Plant factory environmental control system and method based on artificial intelligence

Abstract

The present invention relates to the field of plant strategy management, and in particular to an artificial intelligence-based plant factory environmental control system and method, comprising an environment extraction module, which models the plant factory according to the environmental parameters and plant growth state; an experience pool formation module, which divides the historical cultivation strategy of the plant factory into multiple experience layers to form a multi-layer experience pool; a strategy determination module, which sets the plant target growth state, determines the intelligent control strategy and the experience control strategy of the plant growth stage; a strategy adjustment module, which dynamically adjusts the intelligent control strategy and the experience control strategy using a dual-path mechanism to determine the target control strategy; a growth state acquisition module, which regulates the environment according to the adjustment instruction to obtain the actual growth state; and an experience pool update module, which performs difference analysis on the actual growth state and the target growth state, judges the growth difference stage, determines the difference factor, and updates the content of the multi-layer experience pool based on the difference factor. Production efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of plant strategy management, and in particular to an artificial intelligence-based plant factory environment control system and method. Background Art

[0002] Plant growth is affected by a variety of complex environmental factors. It is still a difficult challenge to achieve the goal of predicting plant growth by setting the relationship between environmental factors.

[0003] A patent document with Chinese patent publication number CN107515531A discloses an intelligent control method and system, and an intelligent monitoring system for a plant factory. The method includes: obtaining an error value between a monitoring parameter and a set parameter of a plant growth environment; using a Q learning algorithm to adjust the proportional gain, integral gain, and differential gain of a PID controller; and outputting a control command for a control system of the plant growth environment based on the error value and the adjusted proportional gain, integral gain, and differential gain.

[0004] In the prior art, the error value between the monitoring parameters and the set parameters of the plant growth environment is obtained, and the control command of the control system of the plant growth environment is output based on the error value. There is no flexible strategy for real-time adjustment, which leads to the problem of low production efficiency. Summary of the invention

[0005] To this end, the present invention provides a plant factory environmental control system and method based on artificial intelligence, which can utilize a dual-channel mechanism to dynamically adjust the intelligent control strategy and the empirical control strategy to flexibly cope with the real-time changing environment and thus solve the problem of low production efficiency.

[0006] To achieve the above objectives, the present invention provides, on one hand, a plant factory environmental control system based on artificial intelligence, the system comprising:

[0007] The environment extraction module is used to model the plant factory based on the current environmental parameters of the plant factory and the growth status of the plants to obtain the plant factory model;

[0008] An experience pool forming module, for dividing the historical cultivation strategy of the plant factory into multiple experience layers according to the gap of historical growth parameters and the confidence of the strategy, so as to form a multi-layer experience pool;

[0009] A strategy determination module, connected to the environment extraction module and the experience pool forming module respectively, for determining the intelligent control strategy for each growth stage of the plant under the target growth state based on the target growth state of the plant in the set plant factory model, and for screening the multi-layer experience pool based on the strategy confidence and target matching degree to obtain the experience control strategy, wherein the intelligent control strategy is the minimum energy consumption strategy for achieving the target growth state, and the experience control strategy is the actual cultivation strategy for achieving the target growth state;

[0010] A strategy adjustment module, connected to the strategy determination module, for dynamically adjusting the intelligent control strategy and the empirical control strategy using a dual-path mechanism to face the real-time changing plant factory environment, thereby determining a target control strategy;

[0011] A growth state acquisition module, connected to the strategy adjustment module, for regulating the environment of the plant factory based on an adjustment instruction to obtain the actual growth state of the plant, wherein the adjustment instruction is generated based on the target control strategy and the environmental parameters of the current plant factory;

[0012] The experience pool updating module is connected to the strategy adjustment module and the growth state acquisition module respectively, and is used to determine the difference factors between the actual growth state and the target growth state, and to update the content of the experience layer based on the difference factors to improve the multi-layer experience pool data.

[0013] Furthermore, the environment extraction module includes:

[0014] An environmental parameter collection unit, which is used to determine the environmental parameters of the plant factory based on the sensor information collected from the plant factory;

[0015] A growth status acquisition unit, connected to the environmental parameter collection unit, for collecting plant images according to the visual sensor to determine the growth status of the plant;

[0016] The model building unit is connected to the growth status acquisition unit and is used to create a plant factory model based on input data to simulate the growth process of plants, wherein the input data includes the environmental parameters and the growth status.

[0017] Furthermore, the experience pool forming module includes:

[0018] The identification unit identifies the gap between the confidence of each training strategy and the historical growth parameters based on historical data;

[0019] The experience pool forming unit divides the historical training strategy into an expert experience layer, a key information experience layer and a general information experience layer based on the confidence and the gap to form a multi-layer experience pool.

[0020] Furthermore, the strategy determination module includes:

[0021] Data collection unit, which uses IoT sensors and image acquisition systems to collect plant factory environmental data and plant growth status data;

[0022] An intelligent strategy determination unit is connected to the data acquisition unit, and is used to input the plant factory environment data and plant growth status data into an intelligent algorithm, perform calculation and analysis on the target growth status as output data, and determine an intelligent control strategy based on the analysis results, wherein the intelligent control strategy is a minimum energy consumption strategy for achieving the target growth status.

[0023] Furthermore, the strategy determination module also includes:

[0024] A sample extraction unit, used to extract target information samples from each layer of experience pool according to the strategy confidence;

[0025] The empirical strategy determination unit is connected to the sample extraction unit and matches the information sample according to the target matching degree to determine the empirical control strategy.

[0026] Furthermore, the strategy adjustment module monitors the environmental parameters and plant growth status in the plant factory in real time. If an abnormal situation occurs, the empirical control strategy in the dual-path mechanism is activated, otherwise the intelligent control strategy is executed to dynamically adjust the intelligent control strategy and the empirical control strategy to determine the target control strategy.

[0027] Furthermore, the growth status acquisition module includes:

[0028] A data acquisition unit, used to collect environmental parameters through sensors;

[0029] An environmental control unit, used for analyzing the environmental parameters and the target control strategy to generate an environmental control instruction, and sending the environmental control instruction to the environmental control equipment of the plant factory for control;

[0030] The growth status acquisition unit analyzes the plants using image recognition technology according to the environmental control results to obtain the actual growth status of the plants.

[0031] Furthermore, the experience pool updating module includes:

[0032] A difference analysis unit compares and analyzes the actual growth state and the target growth state to obtain a difference value;

[0033] A growth difference stage judgment unit, which judges the growth difference stage by using an algorithm according to the difference value, analyzes the plants in the growth difference stage, and identifies the difference factors;

[0034] The strategy formulation unit formulates corresponding adjustment strategies according to the difference factors, and updates the corresponding strategy content in the multi-layer experience pool to improve the multi-layer experience pool.

[0035] Furthermore, the difference analysis unit is a difference value between a target maturity cycle of a plant and an actual maturity cycle of a plant and a difference value between a target maturity yield and an actual maturity yield.

[0036] On the other hand, the present invention also provides a plant factory environment control method based on artificial intelligence, the method comprising:

[0037] Modeling is performed based on the current environmental parameters of the plant factory and the growth status of the plants to obtain a plant factory model;

[0038] Dividing the historical cultivation strategy of the plant factory into multiple experience layers according to the gap between historical growth parameters and the confidence of the strategy to form a multi-layer experience pool;

[0039] Setting a target growth state of the plants in the plant factory model to determine an intelligent control strategy for each growth stage of the plants under the target growth state, wherein the intelligent control strategy is a strategy with the lowest energy consumption for achieving the target growth state, and screening the multi-layer experience pool according to the strategy confidence and target matching degree to obtain an experience control strategy, wherein the experience control strategy is an actual cultivation strategy for achieving the target growth state;

[0040] The intelligent control strategy and the empirical control strategy are dynamically adjusted by using a dual-path mechanism to face the real-time changing plant factory environment, thereby determining a target control strategy;

[0041] Generate adjustment instructions according to the target control strategy and the current environmental parameters of the plant factory to regulate the environment of the plant factory to obtain the actual growth status of the plants;

[0042] The difference factors between the actual growth state and the target growth state are determined, and the content of the experience layer is updated based on the difference factors to improve the multi-layer experience pool data.

[0043] Compared with the prior art, the beneficial effect of the present invention is that the environment extraction module establishes a plant factory model by acquiring the environmental parameters of the plant factory and the growth status of the plant, and more accurately understands the interaction between the plant and the environment. The experience pool formation module forms a multi-layer experience pool by layered sampling of historical cultivation strategies, which can quickly retrieve relevant historical experience and improve the efficiency of strategy formulation. The strategy determination module determines the intelligent control strategy and the experience control strategy by setting the target growth state of the plant for the plant factory model, thereby improving the accuracy and timeliness of decision-making. The growth state acquisition module applies the target control strategy determined by combining the intelligent control strategy and the experience control strategy to the plant factory model, generates adjustment instructions for environmental regulation, so as to increase the possibility of the plant reaching the target growth state, thereby improving production efficiency. The experience pool update module updates the content of the multi-layer experience pool by performing a difference analysis on the actual growth state and the target growth state, so that the strategy in the experience pool can continuously adapt to the new environment and conditions, thereby improving the accuracy of the system strategy.

[0044] In particular, by collecting environmental parameters and obtaining growth status to establish a plant factory model, the growth process of plants can be simulated more accurately, thereby improving the accuracy of predicting plant growth and avoiding waste and excessive consumption of resources.

[0045] In particular, the historical training strategies can be finely classified through experience layer division, so that relevant experience can be selected more accurately as a reference when formulating strategies, which helps to improve the effectiveness of strategy formulation and reduce the cost of trial and error.

[0046] In particular, by collecting plant growth environment data and growth status data, intelligent control strategies can be formulated to optimize planting conditions, improve plant growth rate and quality, help shorten the growth cycle and improve production efficiency.

[0047] In particular, by extracting samples that are highly relevant to the target growth state from a multi-layer experience pool, the time for processing irrelevant information can be reduced, and more focus can be placed on information that has a direct impact on the target state, thereby improving implementation efficiency. The information samples are analyzed to determine the empirical control strategy, thereby improving the scientificity and accuracy of decision-making.

[0048] In particular, the addition of empirical control strategies makes the system more stable when facing emergencies and complex environments, and can ensure the continuity and stability of production. The application of the dual-pathway mechanism provides support for the further development of intelligent agricultural technology.

[0049] In particular, precise regulation based on real-time environmental parameters and target control strategies ensures that the plant's growth environment always remains in the best state, thereby improving the growth rate and quality of the plants. It can quickly and accurately obtain a large amount of plant growth status information, improving detection efficiency and accuracy.

[0050] In particular, timely analysis and resolution of problems that arise during plant growth and updating of corresponding control strategies can help reduce planting risks caused by environmental unsuitability or improper management and ensure the healthy growth of plants.

[0051] In particular, by improving the plant maturity cycle, planting plans can be reasonably arranged, land use efficiency and crop turnover can be improved. Mature yield analysis not only focuses on whether the yield reaches the target, but also achieves improvements in yield and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A schematic diagram of the structure of a plant factory environmental control system based on artificial intelligence provided by an embodiment of the present invention;

[0053] Figure 2 A schematic diagram of the structure of an environment extraction module of an artificial intelligence-based plant factory environment control system provided by an embodiment of the present invention;

[0054] Figure 3 A schematic diagram of the structure of a strategy determination module of an artificial intelligence-based plant factory environmental control system provided in an embodiment of the present invention;

[0055] Figure 4 A schematic diagram of the structure of a growth status acquisition module of an artificial intelligence-based plant factory environmental control system provided in an embodiment of the present invention;

[0056] Figure 5 A schematic diagram of the structure of an experience pool update module of an artificial intelligence-based plant factory environmental control system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0058] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0059] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be understood as a limitation on the present invention.

[0060] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0061] See also Figure 1 As shown, the present invention provides an artificial intelligence-based plant factory environment control system, the system comprising:

[0062] The environment extraction module 10 is used to model the plant factory according to the current environmental parameters of the plant factory and the growth status of the plants to obtain a plant factory model;

[0063] An experience pool forming module 20, for dividing the historical cultivation strategy of the plant factory into multiple experience layers according to the gap of historical growth parameters and the confidence of the strategy, so as to form a multi-layer experience pool;

[0064] A strategy determination module 30, connected to the environment extraction module and the experience pool forming module, respectively, for determining the intelligent control strategy for each growth stage of the plant under the target growth state based on the target growth state of the plant in the set plant factory model, and for screening the multi-layer experience pool based on the strategy confidence and target matching degree to obtain the experience control strategy, wherein the intelligent control strategy is the minimum energy consumption strategy for achieving the target growth state, and the experience control strategy is the actual cultivation strategy for achieving the target growth state;

[0065] A strategy adjustment module 40, connected to the strategy determination module, is used to dynamically adjust the intelligent control strategy and the empirical control strategy using a dual-path mechanism to face the real-time changing plant factory environment, thereby determining a target control strategy;

[0066] A growth state acquisition module 50, connected to the strategy adjustment module, is used to adjust the environment of the plant factory based on the adjustment instruction to obtain the actual growth state of the plant, wherein the adjustment instruction is generated based on the target control strategy and the environmental parameters of the current plant factory;

[0067] The experience pool updating module 60 is connected to the strategy adjustment module and the growth state acquisition module respectively to determine the difference factors between the actual growth state and the target growth state, and updates the content of the experience layer based on the difference factors to improve the multi-layer experience pool data.

[0068] Specifically, the current environmental parameters and plant growth status of the plant factory are modeled to simulate the growth process of the plant. The historical cultivation strategy of the plant factory is divided into multiple experience layers (expert experience layer, key information experience layer and general information experience layer) according to the gap between historical growth parameters and the confidence of the strategy to form a multi-layer experience pool. The target growth state of the plant in the plant factory model (plant maturity cycle and mature yield) is set, and the intelligent control strategy of each growth stage of the plant is determined by an intelligent algorithm. The intelligent control strategy includes the adjustment of environmental parameters (temperature and humidity values, light intensity and carbon dioxide concentration, etc.) and the nutrient solution concentration ratio. Then, the target information samples of the target growth state are extracted from each experience layer through the multi-layer experience pool, and the information samples are analyzed to determine the experience control strategy. The intelligent control strategy and the experience control strategy are converted to each other through a dual-channel mechanism to determine the target control strategy. According to the target control strategy, the adjustment instructions are generated to perform environmental regulation of the plant factory model to obtain the actual growth state of the plant; the difference between the actual growth state and the target growth state is analyzed, and the content of the multi-layer experience pool is updated according to the analysis results to improve the multi-layer experience pool data.

[0069] Specifically, the environment extraction module establishes a plant factory model by acquiring the environmental parameters of the plant factory and the growth status of the plants, so as to more accurately understand the interaction between plants and the environment. The experience pool formation module forms a multi-layer experience pool by stratifying and sampling the historical cultivation strategies, which can quickly retrieve relevant historical experiences and improve the efficiency of strategy formulation. The strategy determination module determines the intelligent control strategy and the empirical control strategy by setting the target growth state of the plants for the plant factory model, thereby improving the accuracy and timeliness of decision-making. The growth state acquisition module applies the target control strategy determined by combining the intelligent control strategy and the empirical control strategy to the plant factory model, generates adjustment instructions for environmental regulation, so as to increase the possibility of plants reaching the target growth state, thereby improving production efficiency. The experience pool update module updates the content of the multi-layer experience pool by performing a difference analysis between the actual growth state and the target growth state, so that the strategies in the experience pool can continuously adapt to new environments and conditions, thereby improving the accuracy of system strategies.

[0070] See also Figure 2 As shown, the environment extraction module 10 includes:

[0071] An environmental parameter collection unit 11, which is used to determine the environmental parameters of the plant factory based on the sensor information collected from the plant factory;

[0072] A growth status acquisition unit 12, connected to the environmental parameter collection unit, for collecting plant images according to the visual sensor to determine the growth status of the plant;

[0073] The model building unit 13 is connected to the growth status acquisition unit and is used to create a plant factory model based on input data to simulate the growth process of plants, wherein the input data includes the environmental parameters and the growth status.

[0074] Specifically, the environmental parameter collection unit determines the environmental parameters of the plant factory by collecting temperature sensors, humidity sensors, light intensity sensors and CO2 concentration sensors. The growth condition acquisition unit uses visual sensors to capture the growth images of plants and analyzes the growth characteristics of plants (leaf color, morphology, growth rate, flowering and fruiting, etc.) through image processing and visual recognition, thereby determining the growth status of plants. The model building unit is used to receive data from the environmental parameter collection unit and the growth condition acquisition unit as input, and create a plant factory model to simulate the growth process of plants under different environmental conditions.

[0075] Specifically, by collecting environmental parameters and obtaining growth status to establish a plant factory model, the growth process of plants can be simulated more accurately, thereby improving the accuracy of predicting plant growth and avoiding waste and excessive consumption of resources.

[0076] Specifically, the experience pool forming module includes:

[0077] The identification unit identifies the gap between the confidence of each training strategy and the historical growth parameters based on historical data;

[0078] The experience pool forming unit divides the historical training strategy into an expert experience layer, a key information experience layer and a general information experience layer based on the confidence and the gap to form a multi-layer experience pool.

[0079] Specifically, confidence is an indicator to measure the reliability of a strategy, which is usually based on the success rate and stability of the strategy in past implementations. The expert experience layer contains training strategies with the highest confidence and the smallest gap. The key information experience layer contains training strategies with higher confidence but certain gaps. Although these strategies perform well, there may still be some room for improvement or areas that require special attention. The general information experience layer contains training strategies with lower confidence or larger gaps, but they still have certain reference value.

[0080] Specifically, these strategies are divided into different experience levels (expert experience layer, key information experience layer and general information experience layer). Among them, the expert experience layer contains planting strategies that have been verified for a long time, have significant effects and are widely recognized; the key information experience layer contains strategies that perform well under certain specific conditions and are of great significance for solving specific problems; the general information experience layer contains a large number of basic and general planting strategies as supplements and references.

[0081] Specifically, historical training strategies are finely classified through experience layer division, so that relevant experience can be selected more accurately as a reference when formulating strategies, which helps to improve the effectiveness of strategy formulation and reduce trial and error costs.

[0082] See also Figure 3 As shown, the strategy determination module 30 includes:

[0083] The data collection unit 31 collects plant factory environment data and plant growth status data using an Internet of Things sensor and an image acquisition system;

[0084] The intelligent strategy determination unit 32 is connected to the data collection unit, and is used to input the plant factory environment data and plant growth status data into the intelligent algorithm, and perform calculation and analysis on the target growth status as output data, and determine the intelligent control strategy according to the analysis results, wherein the intelligent control strategy is the minimum energy consumption strategy for achieving the target growth status.

[0085] Specifically, the intelligent strategy determination unit is used to receive the environmental data and growth status data and input them into a pre-designed intelligent algorithm. The intelligent algorithm uses machine learning or deep learning to perform in-depth analysis. According to the analysis results, the intelligent strategy determination unit can automatically generate a set of minimum energy consumption control strategies for the current situation. The minimum energy consumption control strategy may include adjusting environmental parameters such as light intensity, temperature, humidity and CO2 concentration, as well as providing specific water and fertilizer management measures to enable it to reach or exceed the target growth state (maturity cycle and mature yield) as soon as possible.

[0086] Specifically, by collecting plant growth environment data and growth status data, intelligent control strategies can be formulated to optimize planting conditions, improve plant growth rate and quality, help shorten the growth cycle and improve production efficiency.

[0087] Specifically, the strategy determination module also includes:

[0088] A sample extraction unit, used to extract target information samples from each layer of experience pool according to the strategy confidence;

[0089] The empirical strategy determination unit is connected to the sample extraction unit and matches the information sample according to the target matching degree to determine the empirical control strategy.

[0090] Specifically, the sample extraction unit extracts target information samples from the multi-layer experience pool according to the target growth state. The target information samples are planting strategies and experience data that have successfully achieved or approached the target growth state under similar growth conditions in the past. The sample extraction unit analyzes the specific requirements of the target growth state (maturity cycle and mature yield) through intelligent algorithms, and searches for matching or highly relevant strategy samples in each layer of experience pools (expert experience layer, key information experience layer, and general information experience layer).

[0091] Specifically, the empirical strategy determination unit is used to receive information samples extracted from each layer of the experience pool. The core function of this unit is to analyze the information samples, and analyze multiple factors that need to be considered, such as the success rate of the samples, the similarity of the applicable scenarios, the historical performance of the strategy, etc. Based on the analysis results, the empirical control strategy is determined. The empirical control strategy not only includes specific planting operation suggestions, but also may include environmental parameter adjustment plans, water and fertilizer management optimization measures, etc., aiming to guide the actual planting process.

[0092] Specifically, by extracting samples that are highly relevant to the target growth state from a multi-layer experience pool, the time for processing irrelevant information can be reduced, and more focus can be placed on information that has a direct impact on the target state, thereby improving implementation efficiency. The information samples are analyzed to determine the empirical control strategy, thereby improving the scientificity and accuracy of decision-making.

[0093] Specifically, the strategy adjustment module monitors the environmental parameters and plant growth status in the plant factory in real time. If an abnormal situation occurs, the empirical control strategy in the dual-path mechanism is activated, otherwise the intelligent control strategy is executed to dynamically adjust the intelligent control strategy and the empirical control strategy to determine the target control strategy.

[0094] Specifically, the intelligent control strategy and the empirical control strategy are run simultaneously through a dual-channel mechanism, and both are dynamically adjusted according to real-time environmental data. During the adjustment process, if a large number of pests suddenly appear on the plants or an emergency situation occurs, the empirical control strategy will be switched immediately. The strategy determination unit will also evaluate the performance of the two strategies in the current environment, and comprehensively consider factors such as planting objectives, resource utilization efficiency, and cost-effectiveness to determine the target control strategy.

[0095] Specifically, the addition of empirical control strategies makes the system more stable when facing emergencies and complex environments, and can ensure the continuity and stability of production. The application of the dual-channel mechanism provides strong support for promoting the further development of agricultural intelligent technology.

[0096] See also Figure 4 As shown, the growth status acquisition module 50 includes:

[0097] A data acquisition unit 51, used to collect environmental parameters through sensors;

[0098] An environmental control unit 52, connected to the data acquisition unit, is used to analyze according to the environmental parameters and the target control strategy to generate an environmental control instruction, and send the environmental control instruction to the environmental control equipment of the plant factory for control;

[0099] The growth status acquisition unit 53 is connected to the environment control unit, and uses image recognition technology to analyze the plant according to the environment control result to obtain the actual growth status of the plant.

[0100] Specifically, the environmental control unit is responsible for analyzing and calculating the environmental parameters and target control strategies provided by the data acquisition unit, generating corresponding environmental adjustment instructions, and sending them to the environmental control equipment in the plant factory (such as lighting system, temperature and humidity control system, ventilation system, etc.) to achieve the control of environmental parameters. The environmental control unit continuously optimizes the control strategy to ensure that the plants are always in the most suitable growth environment.

[0101] Specifically, precise regulation is carried out according to real-time environmental parameters and target control strategies to ensure that the growth environment of the plants always remains in the best state, thereby improving the growth rate and quality of the plants. It can quickly and accurately obtain a large amount of plant growth status information, improving detection efficiency and accuracy.

[0102] See also Figure 5 As shown, the experience pool updating module 60 includes:

[0103] A difference analysis unit 61 compares and analyzes the actual growth state and the target growth state to obtain a difference value;

[0104] A growth difference stage judgment unit 62 is connected to the difference analysis unit, and judges the growth difference stage by using an algorithm according to the difference value, analyzes the plants in the growth difference stage, and identifies the difference factors;

[0105] The strategy formulation unit 63 is connected to the growth difference stage judgment unit, formulates corresponding adjustment strategies according to difference factors, and updates corresponding strategy contents in the multi-layer experience pool to improve the multi-layer experience pool.

[0106] Specifically, the growth difference stage judgment unit receives the difference values ​​from the difference analysis unit, and uses a preset algorithm to analyze these difference values ​​to determine the growth difference stage of the plant. This process not only takes into account the size of the difference value, but also combines multiple factors such as the plant's growth cycle and physiological characteristics. By deeply analyzing the state and changes of the plant itself during the growth difference stage, the unit can identify the key factors that lead to growth differences, such as environmental factors, nutritional deficiencies, and pests and diseases.

[0107] Specifically, the strategy formulation unit formulates corresponding adjustment strategies based on the difference factors identified by the growth difference stage judgment unit. These strategies involve the adjustment of environmental parameters (such as light intensity, temperature, humidity, etc.), optimization of nutrient supply, implementation of pest and disease control measures, etc. At the same time, the strategy formulation unit will also compare and update these newly formulated strategies with the contents of the multi-layer experience pool to continuously improve and optimize the multi-layer experience pool.

[0108] Specifically, by timely analyzing and solving problems that arise during plant growth and updating corresponding control strategies, it helps reduce planting risks caused by environmental unsuitability or improper management and ensures the healthy growth of plants.

[0109] Specifically, the difference analysis unit is the difference value between the target maturity cycle of the plant and the actual maturity cycle of the plant and the difference value between the target maturity yield and the actual maturity yield.

[0110] Specifically, by improving the plant maturity cycle, planting plans can be reasonably arranged, land use efficiency and crop turnover rate can be improved. Mature yield analysis not only focuses on whether the yield reaches the target, but also achieves improvements in yield and quality.

[0111] Specifically, the plant factory environment control method based on artificial intelligence provided by the embodiment of the present invention includes:

[0112] Modeling is performed based on the current environmental parameters of the plant factory and the growth status of the plants to obtain a plant factory model;

[0113] Dividing the historical cultivation strategy of the plant factory into multiple experience layers according to the gap between historical growth parameters and the confidence of the strategy;

[0114] Setting a target growth state of the plants in the plant factory model to determine an intelligent control strategy for each growth stage of the plants under the target growth state, wherein the intelligent control strategy is a strategy with the lowest energy consumption for achieving the target growth state, and screening the multi-layer experience pool according to the strategy confidence and target matching degree to obtain an experience control strategy, wherein the experience control strategy is an actual cultivation strategy for achieving the target growth state;

[0115] The intelligent control strategy and the empirical control strategy are dynamically adjusted by using a dual-path mechanism to face the real-time changing plant factory environment, thereby determining a target control strategy;

[0116] Generate adjustment instructions according to the target control strategy and the current environmental parameters of the plant factory to regulate the environment of the plant factory to obtain the actual growth status of the plants;

[0117] The difference factors between the actual growth state and the target growth state are determined, and the content of the experience layer is updated based on the difference factors to improve the multi-layer experience pool data.

[0118] The method in the embodiment of the present invention is applied in the above-mentioned artificial intelligence-based plant factory environmental control system, and can achieve the same technical effect, which will not be repeated here.

[0119] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0120] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A plant factory environmental control system based on artificial intelligence, characterized in that: include: The environment extraction module is used to model the plant factory based on the current environmental parameters of the plant factory and the growth status of the plants to obtain the plant factory model; An experience pool forming module, for dividing the historical cultivation strategies of the plant factory into multiple experience layers according to the gaps between historical growth parameters and the confidence of the strategies, so as to form a multi-layer experience pool; A strategy determination module, connected to the environment extraction module and the experience pool forming module respectively, for determining the intelligent control strategy for each growth stage of the plant under the target growth state based on the target growth state of the plant in the set plant factory model, and for screening the multi-layer experience pool based on the strategy confidence and target matching degree to obtain the experience control strategy, wherein the intelligent control strategy is the minimum energy consumption strategy for achieving the target growth state, and the experience control strategy is the actual cultivation strategy for achieving the target growth state; A strategy adjustment module, connected to the strategy determination module, for dynamically adjusting the intelligent control strategy and the empirical control strategy using a dual-path mechanism to face the real-time changing plant factory environment, thereby determining a target control strategy; A growth state acquisition module, connected to the strategy adjustment module, for regulating the environment of the plant factory based on an adjustment instruction to obtain the actual growth state of the plant, wherein the adjustment instruction is generated based on the target control strategy and the environmental parameters of the current plant factory; An experience pool updating module, connected to the strategy adjustment module and the growth state acquisition module respectively, for determining the difference factors between the actual growth state and the target growth state, and updating the content of the experience layer based on the difference factors to improve the multi-layer experience pool data; The strategy determination module includes: Data collection unit, which uses IoT sensors and image acquisition systems to collect plant factory environmental data and plant growth status data; an intelligent strategy determination unit, connected to the data collection unit, for inputting the plant factory environment data and plant growth state data into an intelligent algorithm, performing operation analysis on the target growth state as output data, and determining an intelligent control strategy according to the analysis result, wherein the intelligent control strategy is a minimum energy consumption strategy for achieving the target growth state; The strategy determination module also includes: A sample extraction unit, used to extract target information samples from each layer of experience pool according to the strategy confidence; An empirical strategy determination unit, connected to the sample extraction unit, matches the target information sample according to the target matching degree to determine the empirical control strategy; The strategy adjustment module monitors the environmental parameters and plant growth status in the plant factory in real time. If an abnormal situation occurs, the empirical control strategy in the dual-path mechanism is activated, otherwise the intelligent control strategy is executed to dynamically adjust the intelligent control strategy and the empirical control strategy to determine the target control strategy.

2. The plant factory environmental control system based on artificial intelligence according to claim 1 is characterized in that: The environment extraction module comprises: An environmental parameter collection unit, which is used to determine the environmental parameters of the plant factory based on the collected sensor information of the plant factory; A growth status acquisition unit, connected to the environmental parameter collection unit, for collecting plant images according to the visual sensor to determine the growth status of the plant; The model building unit is connected to the growth status acquisition unit and is used to create a plant factory model based on input data to simulate the growth process of plants, wherein the input data includes the environmental parameters and the growth status.

3. The plant factory environmental control system based on artificial intelligence according to claim 2 is characterized in that: The experience pool forming module comprises: The identification unit identifies the gap between the confidence of each training strategy and the historical growth parameters based on historical data; The experience pool forming unit divides the historical training strategy into an expert experience layer, a key information experience layer and a general information experience layer based on the confidence and the gap to form a multi-layer experience pool.

4. The plant factory environmental control system based on artificial intelligence according to claim 1 is characterized in that: The growth status acquisition module includes: A data acquisition unit, used to collect environmental parameters through sensors; An environmental control unit, connected to the data acquisition unit, is used to analyze according to the environmental parameters and the target control strategy to generate an environmental control instruction, and send the environmental control instruction to the environmental control equipment of the plant factory for control; The growth state acquisition unit is connected to the environment control unit and uses image recognition technology to analyze the plant according to the environment control result to obtain the actual growth state of the plant.

5. The artificial intelligence-based plant factory environmental control system according to claim 4 is characterized in that: The experience pool update module includes: A difference analysis unit compares and analyzes the actual growth state and the target growth state to obtain a difference value; A growth difference stage judgment unit, connected to the difference analysis unit, judges the growth difference stage by using an algorithm according to the difference value, performs an in-depth analysis on the plants in the growth difference stage, and identifies the difference factors; The strategy formulation unit is connected to the growth difference stage judgment unit, formulates corresponding adjustment strategies according to difference factors, and updates corresponding strategy contents in the multi-layer experience pool to improve the multi-layer experience pool.

6. The artificial intelligence-based plant factory environmental control system according to claim 5 is characterized in that: The difference analysis unit is the difference value between the target maturity cycle of the plant and the actual maturity cycle of the plant and the difference value between the target maturity yield and the actual maturity yield.

7. An artificial intelligence-based plant factory environment control method applied to the artificial intelligence-based plant factory environment control system according to any one of claims 1 to 6, characterized in that: include: Modeling is performed based on the current environmental parameters of the plant factory and the growth status of the plants to obtain a plant factory model; Dividing the historical cultivation strategies of the plant factory into multiple experience layers according to the gaps between historical growth parameters and the confidence of the strategies to form a multi-layer experience pool; Based on the target growth state of the plants in the plant factory model, the intelligent control strategy for each growth stage of the plants under the target growth state is determined, and based on the confidence of the strategy and the target matching degree, the multi-layer experience pool is screened to obtain the experience control strategy, wherein the intelligent control strategy is the minimum energy consumption strategy for achieving the target growth state, and the experience control strategy is the actual cultivation strategy for achieving the target growth state; The intelligent control strategy and the empirical control strategy are dynamically adjusted by using a dual-path mechanism to face the real-time changing plant factory environment, thereby determining a target control strategy; Regulating the environment of the plant factory based on the adjustment instruction to obtain the actual growth status of the plant, wherein the adjustment instruction is generated based on the target control strategy and the environmental parameters of the current plant factory; The difference factors between the actual growth state and the target growth state are determined, and the content of the experience layer is updated based on the difference factors to improve the multi-layer experience pool data.

Images