Linkage intelligent control system and method of earthworm-plant circulating culture system

Abstract

The invention provides a linkage intelligent control system and method for an earthworm-plant circulating culture system, and relates to the technical field of earthworm culture, and the system comprises a plant planting module which is arranged on a plant planting layer and comprises at least one planting area and a soil fertility sensing unit used for monitoring the soil nutrient condition in each planting area, the rotten vegetable metering unit is used for counting the rotten vegetable yield; the earthworm breeding module is arranged on the earthworm breeding layer and comprises at least one breeding bed, a biomass sensing unit used for monitoring the biomass of earthworms in each breeding bed and an earthworm cast bin used for storing and outputting earthworm cast; and the linkage intelligent control module is in communication connection with the plant planting module and the earthworm breeding module and is used for carrying out linkage decision making based on the multi-source data from the plant planting module and the earthworm breeding module and controlling the material processing and conveying equipment to execute corresponding operation so as to realize resource cyclic utilization of breeding wastes. The value of the waste vegetable residues and the earthworm cast fertilizer is greatly utilized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of earthworm breeding, and particularly relates to a linkage intelligent control system and method of an earthworm-plant circular breeding system. BACKGROUND

[0002] The existing circular agriculture mode combining earthworm breeding and plant cultivation usually adopts a simple design of physical space superposition, for example, vegetables are planted in the upper layer and earthworms are bred in the lower layer. However, this mode still has significant defects in actual operation. First, the transfer of material flow (tail vegetables, earthworm manure) in the system depends on artificial experience judgment and timing operation, such as fixed period feeding of tail vegetables or application of earthworm manure, and lacks dynamic perception and response to the actual demand of plants, the real-time state of earthworms and the inventory of materials, resulting in unreasonable allocation of resources and low conversion efficiency. Second, the information of each link (cultivation, breeding) is isolated, the cultivation end does not know the conversion capacity of earthworms and the fertilizer output rhythm, and the breeding end does not master the supply law and component change of tail vegetables, and therefore cannot form efficient and self-adaptive cooperation. In addition, the monitoring of key parameters (such as soil fertility and earthworm biomass) is often missing or relies on artificial sampling, and the data is lagged and discontinuous, which is difficult to support precise management decisions. Therefore, the existing technology cannot realize the intelligent, closed-loop and efficient recycling of waste resources in the true sense. SUMMARY

[0003] In view of the above defects or deficiencies in the prior art, the present application aims to provide a linkage intelligent control system and method of an earthworm-plant circular breeding system.

[0004] In a first aspect, the present application provides a linkage intelligent control system of an earthworm-plant circular breeding system, which comprises a system framework, the system framework comprising a plant cultivation layer and an earthworm breeding layer arranged in a top-bottom manner, and a material processing and conveying device arranged therebetween; and comprising: a plant cultivation module arranged in the plant cultivation layer, comprising at least one cultivation area, a soil fertility sensing unit for monitoring the soil nutrient status in each cultivation area, and a tail vegetable metering unit for counting the tail vegetable yield; an earthworm breeding module arranged in the earthworm breeding layer, comprising at least one breeding bed, a biomass sensing unit for monitoring the earthworm biomass in each breeding bed, and an earthworm manure storage and output device; a linkage intelligent control module in communication connection with the plant cultivation module and the earthworm breeding module, configured to make linkage decisions based on multi-source data from the plant cultivation module and the earthworm breeding module, and control the material processing and conveying device to perform corresponding operations, so as to realize the recycling of breeding waste.

[0005] According to the technical scheme provided in the application, the linkage intelligent control module comprises: a central processing unit; a feeding decision unit connected with the central processing unit, configured to receive and process data from the tail meal metering unit and each biomass sensing unit, to generate and issue a first control instruction to a waste treatment and feeding device in the material treatment and conveying equipment; a fertilization decision unit connected with the central processing unit, configured to receive and process data from each soil fertility sensing unit and the worm manure bin, to generate and issue a second control instruction to a fertilization conveying device in the material treatment and conveying equipment; a cleaning and screening scheduling unit connected with the central processing unit, configured to generate and issue a third control instruction to a cleaning and screening device in the material treatment and conveying equipment according to a preset period or a received external trigger signal.

[0006] According to the technical scheme provided in the application, the feeding decision unit is specifically configured to: obtain a total amount of tail meal inventory reported by the tail meal metering unit and real-time biomass data of each breeding bed reported by the corresponding biomass sensing unit; when the total amount of tail meal inventory reaches a preset feeding trigger threshold, calculate a feeding demand priority coefficient of each breeding bed based on the real-time biomass data of the breeding bed, wherein the higher the real-time biomass data of the breeding bed, the higher the feeding demand priority coefficient of the breeding bed; determine at least one target breeding bed in a current feeding period from all breeding beds according to the feeding demand priority coefficient of each breeding bed; calculate an accurate feeding amount allocated to each target breeding bed based on the total amount of tail meal inventory and the real-time biomass data of each target breeding bed; generate the first control instruction comprising an identifier of each target breeding bed and the corresponding accurate feeding amount.

[0007] According to the technical scheme provided in the application, the fertilization decision unit is specifically configured to: obtain at least one key nutrient content data reported by the soil fertility sensing unit corresponding to each planting area and real-time inventory data of the worm manure bin; when the real-time inventory data is higher than a preset safe inventory threshold, perform the following loop decision for each planting area: compare the key nutrient content data of the planting area with a preset optimal interval of the nutrient; if the key nutrient content data is lower than the lower limit of the preset optimal content interval, marking the planting area as a candidate fertilization area in the current fertilization cycle, and calculating a fertilization urgency level of the candidate fertilization area based on a soil nutrient deficiency amount, the soil nutrient deficiency amount being a difference between the key nutrient content data and the lower limit of the preset optimal content interval; summarizing the fertilization urgency levels of all candidate fertilization areas, and determining at least one target fertilization area to be executed this time according to the levels from high to low; calculating a recommended fertilization amount for each target fertilization area this time based on the soil nutrient deficiency amount of the target fertilization area, real-time inventory data of the worm manure warehouse, and a preset single fertilization upper limit amount; generating the second control instruction including the identification of each target fertilization area and the corresponding recommended fertilization amount.

[0008] According to the technical scheme provided in the present application, the cleaning and screening scheduling unit is specifically configured to: real-time acquisition of historical feeding records and corresponding biomass growth curves of each breeding bed; based on the historical feeding records and biomass growth curves, constructing and training a biomass growth-feeding response model for each breeding bed, and using the model to predict the biomass growth inflection point of the breeding bed; the biomass growth inflection point is defined as the theoretical time point at which the biomass growth rate begins to slow down continuously; for any breeding bed, when the system time enters the monitoring window period after the predicted biomass growth inflection point, starting high-frequency evaluation of its screening demand; when the same breeding bed is marked as efficiency decline and growth stagnation at the same time in the high-frequency evaluation, and the cumulative running time since the last screening operation has reached a preset minimum operation interval, generating the third control instruction.

[0009] According to the technical scheme provided in the present application, the high-frequency evaluation includes: efficiency decline evaluation: calculating the worm manure output efficiency of the breeding bed in the monitoring window period, the worm manure output efficiency being the ratio of the cumulative worm manure output amount to the cumulative feeding amount in the window period; if the current output efficiency decreases by more than a first preset threshold compared with the benchmark period before the inflection point, it is marked as efficiency decline; growth stagnation evaluation: obtaining real-time biomass data of the breeding bed in the monitoring window period, if the biomass growth value is lower than a preset extremely low growth threshold, it is marked as growth stagnation.

[0010] According to the technical scheme provided in the present application, the central processing unit is further configured to: monitoring and calculating the inventory change rate of the worm manure bin as the actual output rate of worm manure; at the same time, based on the historical recommended fertilization amount in the second control instruction of the fertilization decision unit, calculating the average fertilization consumption rate; calculating the difference between the actual output rate of worm manure and the average fertilization consumption rate as the worm manure supply and demand balance deviation; when the worm manure supply and demand balance deviation continuously exceeds the preset reasonable range, generating a parameter adjustment instruction to the feeding decision unit; wherein, the parameter adjustment instruction is used to dynamically adjust the preset feeding trigger threshold and / or the calculation weight of the accurate feeding amount in the feeding decision unit, so as to indirectly adjust the worm manure output rate by adjusting the feeding amount, so that the actual output rate of worm manure approaches the average fertilization consumption rate.

[0011] According to the technical scheme provided in the present application, the feeding decision unit is further configured to: Before calculating the feeding demand priority coefficient based on the real-time biomass data of each breeding bed, the worm manure output efficiency is obtained; Based on the real-time biomass data and the worm manure output efficiency of the corresponding breeding bed, an efficiency corrected biomass value is obtained; Based on the efficiency corrected biomass value, the feeding demand priority coefficient of each breeding bed is calculated, wherein the higher the efficiency corrected biomass value of a breeding bed, the higher the feeding demand priority coefficient of the breeding bed.

[0012] According to the technical scheme provided in the present application, the feeding decision unit is further configured to: When calculating the accurate feeding amount allocated to the target breeding bed, it is inquired whether the target breeding bed has entered the monitoring window period after the predicted biomass growth inflection point; If not, it is determined that the breeding bed is in a rapid growth period dominated by juvenile worms, and a first feeding amount calculation model is used, which is based on the efficiency corrected biomass value or real-time biomass data and gives a higher unit biomass feeding coefficient; If yes, it is determined that the breeding bed has entered a stable transformation period dominated by adult worms, and a second feeding amount calculation model is used, which is based on the efficiency corrected biomass value or real-time biomass data and gives a lower unit biomass feeding coefficient; Wherein, the single feeding amount calculated by the first feeding amount calculation model is higher than the value calculated by the second feeding amount calculation model under the same efficiency corrected biomass value or real-time biomass data.

[0013] In a second aspect, the present application provides a linkage intelligent control method for earthworm-plant recycling breeding system, which is realized based on the linkage intelligent control system as described above, comprising the following steps: Collect tail-cabbage yield data and soil fertility data, and collect earthworm biomass data and earthworm manure inventory data at the same time; Based on the tail-cabbage yield data, soil fertility data, earthworm biomass data and earthworm manure inventory data, a combined control strategy for coordinating the control of the material processing and conveying equipment is generated; According to the combined control strategy, the material processing and conveying equipment is driven to perform corresponding material processing and conveying operations, so that the tail-cabbage produced by the plant planting layer is converted into bait for the earthworm breeding layer, and the earthworm manure produced by the earthworm breeding layer is conveyed to the plant planting layer as fertilizer, realizing the recycling and utilization of breeding waste in the system.

[0014] Compared with the prior art, the beneficial effects of the present application are: I. Intelligent linkage and precise control of closed-loop substances are realized: By setting the upper and lower system framework and configuring the linkage intelligent control module in communication connection with the plant planting module and the earthworm breeding module respectively, the system can obtain multi-source data such as tail-cabbage yield, soil fertility, earthworm biomass and earthworm manure inventory in real time. Based on these data, the linkage intelligent control module makes comprehensive analysis and makes unified decision, directly controls the material processing and conveying equipment to perform corresponding operations. This changes the isolated status of each link in the traditional mode, integrates the material circulation process of tail-cabbage production→earthworm conversion→earthworm manure output→plant fertilization into an intelligent closed loop under control, greatly improving the automation degree and overall efficiency of resource circulation.

[0015] II. The precision and timeliness of resource utilization are improved: the plant planting module includes soil fertility sensing unit and tail-cabbage metering unit, and the earthworm breeding module includes biomass sensing unit. These sensing units provide accurate and real-time source data. The linkage intelligent control module makes decisions based on this, can ensure that the triggering time and execution amount of feeding, fertilizing and other operations are more scientific and accurate, avoid the waste of resources (such as over-fertilization) or insufficient supply (such as earthworm bait shortage) caused by traditional manual or timing operation, and make the value of tail-cabbage waste and earthworm manure be greatly utilized. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The structure diagram of the linkage intelligent control system of the earthworm-plant circulation breeding system provided by the present application.

[0017] The text annotations in the figure represent: 1, plant planting module; 2, linkage intelligent control module; 3, material processing and conveying equipment; 4, earthworm breeding module. DETAILED DESCRIPTION

[0018] The application will be described in further detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related application, and not to limit the application. In addition, it should be noted that only the parts related to the application are shown in the drawings for ease of description.

[0019] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the drawings and embodiments.

[0020] Embodiment 1 As mentioned in the background, in order to solve the problems in the prior art, the present application proposes a linkage intelligent control system of an earthworm-plant circular breeding system, which comprises a system framework, the system framework comprises a plant planting layer and an earthworm breeding layer arranged above and below, and a material processing and conveying device 3 arranged therebetween; as shown in Figure 1 , comprising: a plant planting module 1 arranged in the plant planting layer, comprising at least one planting area, a soil fertility sensing unit for monitoring the soil nutrient status in each planting area, and a tail vegetable metering unit for counting the tail vegetable yield; an earthworm breeding module 4 arranged in the earthworm breeding layer, comprising at least one breeding bed, a biomass sensing unit for monitoring the earthworm biomass in each breeding bed, and an earthworm manure storage and output device; a linkage intelligent control module 2, which is in communication connection with the plant planting module 1 and the earthworm breeding module 4, is configured to make linkage decisions based on multi-source data from the plant planting module 1 and the earthworm breeding module 4, and control the material processing and conveying device 3 to perform corresponding operations, so as to realize the recycling utilization of breeding waste.

[0021] Specifically, the system architecture comprises a plant planting layer and an earthworm breeding layer arranged above and below. The plant planting layer generally refers to a physical space for hydroponic, substrate or soil cultivation of plants, such as a planting rack, a planting tank or a planting bed. The earthworm breeding layer refers to a separate space below the plant planting layer for accommodating earthworm breeding beds, and the two are structurally separated but associated through the material channel designed by the system. The material processing and conveying device 3 arranged therebetween is a general concept, which covers all mechanical and electrical devices for processing and transferring materials between the two layers, such as a crusher and a screw feeder for crushing and conveying tail vegetables, a pneumatic conveying pipeline or a belt conveyor for conveying earthworm manure, and a vibrating screen and a cleaning spray device for screening earthworms.

[0022] The system contains three core modules. The plant cultivation module 1 is located in the upper plant cultivation layer, which contains at least one logically or physically independent cultivation area, such as different hydroponic tanks or soil partitions. The module integrates soil fertility sensing units and tail vegetable metering units. The soil fertility sensing unit is a combination of sensors for in-situ or online monitoring of the content of one or more key nutrients (such as nitrogen, phosphorus, potassium), pH value or conductivity in the planting medium. The tail vegetable metering unit is used to automatically count the amount of waste tail vegetables produced during the plant cultivation process, which can be implemented by a weighing sensor, a volume measuring device or an image recognition combined volume estimation system installed at the tail vegetable collection port.

[0023] The worm breeding module 4 is located in the lower worm breeding layer, which contains at least one independent breeding bed, i.e. a container or area for breeding earthworms. The module integrates biomass sensing units and vermicompost warehouses. The biomass sensing unit is used for non-invasive or indirect monitoring of the total biomass of earthworms in each breeding bed, which can be implemented by periodically weighing the weight change of the entire breeding bed and deducting the base weight, or indirectly evaluating the biomass content by using a specific frequency dielectric sensor. The vermicompost warehouse is a container for temporarily storing the vermicompost produced by the conversion of earthworms, usually equipped with a level sensor to monitor inventory, and a controllable discharge port.

[0024] The linkage intelligent control module 2 can be an industrial computer, PLC (Programmable Logic Controller) or embedded controller. The module establishes data connection with the plant cultivation module 1 and the earthworm breeding module 4 through wired (such as RS485, CAN bus) or wireless (such as LoRa, Wi-Fi) communication. Its implementation is as follows: the module continuously receives tail vegetable yield data and soil fertility data from the plant cultivation module 1, and receives earthworm biomass data and vermicompost inventory data from the earthworm breeding module 4. Based on the fusion analysis of these multi-source and heterogeneous data, the module runs the built-in decision algorithm to generate one or more control instructions. These instructions are sent to the material handling and conveying equipment 3 to drive it to perform precise operations. For example, when the decision considers that feeding is needed, the waste handling equipment is controlled to crush the tail vegetables and transport them to the designated breeding bed; when the decision considers that fertilization is needed, the conveying equipment is controlled to transport a certain amount of vermicompost from the vermicompost warehouse to the designated cultivation area.

[0025] This implementation addresses the core problems of traditional earthworm-plant symbiosis systems, such as extensive material flow management, reliance on manual experience, and disconnected processes. It constructs a data-driven, closed-loop intelligent control system for material circulation. The technical principle lies in clarifying the material circulation path through a hierarchical physical architecture design (upper and lower layers), and providing a data foundation for intelligent decision-making through real-time quantitative sensing of key parameters at both the planting end (waste vegetables, soil) and the breeding end (earthworms, vermicompost). The linked intelligent control module 2, acting as the central hub, performs linked decision-making based on multi-source data. This means it doesn't process signals from each link independently, but rather considers the waste vegetable supply capacity, plant nutrient requirements, earthworm carrying capacity, and fertilizer inventory as a holistic variable. For example, it might decide whether to feed and the amount based on waste vegetable yield and earthworm biomass, and whether to fertilize and the amount based on soil fertility and vermicompost inventory.

[0026] Secondly, this solution automates and refines the resource recycling process for livestock waste. The underlying technology utilizes sensors to replace manual observation and measurement, enabling continuous and objective acquisition of key status information and overcoming the lag and subjectivity of manual judgment. The linked intelligent control module 2 automatically determines the timing and intensity of operations based on a preset decision-making model or algorithm, and accurately completes the operation by controlling the execution equipment, thus seamlessly connecting the entire chain of waste production, conversion, and fertilizer application. This not only significantly reduces labor costs, but more importantly, through precise control, it avoids problems that may occur in traditional methods, such as overfeeding (leading to environmental degradation), insufficient or excessive fertilization (affecting plant growth or causing waste), significantly improving the conversion efficiency and utilization value of resources (vegetable waste, vermicompost).

[0027] In a preferred embodiment, the linkage intelligent control module 2 includes: Central processing unit; The feeding decision unit is connected to the central processing unit and is configured to receive and process data from the vegetable waste metering unit and each of the biomass sensing units to generate and send a first control command to the waste treatment and feeding device in the material handling and conveying equipment 3. The fertilization decision unit is connected to the central processing unit and is configured to receive and process data from each of the soil fertility sensing units and the vermicompost bins to generate and send a second control command to the fertilization conveying device in the material handling and conveying equipment 3. The cleaning and screening scheduling unit is connected to the central processing unit and is configured to generate and send a third control command to the cleaning and screening device in the material handling and conveying equipment 3 according to a preset cycle or a received external trigger signal.

[0028] Specifically, the feeding decision unit, the fertilization decision unit and the washing and screening scheduling unit are three specialized functional sub-modules. They can be independent software threads, tasks or functional modules running on the central processing unit, or hardware functional units assisted by dedicated processing chips. These units are connected with the central processing unit, meaning that there is a two-way data and instruction exchange channel between them. The feeding decision unit is configured to receive and process data from the tail meal metering unit and the biomass sensing units. In implementation, the unit regularly or event-triggered acquires the latest total tail meal inventory and real-time biomass data of each cultivation bed from the central processing unit. It contains decision logic for determining whether the feeding condition is met and calculating the feeding details. Finally, it generates a structured first control instruction, which is sent to the waste treatment and feeding device in the material handling and conveying equipment 3 through the central processing unit or directly. The device usually includes a crusher and a precision feeder, and the instruction may contain information such as target cultivation bed number, feeding material type (such as tail meal type), feeding weight, etc.

[0029] The working mode of the fertilization decision unit is similar. It receives soil nutrient data from the soil fertility sensing units and real-time inventory data from the worm manure warehouse. Its decision logic is based on these data to determine whether fertilization is needed, where to fertilize and how much to fertilize. The generated second control instruction is sent to the fertilization conveying device, which may include a metering feeder under the worm manure warehouse, a conveying pump and a pipeline valve network, and the instruction specifies the target planting area and the amount of fertilizer.

[0030] The washing and screening scheduling unit is responsible for managing the separation of earthworms and worm manure. Its trigger conditions include a preset period (such as once every 30 days) or a received external trigger signal (such as a forced screening command from the administrator or a cooperative request from other modules). The third control instruction generated by it is sent to the washing and screening device, which usually includes a vibrating screen that can pour the material in the cultivation bed, a water spraying system and a collection tank for earthworms and worm manure, and the instruction may contain target cultivation bed identification and screening intensity parameters.

[0031] In a preferred embodiment, the feeding decision unit is specifically configured to: obtain the total tail meal inventory reported by the tail meal metering unit and the real-time biomass data reported by the biomass sensing unit corresponding to each cultivation bed in real time; when the total tail meal inventory reaches a preset feeding trigger threshold, calculate the feeding demand priority coefficient of each cultivation bed based on the real-time biomass data of each cultivation bed, wherein the higher the real-time biomass data of a cultivation bed, the higher the feeding demand priority coefficient of the cultivation bed; determine at least one target cultivation bed in the current feeding period from all cultivation beds according to the feeding demand priority coefficient of each cultivation bed; based on the total inventory of the stale food and real-time biomass data of each of the target cultivation beds, calculate the precise feeding amount allocated to each of the target cultivation beds; generate the first control instruction including the identification of each target cultivation bed and its corresponding precise feeding amount.

[0032] Specifically, first of all, this unit needs to obtain the total inventory of stale food reported by the stale food metering unit in real time. Real-time here means obtaining the latest cumulative stale food weight data at a set sampling frequency. At the same time, it needs to obtain the real-time biomass data reported by the biomass sensor unit corresponding to each cultivation bed, i.e. the current estimated weight of earthworms in each independent cultivation bed.

[0033] Secondly, a preset feeding trigger threshold is provided in this unit. This threshold is a key control parameter, representing the minimum stale food inventory at which the system considers that a feeding operation can be started. In practice, the unit continuously compares the obtained total inventory of stale food with this threshold.

[0034] When the condition is met, i.e. the total inventory of stale food reaches or exceeds the preset feeding trigger threshold, the feeding decision unit starts the core decision-making process. The first step is to calculate the feeding demand priority coefficient of each cultivation bed based on its real-time biomass data. Its calculation rule is clearly defined as follows: the higher the real-time biomass data of a cultivation bed, the higher its feeding demand priority coefficient. This is a simple and effective weight distribution strategy, which implies the biological principle that the larger the biomass of earthworm population, the higher the overall metabolic demand for maintaining life activities and growth, and therefore it should be given priority to obtain feed supply. In software implementation, the priority coefficient can be obtained by dividing the biomass of each bed by the total biomass of all beds, or by directly normalizing the biomass to obtain the priority coefficient.

[0035] Next, according to the calculated feeding demand priority coefficients of each cultivation bed, the system determines at least one target cultivation bed in the current feeding cycle from all cultivation beds. The determination method can be to select the top N beds with the highest priority coefficients, or to select all beds with priority coefficients exceeding a certain reference value.

[0036] Then, the system needs to calculate the precise feeding amount allocated to each target cultivation bed. The calculation is based on two core inputs: the total available stale food inventory and the real-time biomass data of each target cultivation bed. A typical implementation algorithm is as follows: first, according to the proportion of the biomass of each target bed in the total target bed biomass, the feeding amount is preliminarily allocated; secondly, considering constraints such as not exceeding the processing capacity of the cultivation bed in a single feeding, fine-tuning is performed; finally, ensuring that the total feeding amount does not exceed the current total inventory of stale food.

[0037] Finally, the feeding decision unit generates the first control command. This command must include the identifier (such as a number or location code) of each target breeding bed and its corresponding precise feeding amount, calculated as described above. The command is then formatted and sent to the executing mechanism.

[0038] Furthermore, a method for calculating the precise feeding amount for the target breeding bed is given, assuming the system has a total of n breeding beds.

[0039] Obtain real-time biomass data for each bed, denoted as B_i (i = 1, 2, ..., n).

[0040] Calculate the total real-time biomass of all culture beds: B_total = B_1 + B_2 + ... + B_n.

[0041] Calculate the feeding priority coefficient P_i for each breeding bed i: P_i = B_i / B_total.

[0042] Bed selection: Set a priority threshold P_threshold (e.g., 0.1). Select all beds with P_i ≥ P_threshold as target beds for the current feeding cycle. Assume there are m target beds (m ≤ n).

[0043] Precise feeding amount calculation: Get the current total inventory of leftover vegetables, denoted as S_total.

[0044] Calculate the sum of real-time biomass for all target culture beds: B_target_sum = B_t1 + B_t2 + ... + B_tm (where t1, t2, ..., tm are the indexes of the target beds).

[0045] For each target culture bed tj, calculate its precise feeding amount F_tj: F_tj = S_total ×(B_tj / B_target_sum) ×α. In this formula, α is a preset global feeding coefficient (0 < α ≤ 1), used to ensure that the total feeding amount in a single batch does not exceed S_total or meets other system constraints. It is directly multiplied by the initial feeding amount allocated according to the biomass ratio, and its value directly affects (proportionally amplifies or reduces) the final total feeding amount for all target beds. Therefore, α functionally controls the weighting parameter for the overall scale of feeding amount calculation. B_tj / B_target_sum represents the biomass weight of the target bed within the target population.

[0046] In a preferred embodiment, the fertilization decision unit is specifically configured to: acquire, in real time, at least one key nutrient content data reported by a soil fertility sensing unit corresponding to each planting area, and real-time inventory data of the worm manure warehouse; When the real-time inventory data is higher than a preset safe inventory threshold, for each planting area, the following loop decision is executed: Compare the key nutrient content data of the planting area with a preset optimal content interval of the nutrient; If the key nutrient content data is lower than the lower limit of the preset optimal content interval, mark the planting area as a candidate fertilization area in the current fertilization cycle, and calculate the fertilization urgency level of the candidate fertilization area based on soil nutrient deficiency amount; the soil nutrient deficiency amount is the difference between the key nutrient content data and the lower limit of the preset optimal content interval; Summarize the fertilization urgency levels of all candidate fertilization areas, and determine at least one target fertilization area to be executed according to the level; Based on the soil nutrient deficiency amount of the target fertilization area, the real-time inventory data of the worm manure warehouse, and a preset single fertilization upper limit, calculate the recommended fertilization amount of each target fertilization area; Generate the second control instruction including the identification of each target fertilization area and the corresponding recommended fertilization amount.

[0047] Specifically, first, the unit needs to acquire two aspects of data in real time: one is at least one key nutrient content data reported by a soil fertility sensing unit corresponding to each planting area, such as nitrogen content (in mg / kg); the other is the real-time inventory data of the worm manure warehouse, usually expressed in volume or weight.

[0048] Secondly, a preset safe inventory threshold is provided in the unit. This threshold represents the minimum inventory amount that the system considers necessary for the worm manure warehouse to start the fertilization operation, which is used to ensure the continuity of the fertilization operation and to deal with uncertainties. The starting condition of the decision-making process is that the real-time inventory data is higher than the preset safe inventory threshold, that is, the inventory is sufficient, which is the premise of fertilization.

[0049] When the inventory condition is satisfied, the system executes a loop decision sub-process for each planting area: first, compare the reported key nutrient content data of the planting area with the pre-set optimal content interval of the nutrient. This optimal interval is a desired nutrient concentration range pre-set based on the growth characteristics of the planted crops, for example, nitrogen content is set to 150-200 mg / kg. Second, make a judgment: if the key nutrient content data of the planting area is lower than the lower limit of the pre-set optimal content interval (for example, nitrogen content is lower than 150 mg / kg), mark the planting area as a candidate fertilization area in this round of fertilization cycle. Third, calculate the soil nutrient deficiency of the candidate fertilization area. The deficiency is specifically defined as the difference between the key nutrient content data and the lower limit of the pre-set optimal content interval. For example, if the measured nitrogen content is 120 mg / kg and the lower limit is 150 mg / kg, the deficiency is 30 mg / kg. Fourth, based on the calculated soil nutrient deficiency of the candidate fertilization area, calculate the fertilization urgency level of the candidate fertilization area. The calculation rule can be linear (the larger the deficiency, the higher the level) or segmented (set several deficiency thresholds corresponding to different levels). After traversing all planting areas and completing the above loop decision, the system summarizes the fertilization urgency levels of all marked candidate fertilization areas. Then, according to the level, determine at least one target fertilization area for this execution. For example, select the top few areas with the highest urgency level, or all areas with an urgency level exceeding an emergency threshold. Next, the system calculates the recommended fertilization amount for each target fertilization area. The calculation is based on three core parameters: the soil nutrient deficiency of the target fertilization area (representing the demand intensity), the real-time inventory data of the worm manure warehouse (representing the supply upper limit), and the pre-set single fertilization upper limit (representing the capacity limit of a single operation or the safety upper limit set to avoid seedling burning). One implementation algorithm is to first allocate a theoretical fertilization amount according to the deficiency proportion of each target area, and then check if the total theoretical amount exceeds the constraints of real-time inventory and single upper limit. If it exceeds, reduce it in proportion. Finally, the fertilization decision unit generates a second control instruction. The instruction must include the identification of each target fertilization area and the corresponding recommended fertilization amount calculated above.

[0050] In a preferred embodiment, the cleaning and screening scheduling unit is specifically configured to: real-time acquisition of historical feeding records and corresponding biomass growth curves of each breeding bed; based on the historical feeding records and biomass growth curves, construct and train a biomass growth-feeding response model for each breeding bed, and use the model to predict the biomass growth inflection point of the breeding bed; the biomass growth inflection point is defined as the theoretical time point when the biomass growth rate begins to slow down continuously; for any breeding bed, when the system time enters the monitoring window period after the predicted biomass growth inflection point, start high-frequency evaluation of its screening demand; When the same cultivation bed is marked as both efficiency decline and growth stagnation in the high-frequency evaluation, and the cumulative running time since the last screening operation has reached the preset minimum operation interval, the third control instruction is generated.

[0051] Specifically, this unit needs to periodically (e.g., daily) obtain the historical feeding records of each cultivation bed from the system database. Each record should at least include the feeding time, the target cultivation bed identifier, the type and weight of the fed vermicompost. At the same time, the corresponding biomass growth curve needs to be obtained. This curve can be drawn by recording the biomass sensor data of each cultivation bed daily or weekly (e.g., the live weight of earthworms obtained by subtracting the dry weight of the substrate from the total weight of the bed obtained by the bed body weighing sensor) to form a time series of biomass change curve. For each cultivation bed, the system needs to build a biomass growth-feeding response model. This model aims to establish a mathematical relationship between the feeding amount (input) and the biomass growth (output). A simplified implementation is to use a linear regression model or a growth curve model (such as the Logistic model). The training process of the model is as follows: taking a period of historical data (e.g., the past 60 days) as a sample, taking the cumulative feeding amount as the independent variable, and taking the net growth of biomass as the dependent variable, the model parameters are fitted by using least squares method or other algorithms. More complex implementations can consider introducing time delay effect, weight of different feed nutrients, etc. Inflection point prediction: using the trained biomass growth-feeding response model, the system can predict the biomass growth inflection point of the cultivation bed. Here, the biomass growth inflection point is defined as the theoretical time point when the biomass growth rate (i.e., the amount of biomass increase per unit time) starts to slow down continuously. A specific implementation method is: the model itself may be an S-shaped curve, and the inflection point corresponds to the point where the second derivative of the curve is zero, i.e., the turning point where the growth rate changes from acceleration to deceleration. The system can calculate the theoretical cumulative feeding amount or theoretical running time required to reach the inflection point according to the current model parameters, and map it to a future time point. The system maintains the predicted inflection point time for each cultivation bed. When the current time of the system enters the monitoring window period (e.g., 3 days to 15 days after the inflection point) of a certain cultivation bed, the high-frequency evaluation for this bed is started. This means that the system will check the specific performance indicators of this bed at a higher frequency (e.g., once a day, instead of once a week before).

[0052] Further, the high-frequency evaluation includes: Efficiency decline evaluation: calculate the vermicompost output efficiency of the cultivation bed in the monitoring window period, which is the ratio of the cumulative vermicompost output to the cumulative feeding amount in the window period; if the current output efficiency decreases by more than the first preset threshold compared with the baseline period before the inflection point, it is marked as efficiency decline; Growth stagnation assessment: Obtain the real-time biomass data of the cultivation bed in the monitoring window period. If the biomass growth value is lower than a preset extremely low growth threshold, it is marked as growth stagnation.

[0053] Specifically, efficiency decline assessment: In the monitoring window period initiated for the target cultivation bed, the system needs to count two data: one is the cumulative feeding amount of the bed in the window period (which can be directly summed up from the feeding record); The second is the cumulative vermicompost output that can be clearly attributed to the cultivation bed in the window period, which is counted from the system material flow. The attribution method can be realized by setting independent collection channels and weighing scales for the vermicompost produced by each cultivation bed, or by setting output batch labels for each bed and recording the source bed and weight when the vermicompost is stored in the warehouse. Efficiency calculation: Calculate the vermicompost output efficiency in the window period. Its definition is: efficiency = cumulative vermicompost output / cumulative feeding amount. This ratio reflects the ability of unit feed to be converted into fertilizer. The system needs to calculate the vermicompost output efficiency of a pre-kink reference period as a reference. The reference period can be a fixed period of time before the kink prediction time point (for example, 30 days before the kink). Calculate the decline of the efficiency of the current monitoring window period from the reference period efficiency: (reference efficiency - current efficiency) / reference efficiency. If the decline exceeds a first preset threshold (for example, 15% or 20%), the system will mark the cultivation bed as efficiency decline.

[0054] Specifically, growth stagnation assessment: Obtain the real-time biomass data of the cultivation bed in the monitoring window period. Usually take the biomass value at the beginning of the window period (initial value) and the biomass value at the end of the window period (end value). Calculate the biomass growth value, which is the end value minus the initial value. Compare the calculated biomass growth value with a preset extremely low growth threshold. The threshold is a very small positive number, representing a biomass growth that can be ignored in the monitoring window period, for example, 1% of the initial biomass or an absolute weight value (such as 0.5 kg). If the growth value is lower than the threshold, it is determined that the bed biomass growth is basically stagnant, and it is marked as growth stagnation.

[0055] In a preferred embodiment, the central processing unit is further configured to: monitor and calculate the inventory change rate of the vermicompost warehouse in real time as the actual vermicompost output rate; at the same time, based on the historical recommended fertilization amount in the second control instruction of the fertilization decision unit, calculate the average fertilization consumption rate; calculate the difference between the actual vermicompost output rate and the average fertilization consumption rate as the vermicompost supply and demand balance deviation; generate parameter adjustment instructions to the feeding decision unit when the vermicompost supply and demand balance deviation continuously exceeds the preset reasonable range; The parameter adjustment command is used to dynamically adjust the preset feeding trigger threshold and / or the calculation weight of the precise feeding amount in the feeding decision unit, so as to indirectly adjust the earthworm casting production rate by adjusting the feeding amount, so that the actual earthworm casting production rate approaches the average fertilizer consumption rate.

[0056] Specifically, the rate calculation is as follows: Production rate calculation: The central processing unit continuously monitors the inventory level of the vermicompost storage. For example, the inventory weight is recorded every 4 hours. By calculating the change in inventory between two adjacent records (subtracting the previous value from the later value) and dividing by the time interval, an instantaneous inventory change rate can be obtained. Averaging or moving average processing is performed on the instantaneous rates over multiple consecutive time periods (e.g., the past 7 days) to obtain a relatively stable actual vermicompost production rate. Consumption rate calculation: The central processing unit retrieves the recommended fertilization amounts contained in all issued second control commands over a past period (e.g., the past 7 days) from the historical records of the fertilization decision unit. These recommended fertilization amounts are summed over time and then divided by the corresponding time length to calculate the average fertilization consumption rate. Note that this consumption rate is based on the planned value of the decision command and represents the system's fertilization demand intensity.

[0057] Deviation Calculation and Judgment: The deviation in the supply and demand balance of vermicompost is calculated as the actual vermicompost production rate minus the average fertilizer consumption rate. The system sets a preset reasonable range, for example, from -0.5 kg / day to +0.5 kg / day. This means the system allows the production rate to be slightly higher or slightly lower than the consumption rate. The central processing unit continuously monitors this deviation. When the deviation consistently (e.g., for 3 consecutive days) exceeds the above reasonable range, the system is determined to be in an unbalanced state. A consistently positive and excessively large deviation indicates that production far exceeds consumption, and inventory will continue to accumulate; a consistently negative and excessively large deviation indicates that production cannot keep up with consumption, and inventory is at risk of depletion.

[0058] Parameter adjustment command generation and execution: Once an imbalance is determined, the central processing unit generates a parameter adjustment command and sends it to the feeding decision unit. The purpose of this command is to indirectly affect the downstream vermicompost production rate by adjusting the upstream feeding amount.

[0059] Specific adjustment method one: If the deviation is negative (insufficient output), the instruction requires lowering the preset feeding trigger threshold in the feeding decision unit. This makes it easier for the system to trigger feeding operations, increasing the consumption of leftover vegetables and the intake of earthworms, thereby hoping to improve the subsequent rate of earthworm castings production.

[0060] Specific adjustment mode two: the instruction can also require adjustment of the calculation weight of the precise feeding amount. When the central processing unit determines that the feeding amount needs to be adjusted to balance the supply and demand of vermicompost, it generates a parameter adjustment instruction and sends it to the feeding decision unit. The adjustment of the calculation weight in the instruction is reflected in the algorithm as dynamically modifying the value of α. If the vermicompost output needs to be increased (because the actual output rate is lower than the consumption rate), the value of α is adjusted upwards (for example, from 0.8 to 0.9), and under the same inventory and biomass distribution, the system will calculate and execute a larger total feeding amount. If the vermicompost output needs to be reduced (because the actual output rate is higher than the consumption rate), the value of α is adjusted downwards (for example, from 0.8 to 0.7), thereby reducing the total feeding amount. Dynamically adjusting the preset feeding trigger threshold is a different but complementary lever from adjusting α: adjusting the threshold changes the timing and frequency of the system's initiation of feeding operations. Lowering the threshold will cause the system to trigger feeding cycles more frequently; increasing the threshold will lengthen the feeding interval. Adjusting the weight (α) changes the intensity (total amount) of each feeding operation. After each feeding decision is triggered, α determines the proportion of the available inventory for this feeding amount. The combination of the two allows the system to finely control the upstream material input both in terms of macro-rhythm and micro-dose, to more stably achieve the balance goal of downstream output.

[0061] In a preferred embodiment, the feeding decision unit is further configured to: Before calculating the feeding demand priority coefficient based on the real-time biomass data of each breeding bed, the vermicompost output efficiency is obtained; Based on the real-time biomass data and the vermicompost output efficiency of the corresponding breeding bed, an efficiency-corrected biomass value is obtained; Based on the efficiency-corrected biomass value, the feeding demand priority coefficient of each breeding bed is calculated, wherein the higher the efficiency-corrected biomass value of a breeding bed, the higher the feeding demand priority coefficient of the breeding bed.

[0062] Specifically, the vermicompost output efficiency data is obtained: before performing the calculation, the feeding decision unit first needs to obtain the latest vermicompost output efficiency data (denoted as E_i, where i represents the breeding bed number) of each breeding bed. This data is not calculated temporarily, but is a key performance indicator maintained and regularly updated by the system. Its implementation can be consistent with the data source of the efficiency decline evaluation described above: The system maintains a dynamically updated value of earthworm cast production efficiency for each breeding bed. The value is calculated based on a complete statistical period (e.g. the past 30 days) and the formula is: E_i = (cumulative earthworm cast production of this breeding bed in the statistical period) / (cumulative feeding amount of this breeding bed in the statistical period). Cumulative earthworm cast production can be obtained by the metering device equipped in each bed’s independent earthworm cast collection channel, or by associating the source bed with the batch label and weighing the cumulative amount when the earthworm cast warehouse is warehoused. Cumulative feeding amount is directly summed from the historical feeding records of the bed. E_i is a dimensionless ratio value, which is greater than 0, representing the ability of the bed to convert unit feed into fertilizer. The higher the efficiency, the greater the value of E_i.

[0063] Calculate efficiency-corrected biomass value: After obtaining real-time biomass data B_i and corresponding efficiency value E_i, the system performs the following calculation: for each breeding bed i, calculate its efficiency-corrected biomass value B_corrected_i: B_corrected_i = B_i x E_i. The biological and systematic significance of this calculation is that the simple biomass scale (B_i) is weighted by its economic output efficiency (E_i). A population with high biomass but low conversion efficiency will have its corrected value reduced; conversely, a population with medium biomass but extremely high conversion efficiency will have its corrected value improved.

[0064] Calculate feeding demand priority coefficient based on corrected value: This step completely replaces the method of calculating priority coefficient directly based on B_i. Calculate the total value of efficiency-corrected biomass of all breeding beds: B_corrected_total = B_corrected_1 + B_corrected_2 +... + B_corrected_n. Calculate the new feeding demand priority coefficient P_i’ of each breeding bed i: P_i’ = B_corrected_i / B_corrected_total.

[0065] The subsequent decision-making process is connected: after obtaining the priority coefficient P_i' based on efficiency correction, the subsequent target breeding bed determination and accurate feeding amount calculation steps will be completely followed, but the input priority data has changed to P_i'. When determining the target breeding bed, P_i' is compared with the preset priority threshold. When calculating the accurate feeding amount, although the biomass B_tj (used to calculate the biomass weight ratio) is still directly used in the formula F_tj = S_total × (B_tj / B_target_sum) × α, since the target bed screening is based on P_i', the selected target bed group (B_t1, B_t2,... B_tm) is already a group with better conversion efficiency performance. This essentially guides the tailing resources to the earthworm population with higher comprehensive performance (scale × efficiency) in the system.

[0066] In a preferred embodiment, the feeding decision unit is further configured to: When calculating the accurate feeding amount allocated to the target breeding bed, it is queried whether the target breeding bed has entered the monitoring window period after its predicted biomass growth inflection point; If not, it is determined that the breeding bed is in the fast growth period dominated by juvenile worms, and a first feeding amount calculation model is used, which is based on the efficiency corrected biomass value or real-time biomass data and gives a higher unit biomass feeding coefficient; If it has entered, it is determined that the breeding bed has entered the stable conversion period dominated by adult worms, and a second feeding amount calculation model is used, which is based on the efficiency corrected biomass value or real-time biomass data and gives a lower unit biomass feeding coefficient; Wherein, the single feeding amount calculated by the first feeding amount calculation model is higher than the value calculated by the second feeding amount calculation model under the same efficiency corrected biomass value or real-time biomass data.

[0067] Specifically, when the feeding decision unit calculates the accurate feeding amount for a certain determined target breeding bed (denoted as bed k), it first initiates a query to the central processing unit or the cleaning and screening scheduling unit to obtain the key state information of bed k: its predicted biomass growth inflection point time T_inflection_k and whether it has entered the monitoring window period. The system time is compared with T_inflection_k, and combined with the preset monitoring window period length (such as 15 days after the inflection point), the current state of bed k is automatically determined: State A (fast growth period): If the current system time is less than T_inflection_k, or although it is greater than T_inflection_k but does not exceed the window period length, it is determined that the bed has not yet entered the monitoring window period. This stage corresponds to a population dominated by juvenile worms and fast-growing individuals, with physiological needs focusing on building their own biomass (growth).

[0068] State B (Stable Transformation Period): If the current time of the system is greater than or equal to (T_inflection_k + Window Period Length), it is determined that the bed has entered the monitoring window period. This stage corresponds to a significant increase in the proportion of earthworms in the population, a slowdown in growth rate, and a shift in physiological needs to focus on maintaining life activities and efficiently transforming organic matter (producing vermicompost).

[0069] Selection and Application of Differentiated Feeding Models: The system is preset with two sets of feeding amount calculation models, the essential difference between which lies in the different values of the unit biomass feeding coefficient in the models.

[0070] First Feeding Amount Calculation Model (for the Rapid Growth Period): This model uses a higher unit biomass feeding coefficient, denoted as β_high. Let the real-time biomass of bed k be B_k (if the system uses the above process, the efficiency-corrected biomass value B_corrected_k can also be used here. For simplicity of description, B_k is used as an example below). When this model is used, the core part of calculating the feeding amount of bed k is: BaseAmount_k = B_k × β_high. The setting of β_high aims to meet the nutritional needs of fast-growing juvenile earthworms and promote the rapid accumulation of population biomass.

[0071] Second Feeding Amount Calculation Model (for the Stable Transformation Period): This model uses a lower unit biomass feeding coefficient, denoted as β_low, and β_low < β_high. When this model is used, the core part of calculating the feeding amount of bed k is: BaseAmount_k = B_k × β_low. The setting of β_low aims to meet the nutritional needs of adult earthworms for maintaining metabolism and efficient transformation, while avoiding excessive feeding that leads to environmental degradation or resource waste, and using more resources for vermicompost production rather than self-growth.

[0072] Final Feeding Amount Integration: Regardless of which model is used to calculate BaseAmount_k, it will be integrated into the final precise feeding amount calculation formula. For example, the final precise feeding amount F_k of bed k may be calculated as: F_k = S_total × (B_k / B_target_sum) × α, but at this time, B_target_sum used to calculate the biomass weight ratio and B_k used to calculate it already reflect the difference behind the unit demand through β_high or β_low. Alternatively, in a more direct implementation, BaseAmount_k itself, after standardization and total amount constraint adjustment, is directly used as F_k. This judgment and model selection process is embedded in the step of calculating the feeding amount for each target bed by the feeding decision unit and is automatically executed.

[0073] The embodiment identifies and responds to the natural growth stage changes within the earthworm population, making feeding management more in line with biological laws. In the rapid growth period, sufficient nutritional support is provided to promote the healthy development of the population; in the stable transformation period, precise maintenance nutrition is provided to optimize transformation efficiency and save resources. This helps to achieve the optimal balance between growth and output throughout the life cycle of the population, thereby improving the long-term production efficiency and stability of the entire breeding system.

[0074] Embodiment 2 The embodiment proposes a linkage intelligent control method for an earthworm-plant circular breeding system, which is implemented based on the linkage intelligent control system as described in the embodiment, and includes the following steps: Collecting tail vegetable yield data and soil fertility data, as well as earthworm biomass data and vermicompost inventory data; Based on the tail vegetable yield data, soil fertility data, earthworm biomass data, and vermicompost inventory data, a combined control strategy for coordinating the control of the material processing and conveying equipment 3 is generated; According to the combined control strategy, the material processing and conveying equipment 3 is driven to perform corresponding material processing and conveying operations, so that the tail vegetables produced by the plant planting layer are converted into feed for the earthworm breeding layer, and the vermicompost produced by the earthworm breeding layer is transported to the plant planting layer as fertilizer, achieving the recycling and utilization of breeding waste within the system.

[0075] The principles and implementation modes of the present application are described herein using specific examples, and the above examples are only used to help understand the method and its core idea. The above description is only the preferred embodiment of the present application. It should be noted that due to the limited nature of the language expression, there are objectively infinite specific structures, and for ordinary skilled persons in the technical field, without departing from the principles of the present application, a number of improvements, refinements or changes can be made, or the above technical features can be combined in an appropriate manner; these improvements, refinements, changes or combinations, or the direct application of the inventive concept and technical solution to other occasions without improvement, shall be regarded as the protection scope of the present application.

Claims

1. A linkage intelligent control system for an earthworm-plant circular farming system, the earthworm-plant circular farming system comprising a system frame, the system frame comprising upper and lower plant planting layers and earthworm farming layers, and material handling and conveying equipment (3) disposed therebetween; characterized in that, include: Planting module (1), located in the plant planting layer, includes at least one planting area, a soil fertility sensing unit for monitoring the soil nutrient status in each planting area, and a tail vegetable metering unit for calculating the tail vegetable yield. Earthworm farming module (4), located in the earthworm farming layer, includes at least one farming bed, a biomass sensing unit for monitoring the biomass of earthworms in each farming bed, and a worm castings bin for storing and outputting worm castings; The linkage intelligent control module (2) is connected to the plant planting module (1) and the earthworm breeding module (4) respectively. It is configured to make linkage decisions based on multi-source data from the plant planting module (1) and the earthworm breeding module (4), and control the material processing and conveying equipment (3) to perform corresponding operations so as to realize the resource recycling of breeding waste.

2. The intelligent control system for the earthworm-plant recirculating aquaculture system according to claim 1, characterized in that: The linkage intelligent control module (2) includes: Central processing unit; The feeding decision unit is connected to the central processing unit and is configured to receive and process data from the tail vegetable metering unit and each of the biomass sensing units to generate and send a first control command to the waste treatment and feeding device in the material processing and conveying equipment (3). The fertilization decision unit is connected to the central processing unit and is configured to receive and process data from each of the soil fertility sensing units and the vermicompost bins to generate and send a second control command to the fertilization conveying device in the material handling and conveying equipment (3). The cleaning and screening scheduling unit is connected to the central processing unit and is configured to generate and send a third control command to the cleaning and screening device in the material handling and conveying equipment (3) according to a preset cycle or a received external trigger signal.

3. The intelligent control system for the earthworm-plant recirculating aquaculture system according to claim 2, characterized in that: The feeding decision unit is specifically configured for: The total inventory of leftover vegetables reported by the leftover vegetable metering unit and the real-time biomass data reported by the biomass sensing unit corresponding to each breeding bed are obtained in real time. When the total inventory of the leftover vegetables reaches the preset feeding trigger threshold, the feeding demand priority coefficient of each breeding bed is calculated based on the real-time biomass data of each breeding bed. The breeding bed with higher real-time biomass data has a higher feeding demand priority coefficient. Based on the feeding demand priority coefficient of each breeding bed, at least one target breeding bed in the current feeding cycle is determined from all breeding beds; Based on the total stock of the vegetable waste and the real-time biomass data of each of the target breeding beds, the precise feeding amount allocated to each of the target breeding beds is calculated; Generate the first control command, which includes the identifier of each target breeding bed and its corresponding precise feeding amount.

4. The intelligent control system for the earthworm-plant recirculating aquaculture system according to claim 2, characterized in that: The fertilization decision unit is specifically configured for: Real-time acquisition of at least one key nutrient content data reported by the soil fertility sensing unit corresponding to each planting area, as well as real-time inventory data of the vermicomposting bin; When the real-time inventory data is higher than the preset safety stock threshold, the following cyclical decision is executed for each planting area: The key nutrient content data of the planting area are compared with the preset optimal range of the nutrient content; If the key nutrient content data is lower than the lower limit of the preset optimal content range, the planting area is marked as a candidate fertilization area for this fertilization cycle, and the fertilization urgency level of the candidate fertilization area is calculated based on the soil nutrient deficit; the soil nutrient deficit is the difference between the key nutrient content data and the lower limit of the preset optimal content range. Summarize the fertilization urgency levels of all candidate fertilization areas, and determine at least one target fertilization area for this execution based on the level; Based on the soil nutrient deficit in the target fertilization area, the real-time inventory data of the vermicompost, and the preset single fertilization limit, the recommended fertilization amount for each target fertilization area is calculated. Generate a second control instruction that includes the identifier of each target fertilization zone and its corresponding recommended fertilization amount.

5. The intelligent control system for the earthworm-plant recirculating aquaculture system according to claim 2, characterized in that: The cleaning and screening scheduling unit is specifically configured for: Real-time acquisition of historical feeding records and corresponding biomass growth curves for each of the aforementioned breeding beds; Based on the historical feeding records and biomass growth curves, a biomass growth-feeding response model for each culture bed is constructed and trained, and the model is used to predict the inflection point of biomass growth for that culture bed; the inflection point of biomass growth is defined as the theoretical point in time when the biomass growth rate begins to slow down continuously. For any culture bed, when the system time enters the monitoring window period after the predicted biomass growth inflection point, a high-frequency assessment of its screening needs is initiated. The third control command is generated when the same breeding bed is simultaneously marked as having both efficiency decline and growth stagnation in the high-frequency assessment, and the cumulative running time since the last screening operation has reached the preset minimum operation interval.

6. The intelligent control system for the earthworm-plant recirculating aquaculture system according to claim 5, characterized in that: The high-frequency evaluation includes: Efficiency decline assessment: Calculate the vermicompost production efficiency of the breeding bed during the monitoring window period. The vermicompost production efficiency is the ratio of the cumulative vermicompost production to the cumulative feeding amount during the window period. If the current production efficiency decreases by more than a first preset threshold compared to the baseline period before the inflection point, it is marked as efficiency decline. Growth Stagnation Assessment: Obtain real-time biomass data of the aquaculture bed during the monitoring window period. If the biomass growth value is lower than the preset extremely low growth threshold, it is marked as growth stagnation.

7. The intelligent control system for the earthworm-plant recirculating aquaculture system according to claim 2, characterized in that: The central processing unit is also configured to: The inventory change rate of the vermicompost bin is monitored and calculated in real time as the actual vermicompost production rate; at the same time, the average fertilizer consumption rate is calculated based on the historical recommended fertilizer amount in the second control instruction of the fertilization decision unit. The difference between the actual rate of earthworm casting production and the average rate of fertilizer consumption is calculated as the earthworm casting supply and demand balance deviation. When the supply and demand imbalance of vermicompost continues to exceed the preset reasonable range, a parameter adjustment instruction is generated and sent to the feeding decision unit. The parameter adjustment command is used to dynamically adjust the preset feeding trigger threshold and / or the calculation weight of the precise feeding amount in the feeding decision unit, so as to indirectly adjust the earthworm casting production rate by adjusting the feeding amount, so that the actual earthworm casting production rate approaches the average fertilizer consumption rate.

8. The intelligent control system for the earthworm-plant recirculating aquaculture system according to claim 3, characterized in that: The feeding decision unit is also configured to: Before calculating the feeding demand priority coefficient based on the real-time biomass data of each breeding bed, the vermicompost production efficiency is obtained first. Based on the real-time biomass data and the corresponding earthworm castings production efficiency of the breeding bed, an efficiency-corrected biomass value is obtained. Based on the efficiency-corrected biomass value, the feeding requirement priority coefficient for each culture bed is calculated, wherein the culture bed with the higher efficiency-corrected biomass value has a higher feeding requirement priority coefficient.

9. The intelligent control system for the earthworm-plant recirculating aquaculture system according to claim 5, characterized in that: The feeding decision unit is also configured to: When calculating the precise feeding amount allocated to the target culture bed, it is queried whether the target culture bed has entered the monitoring window period after its predicted biomass growth inflection point; If the worms do not enter the breeding bed, it is determined that the breeding bed is in a rapid growth period dominated by juvenile worms. The first feeding amount calculation model is adopted. This model is based on the efficiency correction of biomass value or real-time biomass data and assigns a higher feeding coefficient per unit biomass. If it has entered the stable transition period dominated by adult earthworms, the breeding bed is determined to have entered the stable transition period dominated by adult earthworms. The second feeding amount calculation model is adopted. This model is based on the efficiency correction biomass value or real-time biomass data and assigns a lower unit biomass feeding coefficient. The single feeding amount calculated by the first feeding amount calculation model is higher than the value calculated by the second feeding amount calculation model under the same efficiency correction biomass value or real-time biomass data.

10. A linkage intelligent control method for an earthworm-plant recirculating aquaculture system, implemented based on the linkage intelligent control system as described in any one of claims 1-9, characterized in that: Includes the following steps: Data on vegetable waste yield and soil fertility were collected, along with data on earthworm biomass and vermicompost inventory. Based on the tail vegetable production data, soil fertility data, earthworm biomass data and earthworm castings inventory data, a combined control strategy is generated for coordinating and controlling the material handling and conveying equipment (3). According to the combined control strategy, the material handling and conveying equipment (3) is driven to perform corresponding material handling and conveying operations, so that the tail vegetables produced by the plant planting layer are converted into feed for the earthworm breeding layer, and the earthworm castings produced by the earthworm breeding layer are transported to the plant planting layer as fertilizer, thereby realizing the resource recycling of breeding waste in the system.

Images