Fish and vegetable symbiotic energy internet control method and device, electronic equipment and medium

Abstract

The invention relates to the technical field of energy internet, in particular to a fish and vegetable symbiosis energy internet control method and device, electronic equipment and a medium. According to the current water quality parameter, the current cultivation environment parameter, the current fish and vegetable health condition and the current near-surface meteorological condition parameter, predicting a water quality change condition and a cultivation environment change condition in a preset duration, and determining a fishery breeding early warning grade and a plant cultivation early warning grade; and determining a target control parameter of each breeding controllable electrical load module in combination with the current states of the plurality of breeding controllable electrical load modules, and controlling the corresponding breeding controllable electrical load module. Therefore, the problems that in the prior art, complex climate and sudden risks are difficult to deal with, consequently, the breeding production efficiency is low, and power consumption is sharply increased are solved, and real-time dynamic regulation and control, fault early warning and cost compression are achieved by integrating Internet of Things sensing, an intelligent algorithm and automatic control.

Description

Technical Field

[0001] The present application relates to the field of energy internet technology, and in particular to a method, device, electronic equipment and medium for controlling aquaponics energy internet. Background Art

[0002] Aquaponics, an innovative ecological agricultural model, organically combines aquaculture with plant cultivation to create a closed-loop ecosystem: fish fertilize the water, vegetables purify the water, and water nourishes the fish. This system demonstrates profound value in environmental, economic, social, and sustainable development. However, current aquaponics systems rely primarily on manual environmental control, making them inadequate for complex climates and unexpected risks.

[0003] Water quality monitoring and cultivation environment monitoring are fundamental to the system's early warning and control systems. Traditional manual monitoring methods are not only inefficient but also prone to errors, making it difficult to detect and intervene in a timely manner. Furthermore, aquaponics systems are significantly affected by microclimate fluctuations and environmental conditions. If aquaculture loads are not promptly regulated, environmental indicators may reach warning levels simultaneously, impacting aquaculture performance and ecological balance, and even leading to secondary problems such as a surge in electricity consumption.

[0004] Therefore, in traditional aquaponics systems, the lack of refined and real-time load regulation has hampered aquaculture production efficiency. However, with the deep integration of the Energy Internet and the aquaculture industry, the scale of electrified equipment in actual production, such as temperature control and oxygen production, has rapidly increased, and the coupling relationship between their load characteristics and grid operation has become increasingly significant. Energy Internet optimization for distributed agricultural scenarios such as fisheries urgently needs to be addressed. This demonstrates that traditional aquaculture management methods are unable to respond promptly to changes in the external environment, resulting in inflexible scheduling and delayed decision-making and regulation. Summary of the Invention

[0005] The present application provides a fish-vegetable symbiotic energy Internet control method, device, electronic equipment and medium to solve the problems that the background technology is difficult to cope with complex climate and sudden risks, resulting in low aquaculture production efficiency and surge in electricity consumption. By integrating Internet of Things sensors, intelligent algorithms and automatic control, real-time dynamic regulation, fault warning and cost reduction are achieved.

[0006] The first embodiment of the present application provides a method for controlling an energy internet for aquaponics, including the following steps:

[0007] Obtain current water quality parameters, current cultivation environment parameters, current fish and vegetable health status, current near-ground meteorological condition parameters, and the current status of multiple aquaculture controllable power load modules;

[0008] Based on a preset deep learning time series prediction strategy, the water quality changes and cultivation environment changes within a preset time period are predicted according to the current water quality parameters, the current cultivation environment parameters, the current fish and vegetable health conditions, and the current near-surface meteorological condition parameters;

[0009] Determine the fishery farming warning level based on the water quality changes, and determine the plant planting warning level based on the cultivation environment changes;

[0010] The target control parameters of each aquaculture controllable power load module are determined according to the current status of the multiple aquaculture controllable power load modules, the fishery aquaculture warning level, and the plant planting warning level, and the corresponding aquaculture controllable power load module is controlled according to the target control parameters of each aquaculture controllable power load module.

[0011] According to one embodiment of the present application, determining the target control parameter of each aquaculture controllable power load module according to the current status of the multiple aquaculture controllable power load modules, the fishery aquaculture warning level, and the plant planting warning level includes:

[0012] Obtaining the current node marginal electricity price and current line congestion information, and based on a preset dynamic coordination optimization strategy, inputting the current node marginal electricity price and the current line congestion information into a pre-established power flow optimization sub-model to obtain a power flow optimization result;

[0013] Based on the grid flow optimization results, the current states of the multiple aquaculture controllable power load modules, the fishery aquaculture warning level and the plant planting warning level are respectively input into a pre-built fishery deep reinforcement learning model and a pre-built agricultural deep reinforcement learning model to obtain the target control parameters of each aquaculture controllable power load module.

[0014] According to one embodiment of the present application, the current water quality parameters include at least one of water temperature, dissolved oxygen content, pH, ammonia nitrogen content, nitrite concentration, water flow rate and liquid level.

[0015] According to one embodiment of the present application, determining the fishery farming warning level according to the water quality change includes:

[0016] Determining the interval of the water temperature, the interval of the dissolved oxygen content, the interval of the pH, the interval of the ammonia nitrogen content, the interval of the nitrite concentration, the interval of the water flow rate, and the interval of the liquid level;

[0017] The fishery breeding warning level is determined according to the water quality temperature range, the dissolved oxygen content range, the pH range, the ammonia nitrogen content range, the nitrite concentration range, the water flow rate range and the liquid level range, wherein the fishery breeding warning level includes the first to fourth levels, the warning level of the first level is less than the warning level of the second level, the warning level of the second level is less than the warning level of the third level, and the warning level of the third level is less than the warning level of the fourth level.

[0018] According to one embodiment of the present application, the current cultivation environment parameter includes at least one of air temperature, air humidity, light intensity and carbon dioxide concentration.

[0019] According to one embodiment of the present application, determining the plant planting warning level according to the change in the cultivation environment includes:

[0020] Determining a temperature interval for the air temperature, a humidity interval for the air humidity, a light interval for the light intensity, and a concentration interval for the carbon dioxide concentration;

[0021] The plant planting warning level is obtained according to the temperature range, the humidity range, the light range and the concentration range, wherein the plant planting warning level includes levels five to eight, the warning level of the fifth level is less than the warning level of the sixth level, the warning level of the sixth level is less than the warning level of the seventh level, and the warning level of the seventh level is less than the warning level of the eighth level.

[0022] According to one embodiment of the present application, the current fish and vegetable health status includes at least one of the fish farming status, fish survival rate, fish health status, farmed fish weight, plant planting status, plant survival rate, plant health status and plant planting area.

[0023] According to the energy internet control method for fish-vegetable symbiosis of the embodiment of the present application, based on the preset deep learning time series prediction strategy, the water quality changes and cultivation environment changes within a preset time period are predicted according to the current water quality parameters, the current cultivation environment parameters, the current fish and vegetable health status and the current near-ground meteorological condition parameters, and the fish farming warning level and the plant planting warning level are determined. The target control parameters of each aquaculture controllable power load module are determined in combination with the current status of multiple aquaculture controllable power load modules, and the corresponding aquaculture controllable power load modules are controlled. In this way, the problem that the background technology is difficult to cope with complex climate and sudden risks, resulting in low aquaculture production efficiency and a surge in electricity consumption is solved. By integrating Internet of Things sensors, intelligent algorithms and automated control, real-time dynamic regulation, fault warning, resilience improvement and cost reduction are achieved.

[0024] A second embodiment of the present application provides an aquaponics energy internet control device, comprising:

[0025] An acquisition module is used to obtain current water quality parameters, current cultivation environment parameters, current fish and vegetable health status, current near-ground meteorological condition parameters, and the current status of multiple aquaculture controllable power load modules;

[0026] A prediction module is used to predict changes in water quality and cultivation environment within a preset time period based on the current water quality parameters, the current cultivation environment parameters, the current fish and vegetable health conditions, and the current near-surface meteorological condition parameters based on a preset deep learning time series prediction strategy;

[0027] a determination module, configured to determine a warning level for fishery farming based on the water quality changes, and to determine a warning level for plant cultivation based on the cultivation environment changes;

[0028] A control module is used to determine the target control parameters of each aquaculture controllable power load module according to the current status of the multiple aquaculture controllable power load modules, the fishery aquaculture warning level, and the plant planting warning level, and to control the corresponding aquaculture controllable power load module according to the target control parameters of each aquaculture controllable power load module.

[0029] According to one embodiment of the present application, the control module is configured to:

[0030] Obtaining the current node marginal electricity price and current line congestion information, and based on a preset dynamic coordination optimization strategy, inputting the current node marginal electricity price and the current line congestion information into a pre-established power flow optimization sub-model to obtain a power flow optimization result;

[0031] Based on the grid flow optimization results, the current states of the multiple aquaculture controllable power load modules, the fishery aquaculture warning level and the plant planting warning level are respectively input into a pre-built fishery deep reinforcement learning model and a pre-built agricultural deep reinforcement learning model to obtain the target control parameters of each aquaculture controllable power load module.

[0032] According to one embodiment of the present application, the current water quality parameters include at least one of water temperature, dissolved oxygen content, pH, ammonia nitrogen content, nitrite concentration, water flow rate and liquid level.

[0033] According to one embodiment of the present application, the determining module is configured to:

[0034] Determining the interval of the water temperature, the interval of the dissolved oxygen content, the interval of the pH, the interval of the ammonia nitrogen content, the interval of the nitrite concentration, the interval of the water flow rate, and the interval of the liquid level;

[0035] The fishery breeding warning level is determined according to the water quality temperature range, the dissolved oxygen content range, the pH range, the ammonia nitrogen content range, the nitrite concentration range, the water flow rate range and the liquid level range, wherein the fishery breeding warning level includes the first to fourth levels, the warning level of the first level is less than the warning level of the second level, the warning level of the second level is less than the warning level of the third level, and the warning level of the third level is less than the warning level of the fourth level.

[0036] According to one embodiment of the present application, the current cultivation environment parameter includes at least one of air temperature, air humidity, light intensity and carbon dioxide concentration.

[0037] According to one embodiment of the present application, the determining module is configured to:

[0038] Determining a temperature interval for the air temperature, a humidity interval for the air humidity, a light interval for the light intensity, and a concentration interval for the carbon dioxide concentration;

[0039] The plant planting warning level is obtained according to the temperature range, the humidity range, the light range and the concentration range, wherein the plant planting warning level includes levels five to eight, the warning level of the fifth level is less than the warning level of the sixth level, the warning level of the sixth level is less than the warning level of the seventh level, and the warning level of the seventh level is less than the warning level of the eighth level.

[0040] According to one embodiment of the present application, the current fish and vegetable health status includes at least one of the fish farming status, fish survival rate, fish health status, farmed fish weight, plant planting status, plant survival rate, plant health status and plant planting area.

[0041] According to the energy internet control device for fish-vegetable symbiosis of the embodiment of the present application, based on a preset deep learning time series prediction strategy, the device predicts the water quality changes and cultivation environment changes within a preset time period according to the current water quality parameters, the current cultivation environment parameters, the current fish and vegetable health status and the current near-ground meteorological condition parameters, and determines the fishery farming warning level and the plant planting warning level. In combination with the current status of multiple aquaculture controllable power load modules, the target control parameters of each aquaculture controllable power load module are determined, and the corresponding aquaculture controllable power load module is controlled. In this way, the problem that the background technology is difficult to cope with complex climate and sudden risks, resulting in low aquaculture production efficiency and a surge in electricity consumption is solved. By integrating Internet of Things sensors, intelligent algorithms and automated control, real-time dynamic regulation, fault warning, resilience improvement and cost reduction are achieved.

[0042] A third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the energy Internet control method for fish-vegetable symbiosis as described in the above embodiment.

[0043] The fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the energy Internet control method for fish-vegetable symbiosis as described in the above embodiment.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] (1) The present invention primarily considers the system's ability to monitor the water quality parameters and cultivation environment parameters of the aquaponics system in real time, ensuring the accuracy and timeliness of the data. Furthermore, a deep learning algorithm is used to predict subsequent parameter changes. Based on preset alarm thresholds, the system can automatically issue early warning signals, alerting managers to take timely measures to reduce the impact of environmental anomalies on the cultured organisms.

[0046] (2) The present invention is based on a deep reinforcement learning algorithm, which ensures the accuracy and real-time performance of facility load control optimization. At the same time, through clear data visualization and user-friendly interface design, it provides users with a comprehensive, reliable and easy-to-understand management and control solution with high reliability and strong interpretability.

[0047] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0049] Figure 1 This is a flow chart of a method for controlling an energy internet for aquaponics according to an embodiment of the present application;

[0050] Figure 2 This is a schematic diagram showing the health status of fish according to one embodiment of the present application;

[0051] Figure 3 A schematic diagram showing plant health status according to one embodiment of the present application;

[0052] Figure 4 A schematic diagram of visualizing near-surface meteorological conditions according to one embodiment of the present application;

[0053] Figure 5 Schematic diagram of the power load composition of a system according to one embodiment of the present application;

[0054] Figure 6 This is a schematic diagram of the integration of fishery water quality visualization and early warning decision-making according to one embodiment of the present application;

[0055] Figure 7 This is a schematic diagram of the integration of cultivation environment visualization and early warning decision-making according to one embodiment of the present application;

[0056] Figure 8 This is a warning flow chart according to one embodiment of the present application;

[0057] Figure 9 Schematic diagram of a fishery reinforcement learning model according to one embodiment of the present application;

[0058] Figure 10 Schematic diagram of an agricultural reinforcement learning model according to one embodiment of the present application;

[0059] Figure 11 This is a schematic diagram of a temperature control module of an optimization control facility according to one embodiment of the present application;

[0060] Figure 12 This is a schematic diagram of an oxygen control module of an optimization control facility according to one embodiment of the present application;

[0061] Figure 13 This is a schematic diagram of an ammonia nitrogen content control module of an optimization control facility according to one embodiment of the present application;

[0062] Figure 14 This is a schematic diagram of a microbial control module of an optimized control facility according to one embodiment of the present application;

[0063] Figure 15 Schematic diagram of a liquid level control module of an optimization control facility according to one embodiment of the present application;

[0064] Figure 16 Schematic diagram of a bait feeding amount control module of an optimization control facility according to one embodiment of the present application;

[0065] Figure 17 Schematic diagram of a humidity control module of an optimization control facility according to one embodiment of the present application;

[0066] Figure 18 A schematic diagram of a lighting control module of an optimization control facility according to one embodiment of the present application;

[0067] Figure 19 Schematic diagram of a flow rate control module of an optimization control facility according to one embodiment of the present application;

[0068] Figure 20 This is a schematic diagram of a carbon dioxide control module of an optimization control facility according to one embodiment of the present application;

[0069] Figure 21 This is a flow chart of a method for controlling an energy internet for aquaponics according to one embodiment of the present application;

[0070] Figure 22 Schematic diagram of a block diagram of an energy internet control device for aquaponics according to an embodiment of the present application;

[0071] Figure 23 Schematic diagram of the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0072] The following describes in detail embodiments of the present application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0073] The following describes the energy internet control method, device, electronic device and medium for fish and vegetable symbiosis according to the embodiment of the present application with reference to the accompanying drawings. In response to the problems mentioned in the above background technology that it is difficult to cope with complex climate and sudden risks, resulting in low aquaculture production efficiency and a surge in electricity consumption, the present application provides an energy internet control method for fish and vegetable symbiosis. In this method, based on a preset deep learning time series prediction strategy, the water quality changes and cultivation environment changes within a preset time period are predicted according to the current water quality parameters, the current cultivation environment parameters, the current fish and vegetable health status and the current near-surface meteorological condition parameters, and the fishery farming warning level and the plant planting warning level are determined. The target control parameters of each aquaculture controllable electricity load module are determined in combination with the current status of multiple aquaculture controllable electricity load modules, and the corresponding aquaculture controllable electricity load module is controlled. In this way, the problem that the background technology is difficult to cope with complex climate and sudden risks, resulting in low aquaculture production efficiency and a surge in electricity consumption is solved. By integrating Internet of Things sensors, intelligent algorithms and automated control, real-time dynamic regulation, fault warning, resilience improvement and cost reduction are achieved.

[0074] Specifically, Figure 1 A flow chart of a method for controlling an energy internet for aquaponics provided in an embodiment of the present application.

[0075] like Figure 1 As shown, the aquaponics energy internet control method includes the following steps:

[0076] In step S101, current water quality parameters, current cultivation environment parameters, current fish and vegetable health status, current near-ground meteorological condition parameters and current status of multiple aquaculture controllable power load modules are obtained.

[0077] It is understandable that in the fish-vegetable symbiotic system, the growth of fish is directly related to the water quality, and the growth of vegetables is directly related to the cultivation environment. First, it is necessary to establish a water environment and a cultivation environment model, and model various parameters of the water body and the cultivation environment.

[0078] The current water quality parameters include at least one of water temperature, dissolved oxygen content, pH, ammonia nitrogen content, nitrite concentration, water flow rate and liquid level. The current cultivation environment parameters include at least one of air temperature, air humidity, light intensity and carbon dioxide concentration. The current fish and vegetable health status includes the fish health status and plant health status, such as Figure 2 As shown, the fish health status may include at least one of the fish farming status, fish survival rate, fish health status, and farmed fish weight; Figure 3 As shown, the plant health status may include at least one of the planting status of the plant, the survival rate of the plant, the health status of the plant, and the planting area of ​​the plant.

[0079] Alternatively, as Figure 4 As shown, the current near-surface meteorological condition parameters include at least one of ambient temperature, ambient humidity, light intensity and carbon dioxide concentration.

[0080] Optionally, the multiple controllable power load modules for aquaculture may include at least one of a temperature control module, an oxygen control module, an ammonia nitrogen content control module, a microbial control module, a water flow rate control module, a liquid level control module, a feed amount control module, a humidity control module, a light control module, a carbon dioxide control module and an energy storage device control module.

[0081] In one embodiment of the present application, 12 breeding ponds are used, each with a diameter of 2 meters and a liquid level of 70 centimeters. 29 sturgeons are cultured in each breeding pond. Four plots of land are used for planting, one foam box is used for planting, one hydroponic planting is used, and one column hydroponic planting is used. The area of ​​each plot of land is 54.76 square meters, and the area of ​​each foam box is 53.76 square meters. A multi-parameter water quality collector with a power of 2W is arranged in the fishery breeding area to collect water quality parameters such as temperature, dissolved oxygen content, pH, ammonia nitrogen content, nitrite concentration, and water flow rate. An ultrasonic liquid level meter with a power of 1.5W collects liquid level information. Ensure that the sensor can collect water quality data accurately and in real time. An air temperature and humidity sensor with a power of 0.015W is arranged in the plant planting area to collect air temperature and humidity, a digital illuminometer with a power of 2W collects light intensity, and a carbon dioxide sensor with a power of 0.5W collects carbon dioxide concentration. The system pushes the received water quality parameters and cultivation environment parameters to the system interface in real time. The system interface uses intuitive charts and numbers to display the current values ​​of various parameters, historical change trends and other information, making it convenient for operators to view and monitor at any time.

[0082] Optionally, the current state of the temperature control module includes the switch state, set temperature, etc. of the ultra-low temperature air-energy water heater; the current state of the oxygen control module includes the number of precision oxygen generators that are turned on, the number of turned off, and the power, etc.; the current state of the ammonia nitrogen content control module includes the switch state, power, and start time of the blower in the biochemical tank; the current state of the microbial control module includes the switch state, power, and start time of the ultraviolet lamp; the current state of the water flow rate control module includes the switch state, power, and speed of the variable frequency circulating water pump; the current state of the liquid level control module includes the switch state, power, and start time of the water pump; the current state of the feed amount control module includes the switch state, power, and start time of the feed machine; the current state of the humidity control module includes the switch state, power, and start time of the multi-function spray system; the current state of the lighting control module includes the switch state, power, and start time of the fill light; the current state of the carbon dioxide control module includes the switch state, power, and start time of the carbon dioxide generator; the current state of the energy storage device control module includes the charging and discharging power of the energy storage device, etc.

[0083] For example, the current state of the temperature control module of the embodiment of the present application may include the switch state, set temperature, etc. of the ultra-low temperature air-energy water heater with a cooling capacity of 78KW and a heating capacity of 125KW. The current state of the oxygen control module may include the number of open and closed units, the power, etc. of the precision oxygen generator with a rated power of 1.4kW and an oxygen output of 0-35L / min. The current state of the ammonia nitrogen content control module may include the switch state, power, start time, etc. of the blower with a rated power of 5.5KW; the current state of the microbial control module may include: the switch state, power, start time, etc. of the ultraviolet lamp with a rated power of 0.05kW; the current state of the water flow rate control module may include: the switch state, power, speed, etc. of the variable frequency circulating water pump with a rated power of 2kW; the current state of the liquid level control module may include: the switch state, power, start time, etc. of the water pump with a rated power of 0.37kW; the current state of the feeding amount control module may include: The switch status, power, start time, etc. of the bait thrower with a rated power of 0.75kW; the current status of the humidity control module may include: the switch status, power, start time, etc. of the multi-function spray system with a rated power of 2kW; the current status of the lighting control module may include: the switch status, power, start time, etc. of the fill light with a rated power of 1kW; the current status of the carbon dioxide control module may include: the switch status, power, start time, etc. of the carbon dioxide generator with a rated power of 1kW; the current status of the energy storage device control module may include: the charging and discharging power of the energy storage device, etc.

[0084] The power load composition of this example is as follows Figure 5As shown, in this example, smart meters, power sensors and other equipment are installed on the controllable power load modules of aquaculture, such as ultra-low temperature air-energy water heaters, precision oxygen generators, blowers, ultraviolet lamps, water pumps, variable frequency circulating water pumps, feeding machines, multi-function spray systems, fill lights, and carbon dioxide generators to monitor the operating status (on / off) and power consumption of the equipment in real time. The monitored power load module status data is transmitted to the system server through the communication network and displayed in the system interface in the form of lists, graphics, etc.

[0085] In step S102, based on the preset deep learning time series prediction strategy, the water quality changes and cultivation environment changes within a preset time period are predicted according to the current water quality parameters, current cultivation environment parameters, current fish and vegetable health conditions and current near-ground meteorological conditions parameters.

[0086] The preset duration can be a duration predetermined by a person skilled in the art, such as two hours, and is not specifically limited herein. The predicted water quality changes include water temperature, dissolved oxygen content, pH, ammonia nitrogen content, nitrite concentration, water flow rate, and liquid level. The predicted cultivation environment changes include air temperature, air humidity, light intensity, and carbon dioxide concentration.

[0087] Specifically, in this example, weather station equipment is used to obtain environmental parameters. These water quality parameters, cultivation environment parameters, and environmental input parameters undergo preprocessing, including cleaning, denoising, and normalization. Outliers and missing values ​​are removed, and parameters with varying ranges are uniformly scaled to the [0, 1] interval for subsequent model processing. The model inputs are preprocessed historical water quality parameters, cultivation environment parameters, and environmental input parameters, and the output is a predicted value for each water quality parameter and cultivation environment parameter at 15-minute intervals for the next two hours. The model is trained using a large amount of historical data, and its parameters are continuously adjusted through a backpropagation algorithm, enabling it to accurately learn the patterns of water quality parameter changes.

[0088] Furthermore, the method of the embodiment of the present application for predicting changes in water quality and cultivation environment within a preset time period based on current water quality parameters, current cultivation environment parameters, current fish and vegetable health conditions and current near-ground meteorological conditions parameters is a time series prediction method based on deep learning. First, a one-dimensional convolutional neural network is used to process the original data sequence to extract long-term stable and periodic trends. Then, LSTM (Long Short-Term Memory) is used to learn the characteristics of water quality data changing over time. Then, through a linear layer combined with a channel-independent method, the model predicts each variable separately and can adjust the length of the prediction sequence. Finally, the model combines all the prediction results to generate the final output sequence.

[0089] It should be noted that the training of the prediction model in the embodiment of the present application includes the following steps:

[0090] First, the collected water quality dataset and cultivation environment dataset were preprocessed separately. After preprocessing, the datasets were divided into training and test sets. The training set was used for model training and learning, while the test set was used to evaluate the model's performance and generalization ability.

[0091] Secondly, the original data sequence is processed using CNN (Convolutional Neural Networks). CNN can automatically learn the features in the data and extract the long-term stable and periodic trends, which reflect the periodic changes in the data.

[0092] Thirdly, by leveraging the temporal characteristics of the LSTM learning data and combining it with a linear layer in conjunction with a channel-independent approach, the model is able to output the predicted value of each variable and adjust the length of the prediction sequence as needed.

[0093] Finally, the prediction results of all variables are combined to obtain the final output sequence.

[0094] In step S103, the fishery farming warning level is determined according to the change in water quality, and the plant planting warning level is determined according to the change in the cultivation environment.

[0095] Specifically, the embodiment of the present application uses the prediction results of the current water quality parameters and the prediction results of the cultivation environment parameters to provide an early warning of the subsequent status of the current water quality and cultivation environment parameters in the system.

[0096] Furthermore, in some embodiments, the fishery breeding warning level is determined according to the water quality changes, including: determining the water temperature interval, the dissolved oxygen content interval, the pH interval, the ammonia nitrogen content interval, the nitrite concentration interval, the water flow rate interval and the liquid level interval; the fishery breeding warning level is determined according to the water temperature interval, the dissolved oxygen content interval, the pH interval, the ammonia nitrogen content interval, the nitrite concentration interval, the water flow rate interval and the liquid level interval, wherein the fishery breeding warning levels include the first to fourth levels, the warning level of the first level is less than the warning level of the second level, the warning level of the second level is less than the warning level of the third level, and the warning level of the third level is less than the warning level of the fourth level.

[0097] For example, the first level warning may be a normal state, the second level warning may be a mild warning, the third level warning may be a moderate warning, and the fourth level warning may be a severe warning, which is not specifically limited here.

[0098] For example, the warning status classification method of the fishery farming warning level in the embodiment of the present application can be: first level warning (water temperature is 20-28°C; dissolved oxygen content is above 5mg / L; pH value is between 6.5-8.5; ammonia nitrogen content is lower than 0.2mg / L; nitrite content is lower than 0.1mg / L; water flow rate is 0.1-0.3m / s; liquid level is 70cm, etc.), second level warning (water temperature is 15-20°C or 28-30°C; dissolved oxygen content is between 4-5mg / L; pH value is between 6.0-6.5 or 8.5-9.0; ammonia nitrogen content is between 0.2-0.5mg / L; nitrite content is between 0.1-0.2mg / L; water flow rate is between 0.3-0.35m / s; liquid level is between 60-70cm or 7 Level 1 warning (water temperature is between 10-15℃ or 30-35℃; dissolved oxygen content is between 3-4mg / L; pH value is between 5.5-6.0 or 9.0-9.5; ammonia nitrogen content is between 0.5-1.0mg / L; nitrite content is between 0.2-0.5mg / L; water flow rate is between 0.35-0.4m / s; liquid level is between 50-60cm or 80-90cm, etc.) and level 4 warning (water temperature is less than 10℃ or greater than 35℃; dissolved oxygen content is less than 3mg / L; pH value is less than 5.5 or greater than 9.5; ammonia nitrogen content exceeds 1.0mg / L; nitrite content exceeds 0.5mg / L; water flow rate is less than 0.1m / s or greater than 0.4m / s; liquid level is less than 50cm or greater than 90cm, etc.).

[0099] For example, the integrated interface of visualization and early warning of water quality parameters in this application is as follows: Figure 6 As shown, Figure 6 The parameters corresponding to sequence number 4 are: pH 7.44Ph, temperature 21.54℃, dissolved oxygen content 4.99mg / L, liquid level 70cm, water flow rate 0.2m / s, nitrite content 0.04mg / L, and ammonia nitrogen content 0.10mg / L. The current fishery aquaculture warning level is determined to be a minor warning.

[0100] Furthermore, in some embodiments, the plant planting warning level is determined according to changes in the cultivation environment, including: determining the temperature range of the air temperature, the humidity range of the air humidity, the light range of the light intensity, and the concentration range of the carbon dioxide concentration; obtaining the plant planting warning level according to the temperature range, humidity range, light range, and concentration range, wherein the plant planting warning level includes levels five to eight, the warning level of the fifth level is less than the warning level of the sixth level, the warning level of the sixth level is less than the warning level of the seventh level, and the warning level of the seventh level is less than the warning level of the eighth level.

[0101] For example, the fifth level warning may be a normal state, the sixth level warning may be a mild warning, the seventh level warning may be a moderate warning, and the eighth level warning may be a severe warning, which are not specifically limited here.

[0102] For example, the embodiment of the present application may divide the warning status of plant planting into: fifth-level warning (air temperature is 18-28°C; air humidity is 40%-70%; light intensity is between 1000lux-5000lux; carbon dioxide concentration is between 500ppm-1000ppm, etc.), sixth-level warning (air temperature is 12-18°C or 28-30°C; air humidity is 30%-40% or 70%-80%; light intensity is between 800lux-1000lux or 5000lux-8000lux; carbon dioxide concentration is between 200ppm-500ppm or 1000ppm). -1500ppm, etc.), seventh level warning (air temperature is 8-12℃ or 30-35℃; air humidity is 25%-30% or 80%-85%; light intensity is between 500lux-800lux or 8000lux-10000lux; carbon dioxide concentration is between 150ppm-200ppm or 1500ppm-2000ppm, etc.) and eighth level warning (air temperature is less than 8℃ or greater than 35℃; air humidity is less than 25% or higher than 85%; light intensity is less than 500lux or higher than 10000lux; carbon dioxide concentration is less than 150ppm or higher than 2000ppm, etc.).

[0103] For example, the integrated interface of visualization and early warning of cultivation environment parameters in the embodiment of the present application is as follows: Figure 7 As shown, Figure 7 The parameters corresponding to sequence number 4 are: air temperature of 18.2°C, air humidity of 35.0%, light intensity of 2273 lux, and carbon dioxide concentration of 504 ppm, which means that the current warning status of plant cultivation is determined to be slightly abnormal.

[0104] In this example, the predicted water quality and cultivation environment parameters for the next two hours are compared with their corresponding warning thresholds. If the predicted values ​​exceed the normal range and reach the corresponding warning level threshold, an alert is triggered. Simultaneously, the system displays the alert information using prominent colors, icons, and text, informing operators of the current parameter warning level and potential problems.

[0105] The early warning process of the embodiment of the present application can be as follows Figure 8 As shown, the following steps are included:

[0106] S801, obtaining data transmitted by the sensor.

[0107] S802, visualize water quality parameters and cultivation environment parameters on the platform.

[0108] S803, predicting water quality and cultivation environment based on deep neural network.

[0109] S804, calling the prediction module to implement real-time warning.

[0110] S805: Issue an early warning when an abnormal trend is detected.

[0111] In step S104, the target control parameters of each aquaculture controllable power load module are determined according to the current status of multiple aquaculture controllable power load modules, the fishery aquaculture warning level, and the plant planting warning level, and the corresponding aquaculture controllable power load module is controlled according to the target control parameters of each aquaculture controllable power load module.

[0112] Furthermore, in some embodiments, the target control parameters of each aquaculture controllable power load module are determined based on the current status of multiple aquaculture controllable power load modules, the fishery aquaculture warning level, and the plant planting warning level, including: obtaining the current node marginal electricity price and the current line congestion information, and based on a preset dynamic coordination optimization strategy, inputting the current node marginal electricity price and the current line congestion information into a pre-established power grid flow optimization sub-model to obtain a power grid flow optimization result; based on the power grid flow optimization result, inputting the current status of multiple aquaculture controllable power load modules, the fishery aquaculture warning level, and the plant planting warning level into a pre-constructed fishery deep reinforcement learning model and a pre-constructed agricultural deep reinforcement learning model respectively to obtain the target control parameters of each aquaculture controllable power load module.

[0113] The embodiment of the present application optimizes the power load control scheme by combining the current status of multiple aquaculture controllable power load modules, the fishery aquaculture warning level, and the plant cultivation warning level. The optimization goal is to reduce power consumption, ensure optimal power grid flow, and ensure the best water quality environment and cultivation environment to achieve the best fish and vegetable yield.

[0114] Specifically, a power load model for each controllable power load module of aquaculture will be established for subsequent optimization. This approach uses deep reinforcement learning to develop an intelligent energy internet control method for aquaponics. By learning from existing experience, deep reinforcement learning state spaces for fisheries and agriculture are constructed based on microclimate, water environment, cultivation environment, fish and vegetable health, and power load status. By optimizing power load control schemes and the charge and discharge power of energy storage devices, the control and optimization of the water and cultivation environments are achieved. A power grid flow optimization sub-model is also established, using a dynamic coordinated optimization strategy to implement data interaction and feedback mechanisms to ensure optimal power grid flow.

[0115] In this example, a power load model was established for each controllable power load module in aquaculture. This model takes the operating parameters of electrical equipment (such as the light intensity of a fill light) as input and outputs the corresponding equipment's real-time power. This modeling approach allows for optimization of equipment parameters and power in a virtual environment during reinforcement learning training, thus avoiding the risk of fish, vegetables, and other organisms dying from direct testing in the actual system.

[0116] The power consumed by the temperature control module in the system is: P thermostiact =P tem ×n tem ;

[0117] Among them, P tem is the power of a single ultra-low temperature air energy water heater, in kW, n tem The number of ultra-low temperature air-energy water heater devices turned on.

[0118] The oxygenation power of the system is: P ox =P aerator ×n ox ;

[0119] Among them, P aerato is the rated power of each precision oxygen concentrator, in kW, n ox The number of precision oxygen concentrators enabled.

[0120] The power of the ammonia nitrogen content control module of the system is: P blower =P blo ×n blo ;

[0121] Among them, P blo is the power of a single blower in kW, n blo The number of blowers turned on.

[0122] The power of the microbial control module of the system is: P ultraviolte =P ult ×n ult ;

[0123] Among them, P ult is the power of a single UV lamp in kW, n ult The number of UV lamps turned on.

[0124] The power of the liquid level control module of the system is: P liquid =P liq ×n liq ;

[0125] Among them, P liq is the power of a single water pump in kW, n liq The number of water pumps that are turned on.

[0126] The power of the water flow rate control module of the system is: P flve =P fl ×n fl ;

[0127] Among them, P fl is the power of a single water pump in kW, n fl The number of water pumps that are turned on.

[0128] The power of the system's bait feeding control module is:

[0129] Among them, P feeder is the feeding power of each feeding machine, in kW, n feed is the number of times bait is fed per day on the current typical day, H is the total amount of bait fed, in kg, and H = FRL S Calculate, where L S is the fish survival rate, R is the environmental impact factor, which is 0.990, and F is the empirical feeding amount, in kg, from F = Z × L f Calculate, where L f is the feeding rate, which refers to the percentage of the total mass of the bait fed at this stage to the fish body mass, and is related to the growth stage of the fish. Z is the total mass of the fish, and Z = 10 -3 GDS is obtained, G is the body mass of the fish at the current stage, the unit is g, D is the stocking density, the unit is tail / m 2 , S is the total area of ​​the pond, in m 2 .

[0130] It is understandable that the determination of feeding time and frequency is related to the season. A single feeding time of 1 hour is appropriate. The typical daily feeding time and frequency in different seasons are shown in Table 1:

[0131] Table 1

[0132]

[0133] The power consumed by the multifunctional spray system is: P spray =P spr ×n spr ;

[0134] Among them, P spr is the power of a single spray device in kW, n spr The number of spray devices opened.

[0135] The power consumed by the fill light in the system is: P lightload =P e ×n light ;

[0136] Among them, P e is the power of a single fill light, in kW, n light The number of fill lights turned on.

[0137] The power consumed by the carbon dioxide control module in the system is: P CO2 =P c ×n c ;

[0138] Among them, P c is the power of a single carbon dioxide generator in kW, n c The number of CO2 generators turned on.

[0139] Furthermore, in this example, a power flow optimization sub-model is established. Specifically, by embedding the power flow optimization sub-model within the deep reinforcement learning framework, its objective function is to minimize the total power grid operating cost and network loss. The constraints include:

[0140] (1) Node power balance equation: ( For conventional power output, P 储能 is the charging and discharging power of the energy storage device, P 负荷 is the total aquaculture load demand, P loss is network loss);

[0141] (2) Line transmission capacity limitation: Avoid line overload;

[0142] (3) Voltage safety constraint: V i min ≤V i ≤V i max , ensuring node voltage stability.

[0143] Furthermore, a dynamic coordination optimization strategy is implemented through priority hierarchical control and multi-objective weight adaptation. The dynamic coordination optimization strategy specifically includes:

[0144] The conflicting goals of water quality control and grid power flow optimization are coordinated through:

[0145] (1) Priority hierarchical control: In the event of a grid emergency (such as voltage exceeding the limit or line overload), priority is given to adjusting the energy storage charging and discharging power and interruptible loads (such as non-essential bait dispensers and water pump control boxes) to quickly restore grid security.

[0146] (2) Multi-objective weight adaptation: Dynamically adjust the weight coefficients of "water quality optimization" and "tidal flow optimization" based on real-time electricity prices and grid congestion signals. For example, during peak electricity price periods, priority is given to reducing the peak power of aquaculture loads; during grid congestion periods, priority is given to reducing network losses and balancing line power distribution.

[0147] (3) Data interaction and feedback mechanism: By receiving the node marginal price (LMP) and line congestion information released by the grid dispatching system as optimization input; the optimized load control plan (such as energy storage charging and discharging plan, water pump control box start and stop sequence) is fed back to the grid dispatching center to participate in demand response, thereby realizing data interaction and feedback mechanism to ensure the optimal grid flow.

[0148] Construct fishery deep reinforcement learning models and agricultural deep reinforcement learning models respectively. First, construct the fishery deep reinforcement learning model, such as Figure 9 As shown in FIG, the water environment, cultivation environment, warning information, meteorological information, power load module status and current power state of energy storage equipment are taken as the state space and recorded as S. S={s|s ＝ [Weather information, water quality parameters, early warning information, fishery load status, current energy storage device power]}. The control of power loads such as ultra-low-temperature air-energy water heaters, precision oxygen generators, blowers, UV lamps, water pumps, bait dispensers, variable-frequency circulating water pumps, and carbon dioxide generators, as well as the charge and discharge power of the energy storage device, is used as the action space, denoted as A = {a1, a2, a3, a4…}. The power grid flow results and the output of the fishery module are used as the reward space. A cost-based reward scheme is designed, calculating the output and power grid flow results as costs and using them as the basis for rewards, denoted as R.

[0149] Similar, such as Figure 10 As shown in Figure 1, an agricultural deep reinforcement learning model is constructed, and the cultivation environment, warning information, meteorological information, power load module status, and current power state of the energy storage device are used as the state space and recorded as S. That is, S = {s|s ＝[Weather information, cultivation environment parameters, early warning information, agricultural load status, current energy storage device power]}. The control of loads such as ultra-low-temperature air-energy water heaters, precision oxygen generators, water pumps, carbon dioxide generators, multi-function spray systems, and fill lights, as well as the charge and discharge power of the energy storage device, is used as the action space, denoted as A = {a1, a2, a3, a4…}. The grid flow results and the output of the agricultural module are used as the reward space. A cost-based reward scheme is designed, calculating the output and grid flow results as costs and using them as the basis for rewards, denoted as R.

[0150] For the two models above, we construct an actor network and a critic network, respectively: The actor network uses a three-layer fully connected neural network, with nonlinear activation functions for feature extraction and transformation. Input: state s in state space S. Output: probability distribution π(a|s) of A in action space A, representing the probability of taking action a in state s. The critic network also uses a three-layer fully connected neural network, with nonlinear activation functions for feature extraction and transformation. Input: state s in state space S and action a in action space A. Output: action value Q(s|a), representing the value of taking action a in state s.

[0151] Furthermore, water quality parameters, cultivation environment parameters, electrical equipment status, micro-meteorological forecast results, and warning information are encoded as part of the state and input into the actor network and critic network. The actor network outputs an action probability distribution based on the current state and randomly selects the next action based on the probability distribution. The next action is executed and interacts with the environment to obtain the next state and reward. The state, action, reward, and next state of each step are stored in the experience replay pool.

[0152] Specifically, during the interaction between the model and the environment, for each time step t: the actor network is based on the current state s t Output action probability distribution π(a|s t ), and randomly select action a t . Execute action a t Interact with the environment to get the next state s t+1 and reward r t . The state s of each step t 、Action a t , reward rt and next state s t+1 Stored in the experience replay pool D, that is, D←D∪{(s t ,a t ,r t ,s t+1 )}.

[0153] Randomly draw a certain number of samples {(s t ,a t ,rt ,s t+1 )} for training.

[0154] Update the critic network: calculate the target Q value y i =r i +γQ target (s i+1 ,π target (s i+1 ), where γ is the discount factor, Q target is the target critic network, π target is the target actor network. Using the mean squared error loss function To update the parameters of the critic network, where N is the number of samples. Update the actor network: Use the policy gradient algorithm to update the parameters of the actor network, with the goal of maximizing the long-term reward. The objective function of the policy gradient is Where θ is the parameter of the actor network. The parameter θ of the actor network is updated by gradient ascent method, that is, where α is the learning rate.

[0155] Furthermore, the above interaction steps are repeated until the maximum number of training steps is reached, striving to obtain the maximum long-term reward.

[0156] Finally, the optimized control facilities of the embodiment of the present application are displayed in the system interface.

[0157] For example, the ultra-low temperature air energy water heater of the embodiment of the present application is optimized and controlled. Figure 11 As shown, device 101 is in the on state, the temperature is 20° C., the power consumption is 80KW, and the state is normal; device 102 is in the off state.

[0158] Optimize the control facilities of precision oxygen concentrators such as Figure 12 As shown, device 201 is in the off state; device 202 is in the on state, the oxygen flow rate is 14 L / min, the power consumption is 1 KW, and the state is normal.

[0159] Optimize the control facilities of the blower such as Figure 13 As shown, device 301 is in the closed state; device 302 is in the open state, and the air volume is 5m 3 / min, power consumption is 4.5KW, and the status is normal.

[0160] Optimize the control of UV lamps in facilities such as Figure 14 As shown, the device 701 is in the on state and the UV dose is 15mJ / cm 3 , power consumption is 0.04KW, status is normal; device 702-device 704 is in shutdown state.

[0161] Optimize the control facilities of water pumps such as Figure 15 As shown, the device 401 is in the on state, the pump start liquid level is 60 cm, the power consumption is 0.3 KW, and the state is normal.

[0162] Optimize the control facilities of the bait dispenser such as Figure 16 As shown, device 801 is in the open state, the feeding amount is 300kg / h, the power consumption is 0.04KW, and the state is normal; devices 802-804 are in the closed state.

[0163] Multifunctional spray systems for optimized control facilities such as Figure 17 As shown, device 501 is in the open state, the single nozzle flow rate is 0.5L / min, the power consumption is 1.5KW, and the state is normal; device 502 is in the closed state.

[0164] Optimize the status of the fill light of the control facility such as Figure 18 As shown, device 601 is in the on state, the illumination intensity is 2000 lux, the power consumption is 0.8KW, and the state is normal; devices 602-604 are in the off state.

[0165] Optimize the status of the variable frequency circulating water pump of the control facility Figure 19 As shown, the device 901 is in the on state, the rotation speed is 1500 rpm, the power consumption is 1.5KW, and the status is normal.

[0166] Optimize the status of the CO2 generator in the control facility such as Figure 20 As shown, device 1001 is in the open state, the flow rate is 0.4L / min, the power consumption is 0.5KW, and the state is normal; device 1002 is in the closed state.

[0167] Based on the traditional fish-vegetable symbiosis system, this application carefully depicts the impact of the water quality environment of facility fisheries on fish growth and the impact of the cultivation environment on plant growth status. By optimizing the regulation of power load and the charging and discharging power of energy storage equipment, the optimal power grid flow and the optimal water quality environment and cultivation environment are achieved, thereby realizing intelligent management and efficient operation of fish-vegetable symbiosis, significantly improving the efficiency and safety of fish-vegetable symbiosis, ensuring the stability of the power grid, reducing the impact of abnormal aquaculture environment on farmed organisms, reducing labor costs and labor intensity, and providing strong technical support for the green, ecological and sustainable development of the fish-vegetable symbiosis industry.

[0168] In order to facilitate those skilled in the art to more clearly and intuitively understand the energy Internet control method of fish and vegetable symbiosis proposed in this application, the following is combined with Figure 21 Provide detailed explanation.

[0169] Specifically, if Figure 21 As shown, the aquaponics energy internet control method includes the following steps:

[0170] S2101, obtain water quality parameters, cultivation environment parameters and fish and vegetable health status input by sensors, and display them in the system interface.

[0171] S2102, obtaining near-surface meteorological condition parameters, combining the water quality and cultivation environment parameters of step S2101, and predicting the changes of various parameter values ​​of water quality and cultivation environment within 2 hours.

[0172] S2103: Based on the prediction result of step S2102, an early warning is issued in the system regarding the subsequent status of the parameters of the current water quality and cultivation environment.

[0173] S2104, obtain the status of each controllable power load module of the breeding industry and display it in the system interface.

[0174] S2105 , combining the results of steps S2101 , S2103 and S2104 , and optimizing the power load control scheme.

[0175] Therefore, in order to improve the efficiency of fish-vegetable symbiotic production, the embodiment of the present application needs to realize real-time monitoring and accurate early warning of water quality and cultivation environment, so as to intervene in time and ensure the high yield and effectiveness of fish-vegetable symbiotic production. Wherein water quality is related to water temperature, dissolved oxygen content, pH, ammonia nitrogen content, nitrite concentration, water flow rate and liquid level, while cultivation environment is related to air temperature, air humidity, light intensity and carbon dioxide concentration. In combination with actual conditions, the present invention uses water temperature, dissolved oxygen content, pH, ammonia nitrogen content, nitrite concentration, water flow rate and liquid level as input parameters to realize the visualization of water quality parameters, uses air temperature, air humidity, light intensity and carbon dioxide concentration as input parameters to realize the visualization of cultivation environment parameters, and on this basis, combines relevant algorithms such as reinforcement learning, appropriately sets thresholds, realizes the prediction of water quality parameters and cultivation environment parameters, and performs normal, slight, moderate and severe early warning for the subsequent states of water quality parameters and cultivation environment parameters based on the prediction results.

[0176] At the same time, the system also obtains and displays the status of the controllable power load module of the aquaculture, including temperature, oxygen, ammonia nitrogen content, microorganisms, water flow rate, liquid level, feed amount, humidity, light, and energy storage equipment control module information. In combination with water quality parameters, cultivation environment parameters, early warning information and power load status, the system uses an intelligent control method based on deep reinforcement learning to optimize the power load control plan, aiming to reduce power consumption, ensure optimal grid current and water quality environment, thereby realizing intelligent management and efficient operation of fish and vegetable symbiosis. This not only significantly improves production efficiency and safety, reduces the impact of environmental anomalies on farmed organisms, but also reduces labor costs and labor intensity, providing strong technical support for the green, ecological, and sustainable development of fish and vegetable symbiosis. At the same time, this method is highly reliable and interpretable.

[0177] According to the energy internet control method for fish-vegetable symbiosis of the embodiment of the present application, based on the preset deep learning time series prediction strategy, the water quality changes and cultivation environment changes within a preset time period are predicted according to the current water quality parameters, the current cultivation environment parameters, the current fish and vegetable health status and the current near-ground meteorological condition parameters, and the fishery breeding warning level and the plant planting warning level are determined. The target control parameters of each breeding controllable electricity load module are determined in combination with the current status of multiple breeding controllable electricity load modules, and the corresponding breeding controllable electricity load module is controlled. In this way, the problem that the background technology is difficult to cope with complex climate and sudden risks, resulting in low breeding production efficiency and a surge in electricity consumption is solved. By integrating Internet of Things sensors, intelligent algorithms and automated control, relevant parameters are monitored in real time, problems can be discovered and intervened early, and real-time dynamic regulation, fault warning, resilience improvement and cost reduction are achieved, with high reliability and strong interpretability.

[0178] Next, the energy internet control device for fish-vegetable symbiosis proposed in accordance with an embodiment of the present application will be described with reference to the accompanying drawings.

[0179] Figure 22 It is a block diagram of the energy Internet control device for fish and vegetable symbiosis according to an embodiment of the present application.

[0180] like Figure 22 As shown, the aquaponics energy internet control device 10 includes: an acquisition module 100 , a prediction module 200 , a determination module 300 and a control module 400 .

[0181] Among them, the acquisition module 100 is used to obtain the current water quality parameters, the current cultivation environment parameters, the current fish and vegetable health status, the current near-surface meteorological condition parameters and the current status of multiple aquaculture controllable power load modules; the prediction module 200 is used to predict the water quality changes and cultivation environment changes within a preset time period based on the current water quality parameters, the current cultivation environment parameters, the current fish and vegetable health status and the current near-surface meteorological condition parameters based on the preset deep learning time series prediction strategy; the determination module 300 is used to determine the fishery aquaculture warning level according to the water quality changes, and determine the plant planting warning level according to the cultivation environment changes; the control module 400 is used to determine the target control parameters of each aquaculture controllable power load module according to the current status of multiple aquaculture controllable power load modules, the fishery aquaculture warning level and the plant planting warning level, and control the corresponding aquaculture controllable power load module according to the target control parameters of each aquaculture controllable power load module.

[0182] Furthermore, in some embodiments, the control module 400 is used to: obtain the current node marginal electricity price and the current line congestion information, and based on a preset dynamic coordination optimization strategy, input the current node marginal electricity price and the current line congestion information into a pre-established power flow optimization sub-model to obtain a power flow optimization result; based on the power flow optimization result, input the current status of multiple aquaculture controllable power load modules, the fishery aquaculture warning level and the plant planting warning level into a pre-built fishery deep reinforcement learning model and a pre-built agricultural deep reinforcement learning model respectively to obtain the target control parameters of each aquaculture controllable power load module.

[0183] Furthermore, in some embodiments, the current water quality parameter includes at least one of water temperature, dissolved oxygen content, pH, ammonia nitrogen content, nitrite concentration, and liquid level.

[0184] Further, in some embodiments, the determination module 300 is used to: determine the interval of water temperature, the interval of dissolved oxygen content, the interval of pH, the interval of ammonia nitrogen content, the interval of nitrite concentration, the interval of water flow rate and the interval of liquid level; determine the fishery breeding warning level according to the interval of water temperature, the interval of dissolved oxygen content, the interval of pH, the interval of ammonia nitrogen content, the interval of nitrite concentration, the interval of water flow rate and the interval of liquid level, wherein the fishery breeding warning level includes first to fourth levels, the warning level of the first level is less than the warning level of the second level, the warning level of the second level is less than the warning level of the third level, and the warning level of the third level is less than the warning level of the fourth level.

[0185] Furthermore, in some embodiments, the current cultivation environment parameter includes at least one of air temperature, air humidity, light intensity, and carbon dioxide concentration.

[0186] Furthermore, in some embodiments, the determination module 300 is used to: determine the temperature range of the air temperature, the humidity range of the air humidity, the light range of the light intensity, and the concentration range of the carbon dioxide concentration; obtain the plant planting warning level according to the temperature range, humidity range, light range, and concentration range, wherein the plant planting warning level includes levels five to eight, the warning level of the fifth level is less than the warning level of the sixth level, the warning level of the sixth level is less than the warning level of the seventh level, and the warning level of the seventh level is less than the warning level of the eighth level.

[0187] Further, in some embodiments, the current fish and vegetable health status includes at least one of the fish farming status, fish survival rate, fish health status, farmed fish weight, plant planting status, plant survival rate, plant health status and plant planting area.

[0188] It should be noted that the aforementioned explanation of the embodiment of the fish-vegetable symbiotic energy Internet control method is also applicable to the fish-vegetable symbiotic energy Internet control device of this embodiment, and will not be repeated here.

[0189] According to the energy internet control device for fish-vegetable symbiosis of the embodiment of the present application, based on a preset deep learning time series prediction strategy, the device predicts the water quality changes and cultivation environment changes within a preset time period according to the current water quality parameters, the current cultivation environment parameters, the current fish and vegetable health status and the current near-ground meteorological condition parameters, and determines the fishery farming warning level and the plant planting warning level. In combination with the current status of multiple aquaculture controllable power load modules, the target control parameters of each aquaculture controllable power load module are determined, and the corresponding aquaculture controllable power load module is controlled. In this way, the problem that the background technology is difficult to cope with complex climate and sudden risks, resulting in low aquaculture production efficiency and a surge in electricity consumption is solved. By integrating Internet of Things sensors, intelligent algorithms and automated control, real-time dynamic regulation, fault warning, resilience improvement and cost reduction are achieved.

[0190] Figure 23 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0191] Memory 2301 , processor 2302 , and computer programs stored in the memory 2301 and executable on the processor 2302 .

[0192] When the processor 2302 executes the program, the energy Internet control method for aquaponics provided in the above embodiment is implemented.

[0193] Furthermore, the electronic device further includes:

[0194] The communication interface 2303 is used for communication between the memory 2301 and the processor 2302 .

[0195] The memory 2301 is used to store computer programs that can be run on the processor 2302.

[0196] The memory 2301 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0197] If the memory 2301, processor 2302, and communication interface 2303 are implemented independently, the communication interface 2303, memory 2301, and processor 2302 can be interconnected via a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 23 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0198] Optionally, in a specific implementation, if the memory 2301, the processor 2302 and the communication interface 2303 are integrated on a chip, the memory 2301, the processor 2302 and the communication interface 2303 can communicate with each other through an internal interface.

[0199] The processor 2302 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0200] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned aquaponics energy Internet control method.

[0201] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0202] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0203] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for controlling an energy internet for aquaponics, characterized in that: The following steps are involved: Obtain current water quality parameters, current cultivation environment parameters, current fish and vegetable health status, current near-ground meteorological condition parameters, and the current status of multiple aquaculture controllable power load modules; Based on a preset deep learning time series prediction strategy, the water quality changes and cultivation environment changes within a preset time period are predicted according to the current water quality parameters, the current cultivation environment parameters, the current fish and vegetable health conditions, and the current near-surface meteorological condition parameters; Determine the warning level for fishery farming based on the water quality changes, and determine the warning level for plant cultivation based on the cultivation environment changes; The target control parameters of each aquaculture controllable power load module are determined according to the current status of the multiple aquaculture controllable power load modules, the fishery aquaculture warning level, and the plant planting warning level, and the corresponding aquaculture controllable power load module is controlled according to the target control parameters of each aquaculture controllable power load module.

2. The method according to claim 1, characterized in that The determining of the target control parameters of each aquaculture controllable power load module according to the current status of the plurality of aquaculture controllable power load modules, the fishery aquaculture warning level, and the plant planting warning level includes: Obtaining the current node marginal electricity price and current line congestion information, and based on a preset dynamic coordination optimization strategy, inputting the current node marginal electricity price and the current line congestion information into a pre-established power flow optimization sub-model to obtain a power flow optimization result; Based on the grid flow optimization results, the current states of the multiple aquaculture controllable power load modules, the fishery aquaculture warning level and the plant planting warning level are respectively input into a pre-built fishery deep reinforcement learning model and a pre-built agricultural deep reinforcement learning model to obtain the target control parameters of each aquaculture controllable power load module.

3. The method according to claim 1, characterized in that The current water quality parameters include at least one of water temperature, dissolved oxygen content, pH, ammonia nitrogen content, nitrite concentration, water flow rate and liquid level.

4. The method according to claim 3, characterized in that Determining the fishery aquaculture warning level according to the water quality change includes: Determining the interval of the water temperature, the interval of the dissolved oxygen content, the interval of the pH, the interval of the ammonia nitrogen content, the interval of the nitrite concentration, the interval of the water flow rate, and the interval of the liquid level; The fishery breeding warning level is determined according to the water quality temperature range, the dissolved oxygen content range, the pH range, the ammonia nitrogen content range, the nitrite concentration range, the water flow rate range and the liquid level range, wherein the fishery breeding warning level includes the first to fourth levels, the warning level of the first level is less than the warning level of the second level, the warning level of the second level is less than the warning level of the third level, and the warning level of the third level is less than the warning level of the fourth level.

5. The method according to claim 1, wherein The current cultivation environment parameters include at least one of air temperature, air humidity, light intensity and carbon dioxide concentration.

6. The method according to claim 5, characterized in that Determining the plant planting warning level according to the changes in the cultivation environment includes: Determining a temperature interval for the air temperature, a humidity interval for the air humidity, a light interval for the light intensity, and a concentration interval for the carbon dioxide concentration; The plant planting warning level is obtained according to the temperature range, the humidity range, the light range and the concentration range, wherein the plant planting warning level includes levels five to eight, the warning level of the fifth level is less than the warning level of the sixth level, the warning level of the sixth level is less than the warning level of the seventh level, and the warning level of the seventh level is less than the warning level of the eighth level.

7. The method according to claim 5, characterized in that The current fish and vegetable health status includes at least one of the fish farming status, fish survival rate, fish health status, farmed fish weight, plant planting status, plant survival rate, plant health status and plant planting area.

8. An energy internet control device for fish and vegetable symbiosis, characterized in that: include: An acquisition module is used to obtain current water quality parameters, current cultivation environment parameters, current fish and vegetable health status, current near-ground meteorological condition parameters, and the current status of multiple aquaculture controllable power load modules; A prediction module is used to predict changes in water quality and cultivation environment within a preset time period based on the current water quality parameters, the current cultivation environment parameters, the current fish and vegetable health conditions, and the current near-surface meteorological condition parameters based on a preset deep learning time series prediction strategy; a determination module, configured to determine a warning level for fishery farming based on the water quality changes, and to determine a warning level for plant cultivation based on the cultivation environment changes; A control module is used to determine the target control parameters of each aquaculture controllable power load module according to the current status of the multiple aquaculture controllable power load modules, the fishery aquaculture warning level, and the plant planting warning level, and to control the corresponding aquaculture controllable power load module according to the target control parameters of each aquaculture controllable power load module.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the energy internet control method for aquaponics according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the energy internet control method for aquaponics as described in any one of claims 1 to 7.

Images