Plasma nitrogen preparation hydroponic intelligent integrated regulation and control system and regulation and control method

Abstract

The invention discloses a plasma nitrogen preparation hydroponic intelligent integrated regulation and control system and a regulation and control method. The system comprises a plasma fluidized bed ammonia preparation device, wherein generated ammonia gas is dissolved in water to be changed into ammonium nitrogen; the nitrogen storage module comprises a liquid storage tank III and a liquid storage tank IV; the nitration device is used for converting part of ammonium nitrogen into nitrate nitrogen; the intelligent control unit comprises various sensors used for monitoring the growth environment of crops in the hydroponic device, a nitrogen demand parameter library of all growth stages of common crops is preset, the LSTM prediction model is used for predicting the concentration demand change trend of ammonium nitrogen and nitrate nitrogen in the future, the deviation between the current nitrogen concentration and a target value is calculated in real time based on a PID algorithm, and the current nitrogen concentration is calculated according to the deviation. And the working states of the plasma fluidized bed ammonia preparation device and the nitrification device and the supply rates of the liquid storage tank III and the liquid storage tank IV to the water culture device are dynamically adjusted. The system can intelligently and autonomously decide nitrogen fertilizer supply and respond to the dynamic nature of plant growth requirements, nitrogen is quantitatively supplied in a high-precision mode, and the ammonium-nitrate ratio is controlled in a high-precision mode.

Description

Technical Field

[0001] The present invention relates to the technical field of hydroponic nitrogen fertilizer regulation, and particularly to a plasma nitrogen production hydroponic intelligent integrated regulation system and a regulation method. Background Art

[0002] Hydroponics is a new type of soilless cultivation method for plants, also known as nutrient solution cultivation. Its core is to directly immerse the roots of plants in the nutrient solution, which can replace the soil and provide plants with growth factors such as water, nutrients, and oxygen, enabling plants to grow normally. The common hydroponic techniques are as follows: deep flow technique, nutrient film technique, floating board capillary cultivation technique, and aeroponics technique. In the hydroponics industry, precisely regulating the application rate of nitrogen fertilizer is the core task for achieving high yields and reducing the nitrate content in vegetables. As an essential macronutrient for plant growth, nitrogen is the core component for constructing key biological macromolecules such as proteins, nucleic acids, and chlorophyll. However, excessive application of nitrogen fertilizer will cause crops to absorb it excessively, resulting in abnormal accumulation of nitrates in plants, which not only affects the quality of vegetables but also, because nitrates are converted into nitrites in the human body, can cause methemoglobinemia and even generate nitrosamines with carcinogenic risks, threatening human health.

[0003] However, the existing hydroponic techniques have a low level of intelligence, usually lacking the ability of self-adaptive nitrogen fertilizer regulation and an effective model self-optimization mechanism, resulting in the system being less sensitive to environmental changes, that is, factors such as temperature and light will affect the absorption rate of nitrogen fertilizer by plants, and at the same time, the feedback on the growth state of crops is less sensitive. The management of the nitrogen fertilizer storage solution mostly relies on manual intervention, such as regularly monitoring the nitrogen fertilizer concentration, manually adjusting the fertilization amount, and regularly replacing the storage solution, etc. This method not only increases the management cost but also is prone to inaccurate fertilization due to human errors, limiting the automation level of hydroponics. Therefore, realizing model self-optimization and reducing manual intervention are the keys to improving the intelligence level and management efficiency of the hydroponic system, which helps to improve the nitrogen fertilizer utilization efficiency, reduce the management difficulty and cost, and create a more precise and stable crop growth environment.

[0004] There is also the use of plasma technology for nitrogen fixation in the existing hydroponic techniques. The advantage of using plasma technology to produce nitrogen fertilizer is that it can supply nitrogen fertilizer autonomously and intelligently.

[0005] As disclosed in the Chinese invention patent with the application number CN202310649129.4, the integrated intelligent plasma water mist cultivation system generates liquid-phase plasma active substances in tap water through a plasma discharge unit, realizes nitrogen fixation in water, and transfers the nitrogen-fixed water to a storage tank. The entire process can complete the production and storage of nitrogen fertilizer without manual intervention, effectively avoiding errors and additional costs that may be brought by manual operation. It converts nitrogen into compounds such as ammonium salts, nitrous acid, and nitrites through a plasma device, and the compounds further react to form compounds such as urea and nitrates for plant use. However, the ammonium-nitrate ratio (NH4 + / NO3 - ) of plant nitrogen is uncontrollable. The regulation of the ammonium-nitrate ratio (NH4 + / NO3 - ) is one of the core technologies for optimizing plant growth, yield, and quality. Different crops have significant differences in their responses to nitrogen forms. By dynamically adjusting the ammonium-nitrate ratio, the nitrogen metabolism pathways (such as nitrate reductase activity and amino acid synthesis) and physiological responses (such as chlorophyll content and antioxidant capacity) can be coordinated to achieve the dual goals of high yield and high quality. That is, the technical solution disclosed in this invention patent cannot adjust the ammonium-nitrate ratio and is difficult to meet the different requirements of different plants for ammonium nitrogen and nitrate nitrogen at each growth stage; at the same time, the control logic depends on a fixed time sequence, and the accuracy of nitrogen concentration regulation is limited, and the nitrogen supply strategy cannot be flexibly optimized according to the actual growth conditions of plants.

[0006] Although the Chinese patent with the application number CN202411632424 discloses a self-stabilizing hydroponic cultivation system suitable for bumpy environments, the core of its control method lies in a passive regulation mechanism based on the nitrogen nutrition index. By collecting the nitrogen fertilizer reserves in the nutrient solution, the crop biomass, and the potential nitrogen demand, the nitrogen nutrition index is calculated to determine the nitrogen fertilizer application rate, realizing the feedback regulation of the nitrogen content in the hydroponic environment. The system has multiple sets of control planting analysis functions, and the growth data is collected through a camera, and the control module compares and discriminates the advantages and disadvantages of different nutrient solution schemes. This patented technology relies on the feedback regulation of preset thresholds and fixed parameters, belongs to passive response control and the control logic is fixed, and it cannot autonomously adjust the control strategy according to the dynamic growth of crops.

[0007] It should be specifically noted that the above technical information is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as an admission or an indication in any form that the above technical information constitutes the prior art known to those skilled in the art. Summary of the Invention

[0008] In view of the deficiencies in the above-mentioned background art, the present invention proposes a plasma nitrogen production hydroponic intelligent integrated control system and control method, and the technical problems to be solved are: how to make intelligent autonomous decisions on nitrogen fertilizer supply, as well as respond to the dynamics of plant growth requirements, high-precision quantitative supply of nitrogen, and high-precision control of ammonium-nitrate ratio.

[0009] The technical solution of the present invention is as follows:

[0010] A plasma nitrogen production hydroponic intelligent integrated control system includes a plasma fluidized bed ammonia production device, a nitrogen storage module, a nitrification device, an intelligent control unit, and a hydroponic device.

[0011] The plasma fluidized bed ammonia production device: Air sieves out nitrogen through a vacuum pressure swing adsorption device one, the nitrogen passes through the plasma fluidized bed device to generate ammonia, and the ammonia dissolves in water to become ammonium nitrogen and is stored in storage tank three.

[0012] The nitrogen storage module: includes storage tank three for storing high-concentration ammonium nitrogen and transporting it to the hydroponic device, and storage tank four for storing high-concentration nitrate nitrogen and transporting it to the hydroponic device.

[0013] The nitrification device: is used to convert part of the ammonium nitrogen in storage tank three into nitrate nitrogen and store it in storage tank four.

[0014] The intelligent control unit: includes a nitrogen concentration sensor, a light sensor, a temperature sensor, and a plant growth state detector for monitoring the crop growth environment in the hydroponic device, pre-stores a nitrogen demand parameter library for each growth stage of common crops, uses an LSTM prediction model to predict the future change trend of ammonium nitrogen and nitrate nitrogen concentration requirements, and based on the PID algorithm, calculates the deviation between the current nitrogen concentration and the target value in real time, and dynamically adjusts the working states of the plasma fluidized bed ammonia production device and the nitrification device, as well as the supply rates of storage tank three and storage tank four to the hydroponic device.

[0015] Preferably, the nitrogen storage module further includes storage tank one and storage tank two for storing low-concentration ammonium nitrogen. The plasma fluidized bed ammonia production device, storage tank one, storage tank two, and storage tank three are connected in sequence. Storage tank two is provided with a detection unit connected to the intelligent control unit and used for detecting the liquid volume and ammonium nitrogen concentration.

[0016] Preferably, storage tank two is connected to the plasma fluidized bed device in a circulating manner through a semi-permeable membrane.

[0017] Preferably, the nitrification device is connected to storage tank one, and the by-product oxygen of the plasma fluidized bed device is separated through storage tank one and then transported to the nitrification device for nitrification.

[0018] Preferably, the nitrification device is connected to a second vacuum pressure swing adsorption device, which inhales air and screens out oxygen for delivery to the nitrification device for nitrification.

[0019] Preferably, the plasma fluidized bed ammonia production device includes a quartz glass reactor filled with catalyst particles, a porous gas distribution plate is arranged at the bottom, and the reactor is configured with several groups of tungsten rod-shaped electrodes connected to a high-frequency AC power supply. After nitrogen is screened out by the first vacuum pressure swing adsorption device, it is introduced into the plasma fluidized bed device together with water vapor, and ammonia is generated through dielectric barrier discharge.

[0020] Preferably, it is powered by the local wind energy and / or solar energy of the hydroponic device.

[0021] A control method for a plasma nitrogen production hydroponic intelligent integrated regulation system adopts the plasma nitrogen production hydroponic intelligent integrated regulation system described in the above technical solution, including the generation of ammonium nitrogen, the generation of nitrate nitrogen, and the regulation of nitrogen by the intelligent control unit. Among them, the regulation of nitrogen by the intelligent control unit: with the ESP32 microcontroller as the core, with the help of nitrogen concentration sensors, light sensors and plant growth state detectors to collect environmental and plant physiological data in real time. A nitrogen demand parameter library for each growth stage of common crops is preset in the module. After the user selects the plant species and the current growth stage, the system automatically loads the preset target concentration, supports the user to customize and expand the parameter library, and automatically adjusts the nitrogen demand standard according to the feedback of the plant growth state detector; when regulating nitrogen, the initial LSTM model relies on pre-trained weights and real-time multi-source data to predict nitrogen demand, and calculates the deviation between the current nitrogen concentration and the target value in real time according to the PID algorithm, and dynamically adjusts the working states of the plasma fluidized bed ammonia production device and the nitrification device, as well as the supply rates of liquid storage tanks three and four to the hydroponic device; as the operation time accumulates, the module uses the window data of the past 72 hours to online optimize the LSTM model parameters every 6 hours, gradually realizing fully adaptive precise regulation and precisely meeting the nitrogen demand of plants at different growth stages.

[0022] Preferably, for the generation of ammonium nitrogen: air is introduced into the first vacuum pressure swing adsorption device to obtain high-purity nitrogen and introduced into the plasma fluidized bed device. Water reaches the plasma region from the storage tank through a pressure nozzle to generate ammonia and oxygen. The ammonia gas flows through the liquid storage tank and dissolves in water to become NH4 + solution, and the mixed solution flows into the liquid storage tank for concentration accumulation. The solution flows through the semi-permeable membrane, and the water molecules flow back to the plasma fluidized bed device after passing through the semi-permeable membrane. NH4 + solution flows into liquid storage tank three when the concentration reaches the standard. According to the nitrogen demand predicted by the LSTM model, the plasma fluidized bed ammonia production device is adjusted in advance according to the remaining amount of liquid storage tank three.

[0023] Preferably, for the generation of nitrate nitrogen: Part of the high-concentration ammonium nitrogen in the third storage tank is introduced into the nitrification device. When the demand for nitrate nitrogen is small, the by-product oxygen of the plasma fluidized bed device is separated through the first storage tank and then used for nitrification; when the demand for nitrate nitrogen is large, the second vacuum pressure swing adsorption device is turned on to inhale air and screen out oxygen for nitrification.

[0024] Preferably, in the process of optimizing the LSTM model using 72-hour window data, the model reduces the weights of the information with low prediction value and weak relevance to the plant nitrogen demand in the window data through the forget gate, achieving selective forgetting; at the same time, relying on the input gate and memory mechanism, the data containing the characteristics of the plant growth stage and the changing trend of environmental factors and significant for the prediction of nitrogen demand is integrated into the memory unit and strengthened, thereby adjusting the connection weights and thresholds of the internal neurons of the model, improving the accuracy of the plant nitrogen demand prediction, and optimizing the effect of nitrogen regulation in the hydroponic system.

[0025] The existing hydroponic technology has a low degree of intelligence and lacks an effective model self-optimization mechanism, often requiring manual intervention, which increases the management difficulty and cost. The present invention solves this problem by constructing a new intelligent control system and method, improving the intelligent level of the system. Specifically, the system takes the ESP32 microcontroller as the core, and is equipped with a nitrogen concentration sensor, a light sensor, and a plant growth state detector to collect multi-source data in real time. At the same time, a nitrogen demand parameter library for each growth stage of common crops is preset, supporting users to customize and expand plant data. The system automatically loads the preset target concentration according to the plant species and growth stage selected by the user, and automatically adjusts the nitrogen demand standard according to the feedback of the plant growth state detector. During the regulation process, the initial LSTM model relies on the pre-trained weights and real-time data to predict the nitrogen demand, and the PID controller dynamically adjusts the plasma nitrogen supply device accordingly to stabilize the nitrogen concentration. Moreover, the module uses the past 72-hour window data to optimize the LSTM model parameters online every 6 hours, gradually realizing fully adaptive and precise regulation, effectively promoting the improvement of the plant growth quality and yield.

[0026] Regarding the nitrogen fertilizer storage solution of the existing hydroponic intelligent system, it mainly relies on manual addition or uses nitrogen oxides supplied by nitrogen fixation through plasma technology, resulting in the problem of uncontrollable ammonium-nitrate ratio. This system adopts a plasma fluidized bed nitrogen fertilizer reactor to optimize the nitrogen fertilizer supply process, reduce risks and improve accuracy. Specifically, the plasma fluidized bed nitrogen fertilizer reactor adopts a fluidized bed structure, enhancing the contact efficiency between nitrogen and the catalyst, reducing local polarization phenomena, thereby extending the service life of the reaction catalyst and the device. The ammonia produced is dissolved in water and becomes NH4 +, after separation by the semi-permeable membrane, it flows into the storage tank when the concentration reaches the standard. Combining with the nitrogen demand predicted by the LSTM model, the plasma device for producing nitrogen fertilizer can be adjusted in advance according to the remaining amount in the liquid storage tank to achieve precise supply and reduce the risk of nitrate accumulation. In addition, the nitrification device set in this project can convert ammonium nitrogen into nitrate nitrogen through nitrification according to the demand. When the demand for nitrate nitrogen is small, the by-product oxygen of the plasma ammonia production device is used for nitrification after separation by the first liquid storage tank; when the demand for nitrate nitrogen is large, the vacuum pressure swing adsorption device two is turned on to inhale air and screen out oxygen for nitrification. The whole nitrification process is divided into two stages. In the first stage, nitrite bacteria oxidize ammonia nitrogen (NH4 + ) into nitrite ions (NO2 - ), and in the second stage, nitrifying bacteria further oxidize nitrite ions (NO2 - ) into nitrate ions (NO3 - ). By controlling the reaction process and the supply amount of reactants in different stages, the precise regulation of the ratio of ammonium nitrogen to nitrate nitrogen is realized, meeting the specific requirements of different plant growth stages for the ammonium-nitrate ratio, effectively solving the problem of uncontrollable ammonium-nitrate ratio in the existing system, and improving the growth quality and nutrient absorption efficiency of hydroponic plants.

[0027] Regarding the problem of remote areas with high nitrogen fertilizer transportation costs, this system can generate electricity locally using local resources to produce nitrogen, reducing costs and promoting agricultural development. Specifically, the integrated plasma nitrogen production hydroponic intelligent control system can directly produce nitrogen fertilizer in the target area without long-distance transportation, significantly reducing transportation costs. For example, in the western region of China where there is abundant wind and solar energy resources, the system can use the local rich wind and solar energy resources to provide power for the plasma fluidized bed nitrogen fertilizer reactor, realizing the efficient utilization of clean energy, reducing production costs and environmental pollution, and providing sustainable support for agricultural development in the western region.

[0028] Compared with the prior art in the background art:

[0029] Among them, the plasma nitrogen fertilizer device of the integrated intelligent plasma water mist cultivation system adopts a coaxial quartz glass double-wall structure and operates based on a static aqueous phase environment. Its metal electrode serves as the high-voltage electrode, using dielectric barrier discharge to excite plasma active substances in water. An air pump discharges the water body to create an air discharge environment, and the dissolution of the electrode can adjust the pH and supplement metal elements. In terms of the control method of the present invention, with the ESP32 microcontroller as the core, an intelligent regulation system is constructed by integrating the LSTM time series prediction model and the PID control algorithm. The model parameters are optimized every 6 hours according to the latest data to accurately adapt to environmental changes. The plasma fluidized bed device is used to directly prepare ammonium nitrogen required by plants, and then nitrate nitrogen is prepared as needed through nitrification, and the ammonium-nitrate ratio can be flexibly adjusted as needed. The device adopts a dielectric barrier discharge (DBD) plasma fluidized bed, which is driven by a 0.05 - 100 kHz high-voltage AC power supply. The fluidized particle bed layer is used to enhance the gas (nitrogen)-solid (catalyst) contact area, improve the reaction heat and mass transfer efficiency, and can extend the service life of the device and the catalyst.

[0030] Among them, a self-stabilizing hydroponic cultivation system suitable for bumpy environments relies on the feedback regulation of preset thresholds and fixed parameters, belongs to passive response control with a fixed control logic, and cannot autonomously adjust the control strategy according to the dynamic growth of crops. The present invention adopts active predictive control based on the LSTM time series model, and a dynamic prediction model of nitrogen demand is established through pre-trained weights and online parameter optimization. In addition, this solution realizes the adaptive adjustment to the changes in the plant growth stage by online optimizing the LSTM model parameters every 6 hours using the window data of the past 72 hours. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0032] Figure 1 It is a schematic diagram of the principle of the intelligent integrated regulation system for nitrogen production by plasma in hydroponics;

[0033] Figure 2 It is a control logic diagram of the intelligent integrated regulation system for nitrogen production by plasma in hydroponics.

[0034] Explanation of the reference numerals in the drawings:

[0035] Ammonia production device 1 by plasma fluidized bed, vacuum pressure swing adsorption device one 101, plasma fluidized bed device 102;

[0036] Nitrogen storage module 2, first liquid storage tank 201, second liquid storage tank 202, third liquid storage tank 203, fourth liquid storage tank 204;

[0037] Nitrification device 3, second vacuum pressure swing adsorption device 301;

[0038] Intelligent control unit 4;

[0039] Hydroponic device 5. Specific implementation manner

[0040] 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 core concept of the present invention and the following embodiments, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0041] These embodiments are provided in this application to make this application thorough and complete, and to fully convey the scope of this application to those skilled in the art. It should be noted that: unless otherwise specifically stated, the relative arrangements of components and steps, the components of materials, numerical expressions and values described in these embodiments should be construed as merely exemplary, rather than as limitations.

[0042] It should be noted that the "one", "two", "three", "four" and similar words used in this application do not represent any order, quantity or importance, but are only used to distinguish different parts.

[0043] It should also be noted that in the description of this application, unless otherwise clearly defined and limited, the terms "installed", "connected", "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 directly connected, or indirectly connected through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific situations. When it is described that a specific device is located between a first device and a second device, there may or may not be an intermediate device between the specific device and the first device or the second device.

[0044] All terms used in this application have the same meanings as understood by those of ordinary skill in the art to which this application belongs, unless otherwise specifically defined. It should also be understood that terms defined in a general dictionary, such as those, should be construed to have a meaning consistent with their meaning in the context of the relevant art, and should not be interpreted in an idealized or overly formal sense, unless clearly defined as such here.

[0045] Known technologies, methods, and devices for those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and devices should be regarded as part of the specification.

[0046] To address the deficiencies of existing hydroponic systems in nitrogen supply and comprehensive regulation, including the lack of autonomous nitrogen fertilizer supply capacity, and problems such as insufficient real-time dynamics, low precision of nitrogen quantitative supply, low precision of ammonium-nitrate ratio control, and lack of intelligent autonomous decision-making mechanism when responding to plant growth needs, the present invention proposes a plasma nitrogen production hydroponic intelligent integrated regulation system and its regulation method.

[0047] The system combines a plasma fluidized bed nitrogen fertilizer reactor, a nitrification device, and a hydroponic nitrogen fertilizer regulation device. Based on the ESP32 microcontroller, it undertakes three core functions: real-time data acquisition, concentration prediction, and dynamic regulation. The controller continuously monitors parameters such as temperature and light in the reactor through sensors, analyzes the temporal patterns in historical operation data through the LSTM (Long Short-Term Memory Neural Network) model, such as the delayed impact of continuous high temperature or flow rate changes on ammonia ion concentration, predicts the future change trends of ammonium nitrogen and nitrate nitrogen concentration requirements, and based on the PID (Proportional-Integral-Derivative) algorithm, calculates the deviation between the current concentration and the target value in real time, dynamically adjusts the supply rates of different nitrogen sources, and realizes the precise adaptation to the nitrogen concentration requirements and nitrogen form requirements of different plants.

[0048] The reaction process of plasma fluidized bed ammonia production: Air is introduced into the vacuum pressure swing adsorption device I to obtain high-purity nitrogen and introduced into the plasma fluidized bed device. The water filtered and refluxed through the water storage tank enters the plasma region through a pressure nozzle. The reaction equation is as follows:

[0049]

[0050] After the reaction, ammonia will dissolve in water to become NH4 + solution. After the water molecules in the solution pass through the semi-permeable membrane, they flow back to the plasma fluidized bed ammonia production device. When the concentration of NH4 + solution reaches the standard, it flows into the liquid storage tank. According to the predicted nitrogen demand of the LSTM model, the plasma device for nitrogen fertilizer production can be adjusted in advance according to the remaining amount in the liquid storage tank.

[0051] Nitrification device: High-concentration ammonium nitrogen is introduced into the device. When the demand for nitrate nitrogen is small, the by-product oxygen of the plasma ammonia production device is separated through the first liquid storage tank and then undergoes nitrification. When the demand for nitrate nitrogen is large, the vacuum pressure swing adsorption device II is opened to inhale air and screen out oxygen for nitrification.

[0052] The nitrification process is divided into two stages. The first stage: Nitrosomonas bacteria convert ammonia nitrogen (NH4 +) is oxidized to nitrite ions (NO2-), and the reaction equation is as follows:

[0053]

[0054] The second stage: Nitrifying bacteria further oxidize nitrite ions (NO2 - ) to nitrate ions (NO3 - ), and the reaction equation is as follows:

[0055]

[0056] Hydroponic regulation device: The intelligent control module of the present invention takes the ESP32 microcontroller as the core, and uses a nitrogen concentration sensor, a light sensor, and a plant growth status detector to collect environmental and plant physiological data in real time. A nitrogen demand parameter library for each growth stage of common crops is preset in the module. After the user selects the plant species and the current growth stage, the system automatically loads the preset target concentration, and also supports the user to customize and expand the plant data. It can also automatically adjust the nitrogen demand standard according to the feedback of the plant growth status detector. When regulating nitrogen, in the initial stage, the LSTM model relies on pre-trained weights and real-time multi-source data to predict nitrogen demand, and calculates the deviation between the current nitrogen concentration and the target value in real time according to the PID algorithm, and dynamically adjusts the working states of the plasma fluidized bed ammonia production device and the nitrification device, as well as the supply rates of liquid storage tanks three and four to the hydroponic device; as the operation time accumulates, the module uses the window data of the past 72 hours to online optimize the LSTM model parameters every 6 hours, gradually realizing fully adaptive and precise regulation, and more precisely meeting the nitrogen demand of plants at different growth stages.

[0057] In the process of optimizing the LSTM model using the 72-hour window data, the model reduces the weight of the information in the window data that has low prediction value and weak relevance to the nitrogen demand of plants through the forget gate, realizing selective forgetting; at the same time, relying on the input gate and the memory mechanism, it integrates the data containing the characteristics of the plant growth stage and the changing trend of environmental factors and is of great significance to the nitrogen demand prediction into the memory unit and strengthens it, thereby adjusting the connection weights and thresholds of the internal neurons of the model, improving the accuracy of the plant nitrogen demand prediction, and optimizing the effect of nitrogen regulation in the hydroponic system.

[0058] The specific embodiments are as follows:

[0059] A plasma nitrogen production hydroponic intelligent integrated regulation system, as Figure 1 and Figure 2 shown, includes a plasma fluidized bed ammonia production device 1, a nitrogen storage module 2, a nitrification device 3, an intelligent control unit 4, and a hydroponic device 5.

[0060] The ammonia production device by plasma fluidized bed 1: Air passes through the vacuum pressure swing adsorption device 101 to screen out nitrogen, and the nitrogen passes through the plasma fluidized bed device 102 to generate ammonia. The ammonia is dissolved in water to become ammonium nitrogen and is stored in the third liquid storage tank 203.

[0061] The nitrogen storage module 2: includes the third liquid storage tank 203 for storing high-concentration ammonium nitrogen and transporting it to the hydroponic device 5, and the fourth liquid storage tank 204 for storing high-concentration nitrate nitrogen and transporting it to the hydroponic device 5.

[0062] The nitrification device 3: is used to convert part of the ammonium nitrogen in the third liquid storage tank 203 into nitrate nitrogen and store it in the fourth liquid storage tank 204.

[0063] The intelligent control unit 4: includes a nitrogen concentration sensor, a light sensor, a temperature sensor, and a plant growth state detector for monitoring the crop growth environment in the hydroponic device 5. There is a preset nitrogen demand parameter library for each growth stage of common crops. Using the LSTM prediction model to predict the future change trend of the demand for ammonium nitrogen and nitrate nitrogen concentration, based on the PID algorithm, the deviation between the current nitrogen concentration and the target value is calculated in real time, and the working states of the ammonia production device by plasma fluidized bed 1 and the nitrification device 3 and the supply rates of the third liquid storage tank 203 and the fourth liquid storage tank 204 to the hydroponic device 5 are dynamically adjusted.

[0064] Preferably, the plant growth state detector includes a plant root imaging module, and the plant growth state is judged by monitoring the plant roots.

[0065] Preferably, the nitrogen storage module 2 further includes the first liquid storage tank 201 and the second liquid storage tank 202 for storing low-concentration ammonium nitrogen. The ammonia production device by plasma fluidized bed 1, the first liquid storage tank 201, the second liquid storage tank 202, and the third liquid storage tank 203 are connected in sequence. The second liquid storage tank 202 is provided with a detection unit connected to the intelligent control unit 4 and used for detecting the liquid volume and ammonium nitrogen concentration.

[0066] Preferably, the second liquid storage tank 202 is connected to the plasma fluidized bed device 102 in a circulating manner through a semi-permeable membrane, that is, the aqueous solution in the second liquid storage tank 202 is filtered through the semi-permeable membrane and then returned to the plasma fluidized bed device 102 for the generation of ammonium nitrogen.

[0067] Preferably, the nitrification device 3 is connected to the first liquid storage tank 201, and the by-product oxygen of the plasma fluidized bed device 102 is separated through the first liquid storage tank 201 and then transported to the nitrification device 3 for nitrification.

[0068] Preferably, the nitrification device 3 is connected to a second vacuum pressure swing adsorption device 301. The second vacuum pressure swing adsorption device 301 inhales air and screens out oxygen and transports it to the nitrification device 3 for nitrification.

[0069] Preferably, the plasma fluidized bed ammonia production device 1 includes a quartz glass reactor filled with catalyst particles inside, a porous gas distribution plate is arranged at the bottom, and the reactor is configured with several groups of tungsten rod-shaped electrodes connected to a high-frequency AC power supply. After nitrogen is screened out by the vacuum pressure swing adsorption device 101, it is introduced into the plasma fluidized bed device 102 together with water vapor, and ammonia is generated through dielectric barrier discharge.

[0070] Preferably, it is powered by the local wind energy and / or solar energy of the hydroponic device 5.

[0071] A control method for a plasma nitrogen production hydroponic intelligent integrated regulation system, which uses the above-mentioned plasma nitrogen production hydroponic intelligent integrated regulation system, includes the generation of ammonium nitrogen, the generation of nitrate nitrogen, and the regulation of nitrogen by the intelligent control unit 4. Among them, the regulation of nitrogen by the intelligent control unit 4: with the ESP32 microcontroller as the core, it uses nitrogen concentration sensors, light sensors, and plant growth state detectors to collect environmental and plant physiological data in real time. A nitrogen demand parameter library for each growth stage of common crops is preset in the module. After the user selects the plant species and the current growth stage, the system automatically loads the preset target concentration, supports the user to customize and expand the parameter library, and automatically adjusts the nitrogen demand standard according to the feedback of the plant growth state detector; when regulating nitrogen, the initial LSTM model relies on pre-trained weights and real-time multi-source data to predict nitrogen demand, and calculates the deviation between the current nitrogen concentration and the target value in real time according to the PID algorithm, and dynamically adjusts the working states of the plasma fluidized bed ammonia production device and the nitrification device, as well as the supply rates of the third liquid storage tank and the fourth liquid storage tank to the hydroponic device; as the operation time accumulates, the module uses the window data of the past 72 hours to online optimize the LSTM model parameters every 6 hours, and gradually realizes fully adaptive precise regulation to accurately meet the nitrogen demand of plants at different growth stages.

[0072] Preferably, for the generation of ammonium nitrogen: air is introduced into the vacuum pressure swing adsorption device 101 to obtain high-purity nitrogen and introduced into the plasma fluidized bed device 102. Water reaches the plasma region from the water storage tank through a pressure nozzle to generate ammonia and oxygen, and the ammonia gas flows through the liquid storage tank 201 and dissolves in water to become NH4 + solution, and the mixed solution flows into the liquid storage tank 202 for concentration accumulation. The solution flows through a semi-permeable membrane, and the water molecules flow back to the plasma fluidized bed device 102 after passing through the semi-permeable membrane. The NH4 + solution flows into the third liquid storage tank 203 when the concentration reaches the standard. According to the predicted nitrogen demand of the LSTM model, the plasma fluidized bed ammonia production device 1 is adjusted in advance according to the remaining amount of the third liquid storage tank 203.

[0073] Preferably, for the generation of nitrate nitrogen: Part of the high-concentration ammonium nitrogen in the third liquid storage tank 203 is introduced into the nitrification device 3. When the demand for nitrate nitrogen is small, the by-product oxygen of the plasma fluidized bed device 102 is separated through the first liquid storage tank 201 and then used for nitrification; when the demand for nitrate nitrogen is large, the second vacuum pressure swing adsorption device 301 is turned on, and air is inhaled to screen out oxygen for nitrification.

[0074] Preferably, in the process of optimizing the LSTM model using 72-hour window data, the model reduces the weight of the information with low prediction value and weak correlation for plant nitrogen demand in the window data through the forget gate, realizing selective forgetting; at the same time, relying on the input gate and memory mechanism, the data containing the characteristics of the plant growth stage and the changing trend of environmental factors and significant for nitrogen demand prediction is integrated into the memory unit and strengthened, thereby adjusting the connection weights and thresholds of the internal neurons of the model, improving the accuracy of plant nitrogen demand prediction, and optimizing the effect of nitrogen regulation in the hydroponic system.

[0075] Specific application examples are as follows:

[0076] A plasma nitrogen production hydroponic intelligent integrated control system, as Figure 1 shown, its core consists of a plasma fluidized bed ammonia production device 1, a nitrogen storage module 2, a nitrification device 3, an intelligent control unit 4, and a hydroponic device 5.

[0077] The plasma fluidized bed ammonia production device 1 includes a quartz glass reactor with an inner diameter of 30 cm and a height of 80 cm; it is filled with catalyst particles, 1 - 2 μm, 4 wt% Co-Ni / Mgo catalyst; a porous gas distribution plate is arranged at the bottom with a pore diameter of 3 - 5 μm and an air flow distribution uniformity > 90%. The reactor is equipped with 6 groups of tungsten rod-shaped electrodes with a diameter of 5 mm, a withstand voltage of 30 kV, connected to a high-frequency AC power supply, with an output voltage of 30 - 100 V and a frequency of 5 - 40 kHz, and the new frequency point is about 20 kHz. After air is introduced, nitrogen is screened out through the first vacuum pressure swing adsorption device 101, and nitrogen and water are introduced into the plasma fluidized bed device 102, and ammonia is generated through dielectric barrier discharge. The ammonia dissolves in water to become ammonium nitrogen (NH4 + ), after the hydrolysis solution is separated by a polyethersulfone semipermeable membrane (cut-off molecular weight 100 Da), NH4 + is enriched to the second liquid storage tank 202 with a capacity of 100 L. When the concentration reaches 1.8 mmol / L, it triggers transportation. The water recovered through the semipermeable membrane is further atomized and returned to the plasma fluidized bed device 102. After the nitrogen concentration and volume in the second liquid storage tank 202 reach the standard, the outlet of the first liquid storage tank 201 is closed, and the outlet of the second liquid storage tank 202 is opened. After the high-concentration nitrogen enters the third liquid storage tank 203, the outlet of the first liquid storage tank 201 is opened.

[0078] The nitrification device 3 is connected to the liquid storage tank three 203 and is used to convert high-concentration ammonium nitrogen into nitrate nitrogen and store it in the liquid storage tank four 204. When the demand for nitrate nitrogen is small, the by-product oxygen of the plasma fluidized bed ammonia production device 102 is separated through the liquid storage tank one 201 and then undergoes nitrification; when the demand for nitrate nitrogen is large, the vacuum pressure swing adsorption device two 301 is turned on to inhale air and screen out oxygen for nitrification. The nitrification process is divided into two stages. The first stage: nitrite bacteria oxidize ammonia nitrogen (NH3 or NH4 + ) to nitrite nitrogen (NO2 - );The second stage: nitrifying bacteria further oxidize nitrite nitrogen (NO2 - ) to nitrate nitrogen (NO3 - ).

[0079] The intelligent control unit 4 is centered on the ESP32 microcontroller, and the control process is as Figure 2 , connecting the nitrogen concentration sensor (range 0 - 10 mmol / L, accuracy ±0.5%), the light sensor (range 0 - 65535 lx), and the plant root imaging module (resolution 1600×1200). The controller has an LSTM prediction model built-in, and its mathematical expression is:

[0080] f t =σ(W f ·[h t-1 , x t +b f )

[0081] i t =σ(W i ·[h t-1 , x t +b i )

[0082]

[0083] o t =σ(W o ·[h t-1 , x t +b o )

[0084] h t =o t ⊙tanh(C t )

[0085] t: The current 6-hour time window;

[0086] x t : The input vector at time t (temperature + light + nitrogen concentration);

[0087] h t-1: Hidden state at the previous moment (summary of historical environmental features);

[0088] C t-1 : Cell state at the previous moment (storing long-term rules, such as the decrease in nitrogen absorption rate with high temperature);

[0089] f t : Forgetting gate vector (filtering out noise / invalid data: σ output [0, 1]);

[0090] i t : Input gate vector (strengthening key features: output [0, 1]);

[0091] Candidate cell state (proposal of new rules for the current environment: tanh output [-1, 1]);

[0092] C t : Updated cell state

[0093] o t : Output gate vector (controlling the contribution of C t to the prediction: σ output [0, 1]);

[0094] h t : Current hidden state (o t ⊙tanh(C t ), output prediction value carrier.

[0095] In the formula, the input parameter x t is the input vector at time t, that is, the time series data of temperature (20 - 40 °C), light intensity (200 - 1000 μmol / m 2 / s), and nitrogen concentration (0 - 5 mmol / L) within 72 hours, and the predicted value of nitrogen demand in the next 6 hours is output. The model is optimized every 6 hours. The forgetting gate f t filters out invalid data, such as sensor instantaneous noise, and the input gate i t strengthens key features, such as the rule that the root nitrogen absorption rate decreases by 20% due to continuous high temperature.

[0096] The parameters of the PID control algorithm are set as the proportional coefficient K p = 0.6, the integral time T i = 15 s, and the derivative time T d = 3 s, and the voltage of the plasma device (50 - 150 W) and the nitrogen flow rate of the liquid storage tank three 203 (80 - 150 mL / min) are adjusted in real time. The deviation calculation formula is:

[0097]

[0098] u(t): Control output (signal to the actuator);

[0099] e(t): Deviation (set value - measured value);

[0100] K p = 0.6: Proportional gain - amplifies the current deviation, determines the response speed, and is prone to oscillation if too large;

[0101] K i = K p / T i : Integral gain - eliminates the cumulative deviation (steady - state error), T i = 15 s, which is the integral time;

[0102] K d = K p *T d : Derivative gain - suppresses the change trend (predicts the future), reduces overshoot; T d = 3 s is the derivative time. After the system starts, the user selects the crop type as tomato, the growth stage as the seedling stage, the preset nitrogen concentration target as 2.5 mmol / L, and the initial flow rate of the fluidized bed is set to 1.2 L / min. At this time, the system enters the nitrogen regulation stage in the seedling stage.

[0103] In the seedling stage, the proportion of the plant's demand for ammonium nitrogen is about 60%. The plasma fluidized bed device 102 first operates at high power to generate ammonia, which dissolves in water to form ammonium nitrogen. The nitrogen concentration sensor collects nitrogen data in the liquid storage tank three 203 and the liquid storage tank four 204 at a frequency of once every 10 seconds to monitor the change of nitrogen concentration in real - time. When the liquid level of the liquid storage tank three 203 reaches 85%, the system automatically switches to the low - power mode, and the voltage of the plasma fluidized bed device 102 and the nitrogen flow rate in the liquid storage tank three 203 are adjusted in real - time through the PID control algorithm. According to the deviation calculation formula, the nitrogen concentration is accurately controlled to be 7.0 - 10.5 mmol / L (depending on the growth stage), and the concentration fluctuation is maintained within the range of ±0.5%. When the liquid level of the liquid storage tank three 203 further rises to 90%, the plasma fluidized bed device 102 is turned off; after the liquid level drops to 75%, the low - power mode is restarted to ensure the continuity and stability of nitrogen supply.

[0104] Meanwhile, the nitrification device 3 operates at low power, using the by - product oxygen of the plasma fluidized bed ammonia production device 1 to convert part of the ammonium nitrogen into nitrate nitrogen to meet the basic demand of the plant for nitrate nitrogen. As the liquid level of the liquid storage tank four 204 changes, the system automatically adjusts the power of the nitrification device 3. If the light intensity > 600 μmol / m 2 / s lasts for 1 hour, the LSTM prediction model is based on the temperature (20 - 40 °C) and light intensity (200 - 1000 μmol / m 2Sequential data such as (s) and nitrogen concentration (0 - 5 mmol / L) are input, and a prediction result that the nitrogen demand will increase by 25% in the next 3 hours is output. At this time, the plasma fluidized bed ammonia production device 1 switches to the high-power mode to rapidly increase the production of ammonium nitrogen. At the same time, the nitrification device 3 also increases its power accordingly to accelerate the conversion of ammonium nitrogen to nitrate nitrogen.

[0105] As the tomato enters the growth stage, the total nitrogen demand of the plant increases significantly. The proportion of ammonium nitrogen demand drops to 30%, and the proportion of nitrate nitrogen demand rises to 70%. To meet the nitrogen demand of the plant at this stage, both the plasma fluidized bed device 102 and the nitrification device 3 operate at high power. The vacuum pressure swing adsorption device II 301 is turned on to inhale air and screen out oxygen for nitrification. The real-time data collected by the nitrogen concentration sensor is compared with the preset nitrogen concentration target value of 6.4 - 10.7 mmol / L (depending on the growth stage), and the deviation is input into the PID control algorithm. According to the calculation result, the voltage of the plasma fluidized bed device 102 and the nitrogen flow rate in the liquid storage tank III 203 are precisely adjusted to ensure that the total amount and morphological ratio of nitrogen supply meet the growth requirements of the plant. When the ammonium nitrogen concentration in the liquid storage tank III 203 is too high or the plant's demand for nitrate nitrogen increases, the power of the nitrification device 301 is increased to accelerate the nitrification process; otherwise, the power is reduced or the nitrification device 301 is turned off to achieve precise and dynamic regulation of nitrogen supply, improve the utilization efficiency of nitrogen fertilizer, and promote the healthy growth of tomatoes.

[0106] So far, the inventive concept, embodiments and application examples of the present invention have been described in detail. To avoid obscuring the core inventive concept of the present invention, some details well-known in the art have not been described. Those skilled in the art can clearly understand how to implement the technical solutions disclosed in the above embodiments based on the above description.

[0107] Although some specific embodiments of the present invention have been described in detail by way of examples, those skilled in the art should understand that the above examples are only for illustration purposes and not for limiting the scope of the present invention. Those skilled in the art should understand that the above embodiments can be modified or some technical features can be equivalently replaced without departing from the scope and spirit of the present invention. In particular, as long as there is no conflict in principle, the various technical features mentioned in each embodiment can be combined in any way.

Claims

1. A plasma nitrogen production hydroponic intelligent integrated control system, characterized in that: It includes a plasma fluidized bed ammonia production device (1), a nitrogen storage module (2), a nitrification device (3), an intelligent control unit (4) and a hydroponic device (5); The plasma fluidized bed ammonia production device (1): Air is sieved to obtain nitrogen through a vacuum pressure swing adsorption device I (101), the nitrogen passes through a plasma fluidized bed device (102) to generate ammonia, and the ammonia is dissolved in water to become ammonium nitrogen and then stored in a liquid storage tank III (203); The nitrogen storage module (2): It includes a liquid storage tank III (203) for storing high-concentration ammonium nitrogen and transporting it to the hydroponic device (5), and a liquid storage tank IV (204) for storing high-concentration nitrate nitrogen and transporting it to the hydroponic device (5); The nitrification device (3): It is used to convert part of the ammonium nitrogen in the liquid storage tank III (203) into nitrate nitrogen and store it in the liquid storage tank IV (204); The intelligent control unit (4): It includes a nitrogen concentration sensor, a light sensor, a temperature sensor, and a plant growth state detector for monitoring the crop growth environment in the hydroponic device (5), pre-sets a nitrogen demand parameter library for each growth stage of common crops, uses an LSTM prediction model to predict the changing trend of future ammonium nitrogen and nitrate nitrogen concentration demands, and based on the PID algorithm, calculates the deviation between the current nitrogen concentration and the target value in real time, and dynamically adjusts the working states of the plasma fluidized bed ammonia production device (1) and the nitrification device (3) as well as the supply rates of the liquid storage tank III (203) and the liquid storage tank IV (204) to the hydroponic device (5).

2. The plasma nitrogen production hydroponic intelligent integrated control system according to claim 1, wherein: The nitrogen storage module (2) also includes a liquid storage tank I (201) and a liquid storage tank II (202) for storing low-concentration ammonium nitrogen. The plasma fluidized bed ammonia production device (1), the liquid storage tank I (201), the liquid storage tank II (202), and the liquid storage tank III (203) are connected in sequence. The liquid storage tank II (202) is provided with a detection unit connected to the intelligent control unit (4) and used for detecting the liquid volume and ammonium nitrogen concentration.

3. The plasma nitrogen production hydroponic intelligent integrated control system according to claim 2, characterized in that: The liquid storage tank II (202) is connected in a cycle with the plasma fluidized bed device (102) through a semi-permeable membrane.

4. The plasma nitrogen production hydroponic intelligent integrated control system according to any one of claims 1-3, characterized in that: The nitrification device (3) is connected to the liquid storage tank I (201), and the by-product oxygen of the plasma fluidized bed device (102) is separated through the liquid storage tank I (201) and then transported to the nitrification device (3) for nitrification.

5. The plasma nitrogen production hydroponic intelligent integrated control system according to claim 4, wherein: The nitrification device (3) is connected with a vacuum pressure swing adsorption device II (301), and the vacuum pressure swing adsorption device II (301) inhales air and sieves out oxygen and transports it to the nitrification device (3) for nitrification.

6. The plasma nitrogen production hydroponic intelligent integrated control system according to any one of claims 1-3 and 5, characterized in that: The plasma fluidized bed ammonia production device (1) includes a quartz glass reactor, which is filled with catalyst particles inside, a porous gas distribution plate is arranged at the bottom, the reactor is configured with several groups of tungsten rod-shaped electrodes connected to a high-frequency AC power supply. After the nitrogen is sieved out by the vacuum pressure swing adsorption device I (101), it is introduced into the plasma fluidized bed device (102) together with water vapor, and ammonia is generated through dielectric barrier discharge.

7. The plasma nitrogen production hydroponic intelligent integrated control system according to claim 6, characterized in that: It is powered by the local wind energy and / or solar energy of the hydroponic device (5).

8. A control method for a plasma nitrogen production hydroponic intelligent integrated control system, characterized in that: Adopt the plasma nitrogen production hydroponic intelligent integrated control system described in claim 7, including the generation of ammonium nitrogen, the generation of nitrate nitrogen, and the regulation of nitrogen by the intelligent control unit (4). Among them, the regulation of nitrogen by the intelligent control unit (4): with the ESP32 microcontroller as the core, relying on nitrogen concentration sensors, light sensors, and plant growth state detectors to collect environmental and plant physiological data in real time. A nitrogen demand parameter library for each growth stage of common crops is preset in the module. After the user selects the plant species and the current growth stage, the system automatically loads the preset target concentration, supports user-defined and extended parameter libraries, and automatically adjusts the nitrogen demand standard according to the feedback of the plant growth state detector; when regulating nitrogen, the initial LSTM model relies on pre-trained weights and real-time multi-source data to predict nitrogen demand, and calculates the deviation between the current nitrogen concentration and the target value in real time according to the PID algorithm, and dynamically adjusts the working states of the plasma fluidized bed ammonia production device (1) and the nitrification device (3) and the supply rates of the third liquid storage tank (203) and the fourth liquid storage tank (204) to the hydroponic device; As the operation time accumulates, the module uses the window data of the past 72 hours to online optimize the LSTM model parameters every 6 hours, gradually realizing fully adaptive precise control and precisely meeting the nitrogen demand of plants at different growth stages.

9. The control method of the plasma nitrogen production hydroponic intelligent integrated control system according to claim 8, characterized in that: Generation of the ammonium nitrogen: Air is introduced into the vacuum pressure swing adsorption device I (101), and high-purity nitrogen is obtained and introduced into the plasma fluidized bed device (102). Water reaches the plasma region through a pressure nozzle from the water storage tank to generate ammonia and oxygen. The ammonia gas flows through the liquid storage tank (201) and dissolves in water to become NH4 + solution. The mixed solution flows into the liquid storage tank (202) for concentration accumulation. The solution flows through the semi-permeable membrane, and the water molecules flow back to the plasma fluidized bed device (102) after passing through the semi-permeable membrane. NH4 + solution flows into the third liquid storage tank (203) when the concentration reaches the standard. According to the predicted nitrogen demand of the LSTM model, the plasma fluidized bed ammonia production device (1) is adjusted in advance according to the remaining amount of the third liquid storage tank (203); The generation of nitrate nitrogen: Part of the high-concentration ammonium nitrogen in the third liquid storage tank (203) is introduced into the nitrification device (3). When the demand for nitrate nitrogen is small, the by-product oxygen of the plasma fluidized bed device (102) is separated through the first liquid storage tank (201) and then undergoes nitrification; when the demand for nitrate nitrogen is large, the second vacuum pressure swing adsorption device (301) is opened, and air is inhaled to screen out oxygen for nitrification.

10. The control method of the plasma nitrogen production hydroponic intelligent integrated regulation system according to claim 8 or 9, characterized in that: In the process of optimizing the LSTM model using the 72-hour window data, the model reduces the weight of information with low prediction value and weak relevance to plant nitrogen demand in the window data through the forget gate, realizing selective forgetting; at the same time, relying on the input gate and memory mechanism, data containing the characteristics of the plant growth stage and the change trend of environmental factors and of great significance for nitrogen demand prediction is integrated into the memory unit and strengthened, thereby adjusting the connection weights and thresholds of the internal neurons of the model, improving the accuracy of plant nitrogen demand prediction, and optimizing the effect of nitrogen regulation in the hydroponic system.

Images