Tomato greenhouse planting environment regulation and control method and system based on big data

Abstract

The invention discloses a tomato greenhouse planting environment regulation and control method and system based on big data, and relates to the technical field of greenhouse environment regulation and control. The growth degree of crops is constructed, and if the growth degree is lower than expected environmental condition data in a greenhouse, the abnormal degree of environmental parameters is constructed; and constructing a condition risk degree of the greenhouse according to the abnormity degree of each environment parameter, if the condition risk degree exceeds the expectation, performing regression analysis on the growth environment parameter and the growth degree of the crop, determining a corresponding importance level according to the importance degree of each environment parameter, and adjusting the environment parameters in the greenhouse according to the importance level of each parameter. Acquiring a corresponding environment adjusting scheme; and testing the environment regulation schemes, constructing the efficiency of executing the environment regulation schemes according to test data, and screening the environment regulation schemes. By screening a plurality of schemes, when the optimal scheme is used for adjusting the environmental parameters in the greenhouse, the adjusting effect can be better.

Description

Technical Field

[0001] The present invention relates to the technical field of greenhouse environment regulation, and specifically to a method and system for regulating the tomato greenhouse planting environment based on big data. Background Technique

[0002] Greenhouse planting is a modern agricultural technology that realizes plant growth through scientific management methods under artificially controlled environmental conditions such as temperature, humidity, and light. It can be divided into types such as glass greenhouses, plastic greenhouses, and solar greenhouses according to different building structures and materials. In greenhouse planting, various soilless cultivation modes are widely used, such as coconut coir soilless cultivation and rockwool soilless cultivation. Coconut coir is currently the most ideal organic cultivation substrate, with excellent air-water ratio and bulk density, renewable, sustainable, and low usage cost, suitable for large-scale promotion and use. Rockwool cultivation is to plant plants in pre-made rockwool and can be used for large-scale production in combination with irrigation, drainage and other device accessories. In greenhouse planting, especially when planting tomatoes, when controlling the environmental conditions in the greenhouse, in order to improve the control effect and accuracy, big data technology is usually involved, and with the assistance of big data technology, the greenhouse planting effect is improved.

[0003] In the Chinese invention patent with the application publication number CN117389355A, a method and system for intelligent greenhouse temperature control for tomato planting set multiple monitoring stages for the greenhouse of tomatoes through a control system, monitor the greenhouse environmental status at different time nodes of each monitoring stage through a sensor group, and set a preset temperature for each monitoring stage; preprocess and group the environmental data to obtain different category groups; train a temperature prediction model by combining multiple data to form a training set, obtain temperature prediction data through the model, set a difference threshold and a temperature balance value according to a formula, and calculate the difference between the temperature prediction data and the preset temperature; compare the difference with the difference threshold, and judge whether to adjust the current temperature to the temperature balance value at the adjustment speed or continuously monitor the difference according to the comparison result.

[0004] Combined with the above application and the content in the prior art:

[0005] When high-value cash crops are planted in a greenhouse, in order to keep the crops in a good growth state, it is usually necessary to monitor various environmental parameters in the greenhouse. After setting an alarm threshold, it is judged whether the environmental parameters are abnormal. If there is an abnormality, an alarm is sent to the outside, and after the alarm, corresponding treatment measures are taken in time to adaptively adjust the environmental parameters. However, in this adjustment method, in order to reduce the alarm frequency, the set alarm threshold is usually very wide, and an alarm is sent to the outside only when the abnormality degree of the environmental parameters is very high. This easily leads to the situation that although the environmental parameters in the greenhouse are not abnormal, they are actually not the optimal growth environment for the crops. Therefore, the growth state of the crops fails to reach the expected level. At the same time, when the greenhouse environmental parameters are abnormal, it is also necessary for the management personnel to adjust various environmental parameters according to experience, which not only has poor efficiency but also is not easy to adjust to the optimal state.

[0006] Therefore, the present invention provides a method and system for regulating the tomato greenhouse planting environment based on big data. Summary of the Invention

[0007] (1) Technical Problems to be Solved

[0008] Aiming at the deficiencies of the prior art, the present invention provides a method and system for regulating the tomato greenhouse planting environment based on big data. By adjusting the environmental parameters in the greenhouse according to the importance level of each parameter, the corresponding environmental adjustment plan is obtained; the environmental adjustment plan is tested, and the efficiency degree of implementing the environmental adjustment plan is constructed from the test data, and the environmental adjustment plan is screened. By screening several plans, when the best plan among them is used to adjust the environmental parameters in the greenhouse, the adjustment effect can be better, thus solving the technical problems in the background art.

[0009] (2) Technical Solutions

[0010] To achieve the above object, the present invention is realized through the following technical solutions:

[0011] A method for regulating the tomato greenhouse planting environment based on big data includes monitoring the growth state of the crops in the greenhouse, constructing the growth degree Kp of the crops from the monitoring data, and sending an adjustment instruction to the outside if the growth degree Kp is lower than the expected value;

[0012] Collecting the environmental condition data in the greenhouse, constructing the abnormality degree Yd of the environmental parameters, and then constructing the conditional risk degree Fx of the greenhouse from the abnormality degree Yd of each environmental parameter. If the conditional risk degree Fx exceeds the expected value, a correlation analysis instruction is sent to the outside;

[0013] Perform a regression analysis on the growth environment parameters and growth degree of crops, determine the corresponding importance level Lo according to the importance degree of each environmental parameter, and adjust the environmental parameters in the greenhouse according to the importance level Lo of each parameter to obtain the corresponding environmental adjustment plan; among them, according to the abnormality degree Yd and influence factor Zo of each environmental parameter, and after normalization processing, calculate the corresponding control importance level Lo for each environmental parameter, and the method is as follows:

[0014]

[0015] Weight coefficient: 0 ≤ ρ ≤ 1, 0 ≤ ζ ≤ 1;

[0016] Use the trained crop growth model to test the environmental adjustment plan, construct the efficiency degree Xo of implementing the environmental adjustment plan from the test data, and screen the environmental adjustment plan according to the efficiency degree Xo to obtain the target plan and execute it.

[0017] Furthermore, divide the greenhouse into several sub-regions, randomly select the target crops in each sub-region; monitor the growth state parameters of the target crops, obtain the growth data at the current stage, and after summarizing the obtained growth data, construct a crop growth state set.

[0018] Furthermore, construct the growth degree Kp of the crop from the crop growth state set. Among them, perform linear normalization processing on the plant height Zg and the number of leaves Ys of the crop, and map the corresponding data values to the interval [0, 1] according to the following method:

[0019]

[0020] Weight coefficient: 0 ≤ F1 ≤ 1, 0 ≤ F2 ≤ 1 and F2 + F1 = 1; i = 1, 2,..., k, k is the number of target crops; Zg i is the plant height of the i-th target crop, Zg avg is the average value of the plant height, is the qualified standard value of the plant height; Ys i is the number of leaves of the i-th target crop, Ys avg is the average value of the number of leaves, is the qualified standard value of the number of leaves.

[0021] Furthermore, after receiving the adjustment instruction, select detection points in each sub-region, collect the environmental condition factors in the greenhouse at the detection points, average the data collected at each detection point, and summarize the averaged collected data to construct an environmental condition set; construct the abnormality degree Yd of the environmental parameters from the environmental condition set. Among them, perform normalization processing on the environmental parameter F according to the following method:

[0022]

[0023] where \(i = 1, 2, \ldots, p\), \(p\) is the number of environmental parameter acquisitions, \(F\) i is the value of the \(i\)-th environmental parameter, \(F\) avg is the corresponding mean value, and the weight coefficients: \(0\leq F_1\leq1\), \(0\leq F_2\leq1\), and \(F_1 + F_2 = 1\).

[0024] Furthermore, based on the abnormality degree \(Y_d\) of each environmental parameter in the greenhouse, the conditional risk degree \(F_x\) of the greenhouse is constructed as follows:

[0025]

[0026] where \(i = 1, 2, \ldots, m\), \(m\) is the number of types of environmental parameters, \(Y_d\) avg is the mean value of the abnormality degree, \(Y_d\) i is the abnormality degree of the \(i\)-th type of environmental parameter.

[0027] Furthermore, after receiving the correlation analysis instruction, using the environmental conditions in the greenhouse as independent variables and the growth degree of the crop as the dependent variable, a linear regression analysis is carried out, and the corresponding regression equation is constructed, and the regression coefficient is used as the influence factor of each environmental parameter; taking adjusting the growth state of the crop as the optimization goal, according to the importance level \(Lo\) of each parameter, the trained optimization model is used to adjust the environmental parameters in the greenhouse in turn. After obtaining several environmental adjustment schemes in the greenhouse, a test instruction is sent to the outside.

[0028] Furthermore, a crop growth model is trained using sample data. After receiving the test instruction, the trained crop growth model is used to test the environmental adjustment scheme, the corresponding test data is obtained, and the test data is summarized to construct a test data set; after obtaining the crop growth prediction data at each prediction node, the corresponding growth degree is constructed from the prediction data.

[0029] Furthermore, the growth degrees before and after implementing the environmental adjustment scheme are aligned one by one, and then the efficiency degree \(X_o\) of implementing the environmental adjustment scheme is constructed. The efficiency degree \(X_o\) is used to label each environmental adjustment scheme, and the one with the highest efficiency degree is used as the target scheme, and the second highest is used as the backup scheme, and the target scheme is executed to adjust the environment in the greenhouse.

[0030] Furthermore, the way to obtain the efficiency degree \(X_o\) is as follows:

[0031]

[0032] where \(T_y\) i is the intermediate value of the efficiency degree at the \(i\)-th node, \(T_y\) avg is its mean value, \(i\) is the node serial number, \(i = 1, 2, \ldots, n\), \(n\) is the number of nodes, \(AK_p\) i and \(BK_p\)i The growth lengths on the i-th node before and after optimization, AKp avg and BKp avg are the corresponding means.

[0033] A tomato greenhouse planting environment regulation method and system based on big data, including:

[0034] A growth state monitoring unit that monitors the growth state of crops in the greenhouse, constructs the growth length Kp of the crops from the monitoring data, and sends an adjustment instruction to the outside if the growth length Kp is lower than expected;

[0035] A greenhouse risk analysis unit that collects environmental condition data in the greenhouse, constructs the abnormality degree Yd of environmental parameters, and then constructs the conditional risk degree Fx of the greenhouse from the abnormality degrees Yd of each environmental parameter. If the conditional risk degree Fx exceeds the expectation, it sends a correlation analysis instruction to the outside;

[0036] An environmental parameter optimization unit that performs a regression analysis on the growth environment parameters of the crops and the growth length, determines the corresponding importance level Lo according to the importance degree of each environmental parameter, and adjusts the environmental parameters in the greenhouse according to the importance level Lo of each parameter to obtain the corresponding environmental adjustment plan;

[0037] A plan screening unit that tests the environmental adjustment plan using the trained crop growth model, constructs the efficiency degree Xo of implementing the environmental adjustment plan from the test data, screens the environmental adjustment plan with the efficiency degree Xo, and obtains and executes the target plan.

[0038] (III) Beneficial effects

[0039] The present invention provides a tomato greenhouse planting environment regulation method and system based on big data, having the following beneficial effects:

[0040] 1. The growth state of the current crops is described and evaluated by the growth length. If the growth length of the current crops exceeds the expectation, by monitoring the growth state of the crops in the greenhouse, when the growth state of the crops is poor, an early warning can be issued in time, which can avoid the deterioration of the growth state of the crops, thereby ensuring the normal growth of the crops.

[0041] 2. Whether the growth environment of the crops in the greenhouse is abnormal is judged by the constructed conditional risk degree. If there is an abnormality, the environmental parameters in the greenhouse can be adjusted in time. By constructing the abnormality degree of the environmental parameters, whether each parameter is abnormal is judged, so that targeted processing can also be achieved during processing.

[0042] 3. Evaluate and judge the importance of each environmental parameter. When it is necessary to adjust the environment in the greenhouse, based on the abnormal degree of each environment, calculate and obtain the importance level Lo of each environmental parameter, and determine the weight for adjusting the environmental parameter according to the determined importance level Lo of each parameter, so that the environmental adjustment plan is more in line with the actual situation.

[0043] 4. Based on the obtained change in the growth length after testing, construct the efficiency degree of each environmental adjustment plan from the change state, evaluate and screen the effects of each environmental adjustment plan, so as to obtain the corresponding target plan. By screening several plans, when using the best plan to adjust the environmental parameters in the greenhouse, the adjustment effect can be better. If the effect of the environmental adjustment plan is poor, the environmental adjustment plan can also be optimized. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic flowchart of the method for regulating the tomato greenhouse planting environment of the present invention;

[0045] Figure 2 It is a schematic structural diagram of the tomato greenhouse planting environment regulation system of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0046] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0047] Please refer to Figure 1 , the present invention provides a method for regulating the tomato greenhouse planting environment based on big data, including:

[0048] Step 1. Monitor the growth state of the crops in the greenhouse, and construct the growth length Kp of the crops from the monitoring data. If the growth length Kp is lower than the expected value, send an adjustment instruction to the outside;

[0049] The content of the above Step 1 includes:

[0050] Step 101. Divide the greenhouse into several sub-regions, and randomly select target crops in each sub-region; monitor the growth state parameters of the target crops, obtain the growth data at the current stage, including: the plant height and the number of leaves of the crops, etc. After summarizing the obtained growth data, construct a crop growth state set;

[0051] Step 102: Construct the growth degree Kp of the crop from the set of crop growth states. Among them, perform linear normalization on the plant height Zg and the number of leaves Ys of the crop, and map the corresponding data values to the interval [0, 1] according to the following method:

[0052]

[0053] Weight coefficient: 0 ≤ F1 ≤ 1, 0 ≤ F2 ≤ 1 and F2 + F1 = 1; i = 1, 2, …, k, where k is the number of target crops; Zg i is the plant height of the i-th target crop, and Zg avg is the average value of the plant height, is the qualified standard value of the plant height; Ys i is the number of leaves of the i-th target crop, and Ys avg is the average value of the number of leaves, is the qualified standard value of the number of leaves; The weight coefficient is obtained by referring to the analytic hierarchy process;

[0054] The calculation formula of the growth degree Kp quantifies the overall growth status of the current crop by combining multiple crop growth indicators. For example, perform linear normalization (map to the 0 - 1 interval) on the plant height and the number of leaves of each plant respectively, and then perform weighted summation according to the preset weight coefficient (for example, determine the weights of each indicator through the analytic hierarchy process) to obtain the "growth degree" value reflecting the overall growth level of the crop. The technical role of this growth degree index is to comprehensively evaluate the crop growth status: compared with the prior art that only monitors a single indicator or relies on empirical judgment, this solution uses a formula to fuse multi-dimensional growth data into an intuitive value, enabling the system to objectively judge whether the crop growth meets the standard. When the growth degree is lower than the preset threshold, an alarm can be issued in a timely manner (send an adjustment instruction) to avoid missing the overall poor growth situation due to minor abnormalities in each single indicator, thus having higher sensitivity and accuracy compared with the prior art that relies on experience or loose thresholds;

[0055] According to historical data and the growth management expectations of the crops in the greenhouse, preset the growth degree threshold in advance; if the growth status of the crop lags behind the expectation, that is, when the growth degree is lower than the growth degree threshold, it indicates that the current crop growth status is poor and the growth status of the crop needs to be adjusted. At this time, send an adjustment instruction to the outside;

[0056] When in use, combine the content in Steps 101 and 102:

[0057] Before adjusting and controlling the growth environment of crops in a greenhouse, monitor the growth status of the crops, construct a corresponding growth degree from the monitoring data, and use the growth degree to describe and evaluate the current growth status of the crops. If the growth degree of the current crops exceeds the expectation, it indicates that the current environmental conditions in the greenhouse are good and conducive to the growth of the crops, and no adjustment is needed. On the contrary, if the growth degree is lower than expected, it is necessary to adjust the environmental parameters in the greenhouse, including soil humidity, temperature, etc. By monitoring the growth status of the crops in the greenhouse, when the growth status of the crops is poor, an early warning can be issued in a timely manner, which can avoid the deterioration of the growth status of the crops and thus ensure the normal growth of the crops.

[0058] Combined with the above application and the content in the prior art:

[0059] When high-value cash crops are planted in a greenhouse, in order to keep the crops in a good growth state, it is usually necessary to monitor various environmental parameters in the greenhouse. After setting an alarm threshold, it is judged whether the environmental parameters are abnormal. If there is an abnormality, an alarm is sent to the outside, and after the alarm, corresponding treatment measures are taken in a timely manner to adaptively adjust the environmental parameters. However, in this adjustment method, in order to reduce the alarm frequency, the set alarm threshold is usually very broad, and an alarm is sent to the outside only when the abnormal degree of the environmental parameters is very high. This is likely to result in the situation that although the environmental parameters in the greenhouse are not abnormal, they are actually not the optimal growth environment for the crops, so the growth state of the crops fails to reach the expectation. At the same time, when the environmental parameters in the greenhouse are abnormal, it is also necessary for the management personnel to adjust various environmental parameters based on experience, which not only has poor efficiency but also is not easy to adjust to the optimal state.

[0060] Step 2: Collect the environmental condition data in the greenhouse, construct the abnormality degree Yd of the environmental parameters, and then construct the conditional risk degree Fx of the greenhouse from the abnormality degree Yd of each environmental parameter. If the conditional risk degree Fx exceeds the expectation, send a correlation analysis instruction to the outside;

[0061] The content of the above Step 2 includes the following:

[0062] Step 201: After receiving the adjustment instruction, select detection points in each sub-region, collect the environmental condition factors in the greenhouse at the detection points, average the data collected at each detection point, and obtain the averaged data, including soil humidity and indoor temperature in the greenhouse, etc. Summarize the averaged collected data to construct an environmental condition set;

[0063] Step 202: Construct the abnormality degree Yd of the environmental parameters from the environmental condition set. Among them, perform normalization processing on the environmental parameter F according to the following method:

[0064]

[0065] where \(i = 1, 2, \ldots, p\), \(p\) is the number of environmental parameter acquisitions, \(F\) i is the value of the \(i\)-th environmental parameter, \(F\) avg is the corresponding mean value, and the weight coefficients: \(0\leq F1\leq1\), \(0\leq F2\leq1\), and \(F1 + F2 = 1\); the values of the weight coefficients are the same as the previous values;

[0066] The calculation formula of the abnormality degree \(Y_d\) is used to quantitatively evaluate the deviation degree of each environmental parameter in the greenhouse from the normal state, that is, the "abnormality degree". Specifically, for each environmental parameter (such as soil humidity, air temperature, etc.), its real-time monitoring value can be compared with the historical average value or the qualified standard value of this parameter, and the abnormality degree value in the range of 0 to 1 is obtained through linear normalization calculation (the closer the value is to 1, the more serious the deviation from the normal value). When necessary, the data of multiple acquisition points can be weighted and averaged to obtain the overall abnormal value. This calculation logic enables the system to continuously depict the abnormality degree of environmental parameters, rather than simply alarming when exceeding a fixed threshold. Its technical effect is that through the abnormality degree, subtle environmental deviations can be detected early, and it can be judged which specific environmental factor is abnormal. Compared with the method of triggering an alarm using a preset threshold in the prior art, the abnormality degree formula of the present invention improves the sensitivity of abnormality detection, and even a medium-degree deviation of environmental parameters can be quantified, thus supporting subsequent targeted adjustments;

[0067] Step 203, construct the conditional risk degree \(F_x\) of the greenhouse according to the abnormality degree \(Y_d\) of each environmental parameter in the greenhouse, in the following manner:

[0068]

[0069] where \(i = 1, 2, \ldots, m\), \(m\) is the number of types of environmental parameters, \(Y_d\) avg is the mean value of the abnormality degree, \(T_d\) i is the abnormality degree of the \(i\)-th type of environmental parameter;

[0070] If the conditional risk degree \(F_x\) of the greenhouse exceeds the preset abnormality degree threshold, it indicates that there may be certain abnormalities in the environmental conditions in the greenhouse, which may affect the growth of crops in the greenhouse. At this time, a correlation analysis instruction is sent to the outside;

[0071] The calculation formula of the risk degree Fx constructs an index reflecting the overall environmental risk level of the greenhouse, namely the "conditional risk degree", by summarizing the abnormality degrees of various environmental parameters. For example, the average value, weighted sum or other statistical methods of the abnormality degrees of each parameter can be used to calculate the overall risk degree. The average value of the abnormality degrees of each parameter in the formula is used to describe the overall deviation degree, and the abnormality degree deviations of each environmental parameter jointly contribute to the risk degree value. Its function is to comprehensively evaluate the abnormal risks of the greenhouse environment: when multiple environmental parameters all have small abnormalities, the risk degree will increase cumulatively, thus reminding that the overall greenhouse environment is no longer conducive to crop growth. Compared with the existing technology that only focuses on whether a single parameter exceeds the limit, the formula comprehensively quantifies various environmental abnormalities, avoiding the defect that the overall risk may be missed when each parameter is judged independently. Once the conditional risk degree exceeds the preset threshold, the system will trigger a correlation analysis instruction to prepare for further finding out the key factors affecting crop growth;

[0072] When in use, combine the content in steps 201 to 203:

[0073] When the growth state of the crops in the greenhouse is abnormal, for example, when the growth state is relatively backward, monitor the environmental conditions in the greenhouse, and construct the abnormality degree of each environmental parameter from the monitoring data. On this basis, construct the conditional risk degree in the greenhouse. At this time, use the constructed conditional risk degree to determine whether the growth environment of the crops in the greenhouse is abnormal. If it is abnormal, the environmental parameters in the greenhouse can be adjusted in time. At the same time, by constructing the abnormality degree of the environmental parameters, judge whether each parameter is abnormal, so that targeted processing can also be achieved during processing.

[0074] Step 3: Conduct a regression analysis on the growth environment parameters and growth degree of the crops, determine the corresponding importance level Lo according to the importance degree of each environmental parameter, and adjust the environmental parameters in the greenhouse according to the importance level Lo of each parameter to obtain the corresponding environmental adjustment plan;

[0075] The said step 3 includes the following content:

[0076] Step 301: After receiving the correlation analysis instruction, conduct a linear regression analysis with the environmental conditions in the greenhouse as the independent variable and the growth degree of the crops as the dependent variable, judge the influence of each environmental parameter on the growth degree of the crops, and construct the corresponding regression equation, using the regression coefficient as the influence factor of each environmental parameter;

[0077] Step 302: According to the abnormality degree Yd and influence factor Zo of each environmental parameter, and after normalization, calculate the corresponding control importance level Lo for each environmental parameter in the following way:

[0078]

[0079] Weight coefficients: 0 ≤ ρ ≤ 1, 0 ≤ ζ ≤ 1;

[0080] Step 303: Construct a conditional optimization model using a linear optimization algorithm, with adjusting the growth state of the crop as the optimization goal. Based on the importance level Lo of each parameter, use the trained optimization model to sequentially adjust the environmental parameters in the greenhouse. After obtaining several environmental adjustment schemes in the greenhouse, send a test command to the outside;

[0081] The calculation formula for controlling the importance level Lo is used to calculate the control importance level (i.e., adjustment priority) of each environmental parameter in the regulation. Its core is to determine the importance based on the current abnormal degree of each parameter combined with its influence degree (influence factor) on crop growth. When calculating, the abnormality degree and the corresponding influence factor of the parameter can be normalized respectively, and then added according to a certain weight coefficient (for example, each with a 50% weight, and the sum of the two weights is 1) to obtain the control importance level of each parameter. The importance level obtained in this way reflects information on both the "current deviation situation" and the "contribution to crop growth" of the parameter. Its technical effect is to realize the intelligent evaluation of the adjustment priority order of environmental parameters: parameters with a large abnormal degree and a large influence on crop growth will obtain a higher importance level and be adjusted first; on the contrary, parameters that are abnormal but have a small influence can reduce the priority. Compared with the method of manual experience adjustment or one-size-fits-all treatment of each parameter in the prior art, the scientificity and pertinence of environmental regulation are improved through this formula, ensuring that the adjustment resources are preferentially used for the link with the greatest impact on growth, thereby optimizing the overall regulation efficiency;

[0082] When in use, combine the content in Steps 301 to 303:

[0083] When the environmental conditions in the greenhouse already have abnormalities, analyze the correlation between each environmental parameter and the crop growth state, and thereby obtain the influence degree of each environmental parameter on crop growth. Thus, it is possible to evaluate and judge the importance degree of each environmental parameter. When it is necessary to adjust the environment in the greenhouse, based on the abnormal degree of each environment, calculate and obtain the importance level Lo of each environmental parameter, and determine the weight when adjusting the environmental parameter according to the determined importance level Lo of each parameter, so that the environmental adjustment scheme is more in line with the actual situation.

[0084] Step Four: Use the trained crop growth model to test the environmental adjustment scheme, construct the efficiency degree Xo of implementing the environmental adjustment scheme from the test data, screen the environmental adjustment scheme with the efficiency degree Xo, and obtain and execute the target scheme;

[0085] The content of the above Step Four includes the following:

[0086] Step 401: Collect the growth status data and growth environment data of the crops in the greenhouse. Use the collected data as sample data to construct an initial model with a neural convolutional network. Train the initial model with the sample data to obtain a trained crop growth model. After receiving a test instruction, use the trained crop growth model to test the environmental regulation plan, obtain the corresponding test data, and summarize the test data to construct a test data set.

[0087] Step 402: After obtaining the crop growth prediction data at each prediction node, construct the corresponding growth degree from the prediction data. Align the growth degrees before and after executing the environmental regulation plan one by one, and then construct the efficiency degree Xo of executing the environmental regulation plan as follows:

[0088]

[0089] where Ty i is the intermediate value of the efficiency degree at the i-th node, Ty avg is its mean value, i is the node serial number, i = 1, 2,..., n, n is the number of nodes, AKp i and BKp i are the growth degrees at the i-th node before and after optimization respectively, AKp avg and BKp avg are the corresponding mean values.

[0090] Mark each environmental regulation plan with the efficiency degree Xo, take the one with the highest efficiency degree as the target plan, and the second highest as the backup plan. Execute the target plan to adjust the environment in the greenhouse.

[0091] The calculation formula of the efficiency degree Xo defines the "efficiency degree" index, which is used to quantitatively evaluate the improvement effect of the environmental regulation plan on crop growth. Its calculation logic is to compare the crop growth degrees before and after the implementation of the plan one by one: calculate the improvement degree (such as the difference or growth ratio) of the crop growth degree after implementing the plan relative to before implementation (in the case of no optimization) at each prediction node, and then statistically process the improvement values of all nodes (such as taking the average value, etc.) to obtain the total efficiency degree value. The intermediate value of the efficiency degree and its mean value at each node in the formula are used to comprehensively measure the overall improvement amplitude. The higher the efficiency degree, the more significant the growth promotion effect brought by the plan during the entire growth cycle. Through the efficiency degree Xo, multiple environmental regulation plans can be objectively evaluated and screened. This approach technically ensures that the selected target plan can maximize the improvement of the crop growth state, rather than relying on manual subjective judgment. Compared with the existing technology that lacks a unified evaluation index and often requires repeated trial and error, the introduction of the efficiency degree formula makes the selection of the optimization plan well-founded and the effect is clearly improved.

[0092] When in use, combine the content in Steps 401 and 402:

[0093] After obtaining multiple environmental adjustment plans, each environmental adjustment plan is tested, and corresponding test data is obtained after simulated execution. Then, a corresponding growth length is constructed from the test data. Based on the changes in the growth length obtained after the test, the efficiency degree of each environmental adjustment plan is constructed from the change state, so as to evaluate and screen the effects of each environmental adjustment plan, thereby obtaining the corresponding target plan. Thus, by screening several plans, when the best plan among them is used to adjust the environmental parameters in the greenhouse, the adjustment effect can be better. If the effect of the environmental adjustment plan is poor, the environmental adjustment plan can be optimized;

[0094] Among them, the crop growth length is constructed from the prediction data: when the trained crop growth model tests the environmental adjustment plan, it will output the crop growth prediction data at each prediction time node. For example, at each prediction node, the model can give the predicted values of the growth state parameters such as the plant height and the number of leaves of the crop. In order to be consistent with the actual growth state evaluation, the corresponding growth length is calculated for these prediction data using the same method as in step 102: that is, the predicted plant height and the number of leaves are linearly normalized respectively (based on references such as the mean and standard values of the training samples) to obtain the standardized growth indicators, and then weighted calculation is performed according to the predetermined weight coefficients to obtain the predicted growth length value corresponding to this node. In this way, the crop growth situation predicted by the model is transformed into a growth length index under a unified scale, making the subsequent comparison of the effects of different plans comparable;

[0095] For each environmental adjustment plan, two sequences of growth length data will be generated during the model test stage, namely before and after the implementation of the plan. The growth length sequence before the implementation of the plan can be sourced from the baseline scenario (for example, the growth length predicted by the model under the condition of maintaining the original environmental parameters unchanged, or the future growth length calculated based on the existing environmental conditions), and the growth length sequence after the implementation of the plan is sourced from the growth length predicted by the model under the condition of applying this adjustment plan. The system aligns these two sequences one by one according to the time nodes: that is, the "unadjusted growth length" at the i-th prediction node is paired and compared with the "growth length after adjustment" at the i-th prediction node. On this basis, the growth improvement value at each node is calculated. For example, using "growth length after adjustment minus growth length before adjustment" or "percentage increase after adjustment compared to before adjustment" as the intermediate value of the efficiency degree. Then, the intermediate values of all nodes are statistically summarized (for example, taking the average value) to obtain the overall efficiency degree index of this plan. The efficiency degree calculated in this way intuitively reflects the comprehensive improvement effect brought by this adjustment plan to the crop growth compared with the non-adjusted situation. Based on this, each candidate environmental adjustment plan is marked and compared, and the plan with the highest efficiency degree is selected as the target plan for execution, and the second highest is used as the backup plan.

[0096] Please refer to Figure 2, the present invention provides a tomato greenhouse planting environment regulation system based on big data, including:

[0097] A growth status monitoring unit that monitors the growth status of crops in the greenhouse, constructs the growth degree Kp of the crops from the monitoring data, and sends an adjustment instruction to the outside if the growth degree Kp is lower than expected;

[0098] A greenhouse risk analysis unit that collects environmental condition data in the greenhouse, constructs the abnormality degree Yd of environmental parameters, and then constructs the conditional risk degree Fx of the greenhouse from the abnormality degree Yd of each environmental parameter. If the conditional risk degree Fx exceeds the expectation, it sends a correlation analysis instruction to the outside;

[0099] An environmental parameter optimization unit that performs a regression analysis on the growth environment parameters of the crops and the growth degree, determines the corresponding importance level Lo according to the importance degree of each environmental parameter, and adjusts the environmental parameters in the greenhouse according to the importance level Lo of each parameter to obtain the corresponding environmental regulation plan;

[0100] A plan screening unit that tests the environmental regulation plan using the trained crop growth model, constructs the efficiency degree Xo of implementing the environmental regulation plan from the test data, screens the environmental regulation plan with the efficiency degree Xo, and obtains and executes the target plan.

[0101] It should be noted that:

[0102] The analytic hierarchy process, abbreviated as AHP, is a decision-making method. Its core idea is to decompose the elements related to the decision-making problem into levels such as goals, criteria, and plans, and conduct qualitative and quantitative analyses on this basis. This method is suitable for dealing with target systems with hierarchical and interleaved evaluation indicators, especially when the target value is difficult to quantitatively describe.

[0103] In decision analysis, it decomposes the problem into different constituent factors, and according to the mutual correlation and influence between factors and the subordination relationship, aggregates and combines the factors into different levels to form a multi-level analysis structure model, so that the problem is finally reduced to the determination of the relative importance weights of the lowest level (plans, measures, etc. for decision-making) relative to the highest level (total goal) or the ranking of relative advantages and disadvantages.

[0104] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0105] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only for some logical function divisions. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0106] As described above, the foregoing is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

[0107] As described above, the foregoing is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application.

Claims

1. A method for regulating the tomato greenhouse planting environment based on big data, characterized in that: Including, monitor the growth status of crops in the greenhouse, construct the growth degree Kp of the crops from the monitoring data, and send an adjustment instruction to the outside if the growth degree Kp is lower than the expectation; collect the environmental condition data in the greenhouse, construct the abnormality degree Yd of the environmental parameters, and then construct the conditional risk degree Fx of the greenhouse from the abnormality degree Yd of each environmental parameter. If the conditional risk degree Fx exceeds the expectation, send a correlation analysis instruction to the outside; conduct a regression analysis on the growth environmental parameters and the growth degree of the crops, determine the corresponding importance level Lo according to the importance degree of each environmental parameter, and adjust the environmental parameters in the greenhouse according to the importance level Lo of each parameter to obtain the corresponding environmental adjustment plan; among them, according to the abnormality degree Yd and the influence factor Zo of each environmental parameter, and after normalization processing, calculate the corresponding control importance level Lo for each environmental parameter, and the method is as follows: Weight coefficient: 0 ≤ ρ ≤ 1, 0 ≤ ζ ≤ 1; Use the trained crop growth model to test the environmental adjustment plan, construct the efficiency degree Xo of implementing the environmental adjustment plan from the test data, screen the environmental adjustment plan with the efficiency degree Xo, and obtain and execute the target plan.

2. The method for regulating the tomato greenhouse planting environment based on big data according to claim 1, characterized in that: Divide the greenhouse into several sub-regions, and randomly select target crops in each sub-region; monitor the growth status parameters of the target crops, obtain the growth data at the current stage, and after summarizing the obtained growth data, construct a crop growth status set.

3. The method for regulating the tomato greenhouse planting environment based on big data according to claim 2, characterized in that: Construct the growth degree Kp of the crops from the crop growth status set, among which, perform linear normalization processing on the plant height Zg and the number of leaves Ys of the crops, and map the corresponding data values to the interval [0, 1], according to the following method: Weight coefficient: 0 ≤ F1 ≤ 1, 0 ≤ F2 ≤ 1 and F2 + F1 = 1; i = 1, 2, …, k, where k is the number of target crops; Zg i is the plant height of the i-th target crop, Zg avg is the average value of the plant height, is the qualified standard value of the plant height; Ys i is the number of leaves of the i-th target crop, Ys avg is the average value of the number of leaves, is the qualified standard value of the number of leaves.

4. The method for regulating the tomato greenhouse planting environment based on big data according to claim 1, characterized in that: After receiving the adjustment instruction, select detection points in each sub-region, collect the environmental condition factors in the greenhouse at the detection points, average the data collected at each detection point, and summarize the averaged collected data to construct an environmental condition set; construct the abnormality degree Yd of the environmental parameters from the environmental condition set, among which, perform normalization processing on the environmental parameter F, according to the following method: where \(i = 1, 2, \ldots, p\), \(p\) is the number of environmental parameter acquisitions, \(F\) i is the value of the \(i\)-th environmental parameter, \(F\) avg is the corresponding mean value, and the weight coefficients: \(0 \leq F_1 \leq 1\), \(0 \leq F_2 \leq 1\), and \(F_1 + F_2 = 1\).

5. The method for regulating the tomato greenhouse planting environment based on big data according to claim 1, characterized in that: Construct the conditional risk degree Fx of the greenhouse according to the abnormality degree Yd of each environmental parameter in the greenhouse, and the method is as follows: where \(i = 1, 2, \ldots, m\), \(m\) is the number of types of environmental parameters, \(Y_d\) avg is the mean of the anomaly degree, \(Y_d\) i is the anomaly degree of the \(i\)-th type of environmental parameter.

6. The method for regulating the tomato greenhouse planting environment based on big data according to claim 5, characterized in that: After receiving the correlation analysis instruction, taking the environmental conditions in the greenhouse as the independent variable and the growth degree of the crop as the dependent variable, a linear regression analysis is carried out, and the corresponding regression equation is constructed, with the regression coefficient as the influencing factor of each environmental parameter; taking adjusting the growth state of the crop as the optimization goal, according to the importance level Lo of each parameter, the trained optimization model is used to adjust the environmental parameters in the greenhouse in turn. After obtaining several environmental adjustment schemes in the greenhouse, a test instruction is sent to the outside.

7. The method for regulating the tomato greenhouse planting environment based on big data according to claim 6, characterized in that: The crop growth model is trained and obtained using sample data. After receiving the test instruction, the trained crop growth model is used to test the environmental adjustment scheme, the corresponding test data is obtained, and the test data is summarized to construct a test data set; after obtaining the crop growth prediction data at each prediction node, the corresponding growth degree is constructed from the prediction data.

8. The method for regulating the tomato greenhouse planting environment based on big data according to claim 7, characterized in that: The growth degrees before and after implementing the environmental adjustment scheme are aligned one by one, and then the efficiency degree Xo of implementing the environmental adjustment scheme is constructed. Each environmental adjustment scheme is marked by the efficiency degree Xo, the one with the highest efficiency degree is used as the target scheme, and the second highest is used as the backup scheme. The target scheme is executed to adjust the environment in the greenhouse.

9. The method for regulating the tomato greenhouse planting environment based on big data according to claim 8, characterized in that: The acquisition method of the efficiency degree Xo is as follows: Among them, Ty i is the intermediate value of the efficiency degree on the i-th node, and Ty avg is its average value. i is the node serial number, i = 1, 2, …, n, where n is the number of nodes. AKp i and BKp i are the growth lengths on the i-th node before and after optimization respectively, and AKp avg and BKp avg are the corresponding average values.

10. A tomato greenhouse planting environment regulation system based on big data, characterized in that: Including: A growth state monitoring unit monitors the growth state of the crops in the greenhouse, and constructs the growth degree Kp of the crops from the monitoring data. If the growth degree Kp is lower than the expectation, a regulation instruction is sent to the outside; A greenhouse risk analysis unit collects the environmental condition data in the greenhouse, constructs the abnormality degree Yd of the environmental parameters, and then constructs the condition risk degree Fx of the greenhouse from the abnormality degree Yd of each environmental parameter. If the condition risk degree Fx exceeds the expectation, a correlation analysis instruction is sent to the outside; An environmental parameter optimization unit conducts a regression analysis on the crop growth environmental parameters and the growth degree, determines the corresponding importance level Lo according to the importance degree of each environmental parameter, and adjusts the environmental parameters in the greenhouse according to the importance level Lo of each parameter to obtain the corresponding environmental adjustment scheme; A scheme screening unit uses the trained crop growth model to test the environmental adjustment scheme, constructs the efficiency degree Xo of implementing the environmental adjustment scheme from the test data, screens the environmental adjustment scheme with the efficiency degree Xo, and obtains and executes the target scheme.

Images