points in the time period at which the ecological parameter is lower than the minimum value of the ideal adaptation range, and the second quantity of time points in the time period at which the ecological parameter exceeds the maximum value of the ideal adaptation range; determining the minimum value in the time period and the second average value of the ecological parameter in the time period at which the ecological parameter is lower than the minimum value of the ideal adaptation range; determining the potential influence trend according to the length of the time period, the first quantity, the second quantity, the minimum value in the time period and the second average value; the formula for calculating the potential influence trend of any one ecological parameter on the current type of crop in the current time period is: In the above formula, denotes the minimum value of the ecological parameter in the i-th time interval, denotes the second mean value of the ecological parameter which is lower than the minimum value of the ideal adaptation range of the ecological parameter in the i-th time interval. the acquisition of the common relationship includes: determining the matching time period of any two ecological parameters based on the dynamic time warping algorithm, and determining the difference between the potential influence trends between the matching time periods of the any two ecological parameters; determining the common relationship between the matching time periods of the any two ecological parameters according to the potential influence trends of each time period of the current ecological parameter, the length of the time period and the difference; or, determining the average influence trend of the potential influence trends of each time period of the current ecological parameter and other ecological parameters except the current ecological parameter, and the average length of all time periods of the current ecological parameter and other ecological parameters except the current ecological parameter; superimposing the differences between the potential influence trends between each of the matching time periods to obtain a superimposed difference; determining the common relationship according to the average influence trend, the average length and the superimposed difference.

Images