Intelligent courtyard ecological collaborative management system and method

Abstract

The invention discloses an intelligent courtyard ecological collaborative management system and method. The system comprises a sensing layer, a control layer and an execution layer. The sensing layer is provided with a multi-mode sensor array, comprises an environment monitoring module, an equipment state monitoring module and a biological recognition module, can collect multi-mode data such as illumination, temperature and humidity, equipment operation parameters and plant growth states, and transmits the multi-mode data to the control layer through wireless communication. The control layer adopts a federated learning framework to fuse multi-modal data, extracts features through a deep residual network, generates an environment state matrix by using an attention mechanism, and operates a Markov decision model to output a decision control command and a response level. And the execution layer adjusts the on-off state and operation parameters of the regulation and control equipment according to the output, so that intelligent cooperative control of the irrigation system, the lighting equipment and the like is realized. According to the system, through multi-modal data acquisition and deep fusion analysis, comprehensive perception and accurate regulation and control of the ecological environment of the courtyard are realized, and the management efficiency and the resource utilization rate are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent devices, and in particular to an intelligent courtyard ecological collaborative management system and method. Background Art

[0002] Traditional garden management systems suffer from numerous shortcomings. First, they rely on single sensors for data collection, capturing only limited environmental parameters. They are unable to comprehensively monitor environmental factors such as light, air quality, and soil fertility, as well as biological activity within the garden and equipment operating parameters, leading to incomplete management decisions. Second, data processing capabilities are weak, and data from different sensor types is independent and lacks fusion analysis, making it difficult to uncover potential correlations between data. Centralized processing architectures also pose privacy risks. Furthermore, system decisions rely on preset thresholds, resulting in delayed responses to sudden environmental changes. Independent operation of each device lacks unified scheduling, leading to significant energy waste and a low level of intelligence, making them unable to meet the needs of efficient garden ecological management. Therefore, a collaborative management system that can achieve multimodal data collection, fusion analysis, and intelligent decision-making is urgently needed. Summary of the Invention

[0003] In view of the above problems, the present invention provides an intelligent courtyard ecological collaborative management system and method to solve the problems that the existing courtyard management system lacks unified scheduling, has a low degree of intelligence, and cannot meet the needs of efficient management of the courtyard ecological environment.

[0004] In a first aspect, the present application provides an intelligent courtyard ecological collaborative management system, comprising:

[0005] Perception layer: deploys a multimodal sensor array to collect multimodal data and send the multimodal data to the control layer via a wireless communication protocol. The multimodal sensor array includes an environmental monitoring module, an equipment status monitoring module, and a biometric recognition module.

[0006] Control layer: This layer uses a federated learning framework for multimodal data fusion, extracts sensor features through a deep residual network, and uses an attention mechanism to weight and generate an environmental state matrix. It also runs a Markov decision model to output response levels and decision control commands.

[0007] Execution layer: deploys control equipment and adjusts the on / off status and operating parameters of the control equipment according to the response level and decision control commands.

[0008] Furthermore, the environmental monitoring module includes a millimeter-wave radar, a spectral sensor, and a soil ion selective electrode array, wherein the data of the millimeter-wave radar and the spectral sensor are fused through a Kalman filter algorithm; the equipment status monitoring module includes a pipeline pressure sensor and a liquid level sensor; the biometric recognition module includes a multispectral camera and a voiceprint sensor for identifying plant pests and diseases and wildlife activities;

[0009] The control equipment includes first type equipment, second type equipment, and third type equipment. The response priorities of the first type equipment, second type equipment, and third type equipment decrease in sequence. The first type equipment includes an audible and visual alarm device and a pressure protection module. The second type equipment includes a spray device, an ultrasonic insect repellent module, and a mosquito lamp module. The third type equipment includes an intelligent irrigation valve and a rainwater collection device.

[0010] Furthermore, the data from the millimeter-wave radar and spectral sensor are fused through the Kalman filter algorithm, including:

[0011] A three-dimensional point cloud model of mosquito flight trajectory is established, and the density of target mosquito swarms is identified by DBSCAN clustering algorithm. When the mosquito density value D is greater than the mosquito density threshold D max When the alarm is on, the linkage spray device releases the disinfectant.

[0012] Furthermore, the input of the Markov decision model includes a state space vector and an action space vector, the state space vector is defined as S = {soil moisture gradient, pest and disease risk index, equipment energy consumption ratio}, the action space vector is defined as A = {irrigation amount, light intensity, insect repellent frequency}, and the output of the Markov decision model is a response level and a decision control command.

[0013] Furthermore, adjusting the on / off state and operating parameters of the control device according to the response level and the decision control command includes dynamically adjusting the irrigation amount according to the soil moisture deviation e(t), and the calculation formula is as follows:

[0014]

[0015] Where u(t) is the irrigation amount, K p ,K i ,K d为 The fuzzy rule base is dynamically loaded based on the soil type classification database, and the fuzzy rule base is automatically optimized and updated through historical irrigation data and plant growth conditions.

[0016] Furthermore, adjusting the on / off state and operating parameters of the control device according to the response level and the decision control command includes dynamically adjusting the attractant release concentration according to the biological activity law of the target mosquito species, and the release amount calculation formula is:

[0017]

[0018] Among them, Q is the amount of attractant released per unit time, t is the diurnal time variable, D is the real-time mosquito density, k1, k2, τ are species characteristic coefficients;

[0019] When a specific mosquito species is identified, the directional ultrasonic insect repellent module is activated to release ultrasonic waves to repel the mosquitoes.

[0020] Furthermore, the biometric recognition module further includes a mosquito density sensor, and the execution layer further includes a high-voltage power grid;

[0021] The mosquito density sensor comprises:

[0022] An electrical signal detection module is used to detect voltage pulse signals of the high-voltage power grid in real time;

[0023] Pulse signal filtering module, used to filter voltage pulse signals with a duration less than 50μs;

[0024] a counting module, electrically connected to the pulse signal filtering module, for counting the number of times that the voltage change in a unit time exceeds a preset amplitude in the voltage pulse signal filtered by the signal filtering module;

[0025] When the count value counted by the counting module exceeds a preset number of times, the decision control command further includes starting a mosquito disinfecting module, and the mosquito disinfecting module includes a spray device or an ultrasonic insect repellent module.

[0026] Furthermore, the execution layer also includes a wind-solar complementary power supply module, including a solar panel, a wind turbine and a storage module. The solar panel is used to convert solar energy into electrical energy, the wind turbine is used to convert wind energy into electrical energy, and the storage module is used to store the electrical energy converted by the solar panel or wind turbine and provide power for the control equipment.

[0027] Furthermore, it also includes:

[0028] The user interaction layer is used to obtain and display the response level and multimodal data collected by the perception layer in real time, as well as receive control commands from the user end to adjust the opening and closing status and operating parameters of the control equipment.

[0029] In a second aspect, the present application further provides a smart courtyard ecological collaborative management method, which is applicable to the smart courtyard ecological collaborative management system as described in the first aspect of the present application, and the method comprises the following steps:

[0030] Collect environmental data at preset periodic intervals through a multimodal sensor array and detect device status data in real time;

[0031] Running the decision model optimized by federated learning to generate a control instruction set, wherein the control instruction set includes control commands for different types of control devices;

[0032] After receiving the control instruction, the control device at the execution layer performs the response action according to the preset priority.

[0033] Different from the existing technology, in the above technical solution, the present invention discloses an intelligent courtyard ecological collaborative management system and method, which includes a perception layer, a control layer and an execution layer. The perception layer deploys a multimodal sensor array, including environmental monitoring, equipment status monitoring and biometric recognition modules, which can collect multimodal data such as light, temperature and humidity, equipment operating parameters, plant growth status, etc., and transmit them to the control layer through wireless communication. The control layer adopts a federated learning framework to fuse multimodal data, extracts features through a deep residual network, uses the attention mechanism to generate an environmental state matrix, runs a Markov decision model to output decision control commands and response levels. The execution layer adjusts the on and off status and operating parameters of the control equipment according to the above output, and realizes intelligent collaborative control of irrigation systems, lighting equipment, etc. Through multimodal data collection and deep fusion analysis, this system realizes comprehensive perception and precise control of the courtyard ecological environment, improves management efficiency and resource utilization, and has distributed deployment and timely dynamic response capabilities.

[0034] The above-mentioned description of the invention content is only an overview of the technical solution of the present invention. In order to enable ordinary technicians in this field to more clearly understand the technical solution of the present invention, and then implement it according to the text of the specification and the contents recorded in the drawings, and to make the above-mentioned objects and other objects, features and advantages of the present invention easier to understand, the following is an explanation in combination with the specific implementation methods and drawings of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings are only used to illustrate the principles, implementations, applications, features, and effects of the specific embodiments of the present invention and other related contents, and are not to be considered as limiting the present invention.

[0036] In the drawings of the specification:

[0037] Figure 1 The module diagram of an intelligent courtyard ecological collaborative management system involved in a specific implementation method is shown.

[0038] Figure 2 The present invention is a flowchart of a smart courtyard ecological collaborative management method involved in a specific implementation method.

[0039] The reference numerals in the above drawings are described as follows:

[0040] 1. Intelligent courtyard ecological collaborative management system;

[0041] 10. Perception layer;

[0042] 20. Control layer;

[0043] 30. Executive layer;

[0044] 40. User interaction layer. DETAILED DESCRIPTION

[0045] In order to explain in detail the possible application scenarios, technical principles, specific solutions that can be implemented, and the purpose and effects of the present invention, the following is a detailed description of the specific embodiments listed in conjunction with the accompanying drawings. The embodiments described herein are only used to more clearly illustrate the technical solutions of the present invention and are therefore only examples and are not intended to limit the scope of protection of the present invention.

[0046] References to "embodiments" herein mean that the specific features, structures, or characteristics described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the term "embodiment" in various places in the specification does not necessarily refer to the same embodiment, nor does it particularly limit its independence or relevance to other embodiments. In principle, in the present invention, as long as there are no technical contradictions or conflicts, the various technical features mentioned in the embodiments can be combined in any manner to form a corresponding implementable technical solution.

[0047] Unless otherwise defined, the technical terms used herein have the same meanings as those generally understood by those skilled in the art to which the present invention belongs. The use of relevant terms herein is only for describing specific embodiments and is not intended to limit the present invention.

[0048] In the description of the present invention, the term "and / or" is used to describe a logical relationship between objects, indicating that three possible relationships exist. For example, A and / or B means: A exists, B exists, and both A and B exist. Furthermore, the character " / " generally indicates that the objects are in a logical "or" relationship.

[0049] In the present invention, terms such as "first" and "second" are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship of quantity, priority or sequence between these entities or operations.

[0050] Without further restrictions, in the present invention, the words "include", "comprise", "have" or other similar expressions used in sentences are intended to cover non-exclusive inclusion. These expressions do not exclude the presence of additional elements in the process, method or product including the elements, so that the process, method or product including a series of elements may include not only those limited elements, but also other elements not explicitly listed, or also include elements inherent to such process, method or product.

[0051] Consistent with the understanding in the Examination Guidelines, in the present invention, expressions such as "greater than," "less than," and "exceed" are understood to exclude the number itself; expressions such as "above," "below," and "within" are understood to include the number itself. Furthermore, in the description of the embodiments of the present invention, "multiple" means two or more (including two), and similar expressions related to "multiple," such as "multiple groups" and "multiple times," are also understood in this manner, unless otherwise specifically defined.

[0052] In the description of the embodiments of the present invention, the space-related expressions used, such as "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "vertical", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicate the orientation or position relationship based on the orientation or position relationship shown in the specific embodiments or drawings, and are only for the convenience of describing the specific embodiments of the present invention or facilitating the reader's understanding, and do not indicate or imply that the device or component referred to must have a specific position, a specific orientation, or be constructed or operated in a specific orientation, and therefore cannot be understood as limiting the embodiments of the present invention.

[0053] Unless otherwise expressly specified or limited, in the description of the embodiments of the present invention, the terms "installed", "connected", "connected", "fixed", "set", etc. used should be understood in a broad sense. For example, the "connection" can be a fixed connection, a detachable connection, or an integrated setting; it can be a mechanical connection, a signal connection, or a communication connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. For those skilled in the art of the technology to which the present invention belongs, the specific meanings of the above terms in the embodiments of the present invention can be understood according to specific circumstances.

[0054] In the first aspect, Figure 1 As shown, the present application provides an intelligent courtyard ecological collaborative management system 1, comprising:

[0055] Perception layer 10: Deploys a multimodal sensor array to collect multimodal data and send the multimodal data to the control layer via a wireless communication protocol. The multimodal sensor array includes an environmental monitoring module, an equipment status monitoring module, and a biometric recognition module.

[0056] Control layer 20: Uses a federated learning framework for multimodal data fusion, extracts sensor features through a deep residual network, uses an attention mechanism to weight and generate an environmental state matrix, and runs a Markov decision model to output response levels and decision control commands.

[0057] Execution layer 30: deploys control equipment and adjusts the on / off status and operating parameters of the control equipment according to the response level and decision control command.

[0058] In some embodiments, the environmental monitoring module includes a millimeter-wave radar, a spectral sensor, and a soil ion selective electrode array, wherein the data from the millimeter-wave radar and the spectral sensor are fused using a Kalman filter algorithm; the equipment status monitoring module includes a pipeline pressure sensor and a liquid level sensor; and the biometric recognition module includes a multispectral camera and a voiceprint sensor for identifying plant pests and diseases and wildlife activities.

[0059] The control equipment includes first type equipment, second type equipment, and third type equipment. The response priorities of the first type equipment, second type equipment, and third type equipment decrease in sequence. The first type equipment includes an audible and visual alarm device and a pressure protection module. The second type equipment includes a spray device, an ultrasonic insect repellent module, and a mosquito lamp module. The third type equipment includes an intelligent irrigation valve and a rainwater collection device.

[0060] In this embodiment, the environmental monitoring module uses millimeter-wave radar to obtain information about the courtyard's spatial structure and object movement (such as the movement of people and animals). Spectral sensors analyze vegetation health (chlorophyll content and signs of pests and diseases). A soil ion-selective electrode array monitors soil nutrients (NPK value, pH value) in real time. Kalman filtering is used to fuse millimeter-wave and spectral data, eliminating environmental noise and improving state estimation accuracy.

[0061] The equipment status monitoring module includes pipeline pressure sensors and liquid level sensors. The pipeline pressure sensors and liquid level sensors work together to monitor the pressure changes of the irrigation system and the water level of the water storage equipment, realizing water leakage warning and optimized water resource scheduling.

[0062] The biometric module includes a multispectral camera, which identifies the characteristics of plant leaf spots through imaging in different bands. The voiceprint sensor captures the sound wave characteristics of biological activities such as birds and insects, and combines it with machine learning algorithms to achieve early warning monitoring of pests and diseases.

[0063] In this embodiment, the control layer uses a federated learning framework for multimodal data fusion, including: each node (such as a courtyard partition controller) locally trains a deep residual network (ResNet) to extract sensor features, and only exchanges model parameters rather than raw data, thereby ensuring user privacy while achieving cross-regional knowledge sharing (such as optimizing irrigation strategies under different climatic conditions).

[0064] In this embodiment, the attention mechanism weighting includes: dynamically weighting the extracted multimodal features (environmental parameters, device status, biological information), for example, increasing the weight of meteorological data during a rainstorm warning, and generating an environmental status matrix that reflects the current ecological status of the courtyard.

[0065] In this embodiment, the Markov decision model predicts the future state transition probability based on the environmental state matrix, combines reinforcement learning to optimize the decision sequence, and outputs graded response instructions (such as emergency, important, and routine) and equipment control parameters.

[0066] When controlling the on / off status or operating parameters of a control device, it is preferred to control according to the priority of the control device, as follows:

[0067] Type 1 equipment (highest priority): When a burst or gas leak in a pipeline is detected, the sound and light alarm device is immediately activated, issuing an audible and visual alarm prompt, and the pressure protection module automatically cuts off the water or gas source to prevent further damage.

[0068] Type 2 equipment (second highest priority): When pests or wildlife damage is identified, the spray device releases biological pesticides, the ultrasonic insect repellent module repels harmful animals, and the mosquito lamp module traps pests.

[0069] Type 3 equipment (conventional priority): Dynamically adjusts the opening of intelligent irrigation valves based on soil moisture and weather forecasts, and rainwater collection devices automatically start storing water during the rainy season to achieve water resource recycling.

[0070] This solution breaks through the limitations of traditional single-sensor monitoring by integrating millimeter-wave radar, spectral analysis, and voiceprint recognition technologies into courtyard management. Through a privacy protection mechanism driven by federated learning, a balance between data privacy and collaborative optimization can be achieved in a distributed intelligent device environment. A dynamic control strategy with hierarchical response can automatically dispatch devices of different priorities based on the urgency of the event, achieving efficient resource utilization. This application is suitable for scenarios such as villa courtyards and agricultural sightseeing parks, and has broad market application prospects.

[0071] In some embodiments, the data of the millimeter wave radar and the spectral sensor are integrated through the Kalman filter algorithm, including: establishing a three-dimensional point cloud model of the mosquito flight trajectory, identifying the density of the target mosquito swarm through the DBSCAN clustering algorithm, and when the mosquito density value D is greater than the mosquito density threshold D max When the alarm is on, the linkage spray device releases the disinfectant.

[0072] In actual application, the millimeter-wave radar scans the space at a frequency of 10Hz to generate point cloud data containing distance, angle, and speed. The spectral sensor synchronously collects images and identifies the unique spectral characteristics of mosquitoes (such as spectral flickering corresponding to the wing vibration frequency) through feature extraction algorithms. The radar and spectral data are then synchronized through timestamps, and the two are converted to the same coordinate system using the external parameter matrix. The Kalman filter algorithm is used to establish a state space model for each detection point, predict the position at the next moment, and update the estimated value to eliminate random noise. When identifying the density of the target mosquito swarm, the spatial neighborhood radius ε = 0.5m and the minimum point density can be set to 5 points / cubic meter to identify dense mosquito swarm areas. The number of mosquitoes per unit volume is then calculated based on the clustering results to generate a three-dimensional density distribution map. When the local area density D>D max (For example, the preset value is 50 pieces / m 3 ), the spray device is triggered. Specifically, the center position of the mosquito swarm can be determined through coordinate mapping, the angle and dosage of the nozzle can be controlled, and the attractant can be released in a targeted manner.

[0073] Preferably, multiple monitoring areas can be set up, and the density can be calculated independently for each monitoring area and the corresponding spray device can be controlled to save resources.

[0074] In some embodiments, the input of the Markov decision model includes a state space vector and an action space vector, the state space vector is defined as S = {soil moisture gradient, pest and disease risk index, equipment energy consumption ratio}, and the action space vector is defined as A = {irrigation amount, light intensity, insect repellent frequency}, and the output of the Markov decision model is a response level and a decision control command.

[0075] In this embodiment, the soil moisture sensor network collects data every 15 minutes and calculates moisture gradients at different depths. Pest and disease monitoring cameras capture images once an hour, analyze leaf images using computer vision algorithms, and calculate a risk index. The device's electric meter collects energy consumption data in real time and calculates the energy consumption ratio (current energy consumption / baseline energy consumption). Specifically, the sensor data can be normalized to the interval [0, 1] to construct a state space vector S = [humidity gradient, pest and disease risk index, energy consumption ratio]. The Markov decision process inference process is as follows:

[0076] The model receives the current state vector, selects the optimal action through the strategy function π(s), calculates the response level (0-100) corresponding to different decision urgency levels, and then generates control commands (irrigation amount, light intensity, and insect repellent frequency) based on the decision results. The control layer controls the execution layer to adjust the corresponding parameters, and records the actual effects for model iterative optimization.

[0077] Through the Markov decision process, the system can automatically adjust the control strategy based on historical data and environmental changes, while optimizing the three goals of soil moisture, pest and disease control, and equipment energy consumption. By adjusting the weight of the reward function, it can flexibly adapt to the needs of different planting scenarios.

[0078] In some embodiments, adjusting the on / off state and operating parameters of the control device according to the response level and the decision control command includes dynamically adjusting the irrigation amount according to the soil moisture deviation e(t), and the calculation formula is as follows:

[0079]

[0080] Where u(t) is the irrigation amount, K p ,K i ,K d为 The fuzzy rule base is dynamically loaded based on the soil type classification database, and the fuzzy rule base is automatically optimized and updated through historical irrigation data and plant growth conditions.

[0081] In this embodiment, a soil type classification database stores parameters such as water retention properties and permeability coefficients for different soils, while a fuzzy rule base constructs irrigation rules based on historical data and expert knowledge. The PID controller first matches the corresponding initial PID parameters from the database based on the current soil type. A fuzzy inference system (FIS) is then used to optimize the PID parameters based on the following input variables: soil moisture deviation e(t), deviation change rate de(t) / dt, and soil type characteristic parameters (such as sand content and organic matter content). During the automatic rule optimization process, historical irrigation data and crop growth indicators (such as yield and plant height) are first regularly collected. A genetic algorithm or gradient descent method is then used to adjust the weights and membership function parameters of the fuzzy rules. The impact of the optimized rules on crop growth is then evaluated, and the rule base is iteratively updated.

[0082] This solution dynamically adjusts irrigation volume based on real-time soil moisture, significantly improving water conservation compared to traditional timed irrigation. It also automatically loads optimal PID parameters for different soil types (sand, loam, and clay), eliminating the need for manual retuning of traditional PID controllers for varying soil conditions.

[0083] In some embodiments, adjusting the on / off state and operating parameters of the control device according to the response level and the decision control command includes dynamically adjusting the attractant release concentration according to the biological activity pattern of the target mosquito species, and the release amount calculation formula is:

[0084]

[0085] Among them, Q is the amount of attractant released per unit time, t is the diurnal time variable, D is the real-time mosquito density, k1, k2, τ are species characteristic coefficients;

[0086] When a specific mosquito species is identified, the directional ultrasonic insect repellent module is activated to release ultrasonic waves to repel the mosquitoes.

[0087] In this embodiment, the attractant release and ultrasonic repellent are performed alternately, which can maximize the resource utilization efficiency and mosquito disinfecting effect.

[0088] In some embodiments, the biometric recognition module further includes a mosquito density sensor, and the execution layer further includes a high-voltage power grid;

[0089] The mosquito density sensor comprises:

[0090] An electrical signal detection module is used to detect voltage pulse signals of the high-voltage power grid in real time;

[0091] Pulse signal filtering module, used to filter voltage pulse signals with a duration less than 50μs;

[0092] a counting module, electrically connected to the pulse signal filtering module, for counting the number of times that the voltage change in a unit time exceeds a preset amplitude in the voltage pulse signal filtered by the signal filtering module;

[0093] When the count value counted by the counting module exceeds a preset number of times, the decision control command further includes starting a mosquito disinfecting module, and the mosquito disinfecting module includes a spray device or an ultrasonic insect repellent module.

[0094] In this embodiment, when the high-voltage power grid is operating, mosquitoes contacting the grid trigger a voltage pulse signal. The electrical signal detection module uses an analog-to-digital converter (ADC) to acquire the grid voltage waveform in real time. The sampling frequency must be ≥4 kHz to ensure signal integrity. The pulse signal filtering module uses a hardware comparator combined with a digital filtering algorithm to filter pulses with a duration of less than 50 μs. This threshold is set based on experimental data on mosquito size (the typical pulse width of a mosquito triggering an electric shock is 50-200 μs), effectively eliminating short-term interference such as rain splash and electromagnetic noise. The counting module then performs a sliding window count on the filtered valid pulses (for example, a 5-minute time window), calculating the number of times the voltage pulse amplitude exceeds a preset value (e.g., ±10V) per unit time. Preferably, the preset number of times can be dynamically adjusted based on the ambient temperature and humidity (e.g., for every 5°C increase in temperature, the preset number of times threshold can be lowered by 10%). This is achieved through the microcontroller's built-in LUT (lookup table) to avoid misjudgments caused by seasonal changes. When the count value exceeds the dynamic threshold, a corresponding control command is triggered, activating the spray device for disinfection.

[0095] The above solution uses 50μs pulse filtering technology to filter out invalid pulses, reduce the number of invalid starts of the high-voltage power grid, and reduce power consumption. At the same time, it can reduce the probability of false triggering due to rain, flying insect debris, etc. to less than 5%, significantly improving the reliability of mosquito density detection.

[0096] In some embodiments, the execution layer also includes a wind-solar complementary power supply module, including a solar panel, a wind turbine, and a storage module. The solar panel is used to convert solar energy into electrical energy, the wind turbine is used to convert wind energy into electrical energy, and the storage module is used to store the electrical energy converted by the solar panel or wind turbine and provide power for the control device. The above solution provides all-weather sustainable energy support for the intelligent courtyard mosquito control system through the wind-solar complementary power supply module, improving the endurance of the system operation.

[0097] In some embodiments, it further includes:

[0098] The user interaction layer 40 is used to obtain and display the response level and multimodal data collected by the perception layer in real time, and receive control commands from the user end to adjust the on / off status and operating parameters of the control equipment.

[0099] Interactive components such as action buttons, sliders, and drop-down menus are provided on the interface to facilitate user input of control commands. For example, a slider for adjusting the irrigation amount and a drop-down menu for selecting the frequency of insect repellent can be provided. Users can use these components to adjust the operating parameters of the control device, thereby achieving remote control of the executing device.

[0100] In the second aspect, Figure 2As shown, the present application also provides a smart courtyard ecological collaborative management method, which is applicable to the smart courtyard ecological collaborative management system as described in the first aspect of the present application, and the method includes the following steps:

[0101] Step S201: Collecting environmental data at preset periodic intervals through a multimodal sensor array and detecting device status data in real time;

[0102] Step S202: running the decision model optimized by federated learning to generate a control instruction set, wherein the control instruction set includes control commands for different types of control devices;

[0103] Step S203: After receiving the control instruction, the control device of the execution layer executes a response action according to a preset priority.

[0104] In this embodiment, the control commands are for different types of control devices, and different types of control devices belong to different ecological modules. The ecological modules include any one or more of the following: intelligent mosquito monitoring module, intelligent pressure monitoring module, intelligent meteorological linkage module, soil moisture closed-loop control module, trace element monitoring and early warning module, intelligent green plant maintenance module, adaptive lighting module, ambient sound field adjustment module, outdoor asset digital management module, intelligent fish pond ecological module, intelligent swimming pool control module, and intelligent pavilion control module. The following is an expanded description of each ecological module and its working principle:

[0105] (1) Intelligent mosquito monitoring module

[0106] This module features a built-in attractant tank, optimized based on mosquito biology and perception mechanisms, for an integrated design that combines mosquito attraction and killing. A built-in monitoring unit collects real-time mosquito population data, triggering a coordinated work process based on scenario requirements, creating a closed "monitoring-response" loop.

[0107] (2) Intelligent pressure monitoring module

[0108] High-precision pressure sensors are deployed at key points in the spray / irrigation equipment pipelines to establish a real-time pressure monitoring network. When pressure deviates from preset thresholds, the system immediately shuts off power and sends multi-level alarms to the user via wireless communication, enabling rapid response to abnormal operating conditions.

[0109] (3) Intelligent meteorological linkage module

[0110] An integrated rain sensor and anemometer form an environmental sensing array, enabling real-time collection and analysis of meteorological parameters. Users can customize control strategies through an interactive interface (e.g., automatically stopping spraying when wind speeds exceed level 5, or triggering sprinkler system shutdown when moderate to heavy rainfall occurs). Weather forecast API data is also integrated, enabling preventive operation mode to be initiated one hour in advance based on a time series prediction model.

[0111] (4) Soil moisture closed-loop control module

[0112] A three-dimensional monitoring network employs soil moisture sensors with adjustable depths (10-50 cm). A fuzzy PID control algorithm dynamically adjusts irrigation flow, ensuring deviations from target moisture within ±3%. The system supports intelligent switching between drip and sprinkler irrigation modes and prioritizes water from rainwater collection systems, enabling adaptive management of water-saving irrigation.

[0113] (5) Trace element monitoring and early warning module

[0114] Equipped with a multi-channel ion concentration sensor array, it can simultaneously monitor the concentrations of key elements such as N, P, K, and Fe in the soil. An expert system based on plant nutrient requirement models automatically generates specialized fertilizer formula recommendations when element imbalances are detected. Remote data calibration via the mini-program improves the maintainability of monitoring accuracy.

[0115] (6) Intelligent green plant maintenance module

[0116] Equipped with a multispectral imaging system, it captures vegetation images and uses a convolutional neural network algorithm to identify pest and disease levels. When the detected pest and disease level reaches a warning threshold, the robotic arm is activated to perform dead branch pruning and precise biopesticide spraying. An integrated environmental control unit dynamically adjusts the light spectrum and CO2 concentration according to the plant's growth cycle, creating a suitable microenvironment for growth.

[0117] (7) Adaptive lighting module

[0118] A composite sensing unit, comprised of a light sensor and an infrared pyroelectric sensor, dynamically adjusts light intensity and color temperature using a full-color RGBW LED strip. The infrared sensing module identifies human movement in real time, automatically switching between safety warning and ambient lighting modes. A built-in astronomical clock algorithm simulates natural sunrise and sunset light curves to meet circadian lighting needs.

[0119] (8) Ambient sound field adjustment module

[0120] A sound pressure level sensor monitors ambient noise levels in real time, dynamically adjusting background music volume based on an adaptive filtering algorithm. Integrated sound source localization technology enables regional directional broadcasting (e.g., independent audio playback in multiple courtyard zones). A visual recognition module also detects harmful bird species, using a bird call database to match natural enemy soundwave patterns. Directional sound waves are then emitted to achieve contactless bird repellency.

[0121] (9) Outdoor asset digital management module

[0122] An asset tracking network is built using UWB ultra-wideband positioning technology, enabling centimeter-level real-time positioning of high-value assets. Image recognition algorithms automatically generate asset inventories and create digital archives. When abnormal asset movement is detected, on-site audio and visual alarms are triggered and data is pushed to the security platform, forming a three-dimensional security system.

[0123] (10) Smart fish pond ecological module

[0124] Equipped with a multi-parameter water quality monitor (pH, ORP, dissolved oxygen, TDS, etc.), water quality parameter thresholds are set based on a fish species database, automatically controlling water exchange and aeration equipment. The feeding system is linked to a visual recognition module, which dynamically adjusts feed intake by analyzing fish feeding behavior, achieving coordinated management of precision aquaculture and water quality ecological balance.

[0125] (11) Intelligent swimming pool control module

[0126] A multi-dimensional monitoring system is built: the water quality monitoring unit measures health indicators such as pH, residual chlorine concentration, and turbidity in real time; the environmental sensing unit collects parameters such as water temperature, humidity, and light intensity, and combines them with meteorological data to optimize operating modes; and the water level detection unit uses ultrasonic and pressure sensors to achieve dynamic water level balance control. Data fusion from each subsystem drives fully automated management of pool operations.

[0127] (12) Intelligent pavilion control module

[0128] The system utilizes an ARM Cortex-A series processor as the main control unit, supporting multi-tasking parallel processing and real-time operating system (RTOS) deployment. The protection module achieves an IP65 dust and water resistance rating and a wide operating temperature range of -20°C to 65°C through structural sealing design, weather-resistant material selection, and environmental adaptability testing. The automatic diagnostic module integrates a fault self-detection algorithm, providing audible and visual alarms and mobile notifications for abnormal conditions such as sensor disconnection and communication interruption, improving system operation and maintenance efficiency.

[0129] This solution enables the integration of diverse sensor monitoring data, establishes a weather-soil-equipment status correlation model, and improves outdoor resource utilization and intelligent decision-making. Through multimodal data collection and deep fusion analysis, this system achieves comprehensive perception and precise control of the courtyard ecosystem, improving management efficiency and resource utilization.

[0130] Finally, it should be noted that although the above embodiments have been described in the specification and drawings of the present invention, this does not limit the scope of patent protection of the present invention. All technical solutions generated by replacing or modifying equivalent structures or equivalent processes based on the essential concepts of the present invention using the contents recorded in the specification and drawings of the present invention, as well as directly or indirectly implementing the technical solutions of the above embodiments in other related technical fields, are included in the scope of patent protection of the present invention.

Claims

1. An intelligent courtyard ecological collaborative management system, characterized in that: include: Perception layer: deploys a multimodal sensor array to collect multimodal data and send the multimodal data to the control layer via a wireless communication protocol. The multimodal sensor array includes an environmental monitoring module, an equipment status monitoring module, and a biometric recognition module. Control layer: This layer uses a federated learning framework for multimodal data fusion, extracts sensor features through a deep residual network, and uses an attention mechanism to weight and generate an environmental state matrix. It also runs a Markov decision model to output response levels and decision control commands. Execution layer: deploys control equipment and adjusts the on / off status and operating parameters of the control equipment according to the response level and decision control commands.

2. The intelligent courtyard ecological collaborative management system according to claim 1, characterized in that: The environmental monitoring module includes a millimeter-wave radar, a spectral sensor, and a soil ion selective electrode array. The data from the millimeter-wave radar and spectral sensor are fused using a Kalman filter algorithm. The equipment status monitoring module includes a pipeline pressure sensor and a liquid level sensor. The biometric recognition module includes a multispectral camera and a voiceprint sensor to identify plant pests and diseases and wildlife activities. The control equipment includes first type equipment, second type equipment, and third type equipment. The response priorities of the first type equipment, second type equipment, and third type equipment decrease in sequence. The first type equipment includes an audible and visual alarm device and a pressure protection module. The second type equipment includes a spray device, an ultrasonic insect repellent module, and a mosquito lamp module. The third type equipment includes an intelligent irrigation valve and a rainwater collection device.

3. The intelligent courtyard ecological collaborative management system according to claim 2, characterized in that: The data from the millimeter-wave radar and spectral sensor are fused through the Kalman filter algorithm, including: A three-dimensional point cloud model of mosquito flight trajectory is established, and the density of target mosquito swarms is identified by DBSCAN clustering algorithm. When the mosquito density value D is greater than the mosquito density threshold D max When the alarm is on, the linkage spray device releases the disinfectant.

4. The intelligent courtyard ecological collaborative management system according to claim 2, characterized in that: The input of the Markov decision model includes a state space vector and an action space vector. The state space vector is defined as S = {soil moisture gradient, pest and disease risk index, equipment energy consumption ratio}, and the action space vector is defined as A = {irrigation amount, light intensity, insect repellent frequency}. The output of the Markov decision model is a response level and a decision control command.

5. The intelligent courtyard ecological collaborative management system according to claim 2, characterized in that: Adjusting the on / off state and operating parameters of the control device according to the response level and the decision control command includes dynamically adjusting the irrigation amount according to the soil moisture deviation e(t), and the calculation formula is as follows: Where u(t) is the irrigation amount, K p ,K i ,K d为 The fuzzy rule base is dynamically loaded based on the soil type classification database, and the fuzzy rule base is automatically optimized and updated through historical irrigation data and plant growth conditions.

6. The intelligent courtyard ecological collaborative management system according to claim 2, characterized in that: Adjusting the on / off state and operating parameters of the control device according to the response level and the decision control command includes: dynamically adjusting the attractant release concentration according to the biological activity law of the target mosquito species, and the release amount calculation formula is: Among them, Q is the amount of attractant released per unit time, t is the diurnal time variable, D is the real-time mosquito density, k1, k2, τ are species characteristic coefficients; When a specific mosquito species is identified, the directional ultrasonic insect repellent module is activated to release ultrasonic waves to repel the mosquitoes.

7. The intelligent courtyard ecological collaborative management system according to claim 1, characterized in that: The biometric recognition module further includes a mosquito density sensor, and the execution layer further includes a high-voltage power grid; The mosquito density sensor comprises: An electrical signal detection module is used to detect voltage pulse signals of the high-voltage power grid in real time; Pulse signal filtering module, used to filter voltage pulse signals with a duration less than 50μs; a counting module, electrically connected to the pulse signal filtering module, for counting the number of times that the voltage change in a unit time exceeds a preset amplitude in the voltage pulse signal filtered by the signal filtering module; When the count value counted by the counting module exceeds a preset number of times, the decision control command further includes starting a mosquito disinfecting module, and the mosquito disinfecting module includes a spray device or an ultrasonic insect repellent module.

8. The intelligent courtyard ecological collaborative management system according to claim 1, characterized in that: The execution layer also includes a wind-solar complementary power supply module, including solar panels, wind turbines and storage modules. The solar panels are used to convert solar energy into electrical energy, the wind turbines are used to convert wind energy into electrical energy, and the storage modules are used to store the electrical energy converted by the solar panels or wind turbines and provide power for the control equipment.

9. The intelligent courtyard ecological collaborative management system according to claim 1, characterized in that: Also includes: The user interaction layer is used to obtain and display the response level and multimodal data collected by the perception layer in real time, as well as receive control commands from the user end to adjust the opening and closing status and operating parameters of the control equipment.

10. A smart courtyard ecological collaborative management method, characterized in that: Applicable to the intelligent courtyard ecological collaborative management system according to any one of claims 1 to 8, the method comprises the following steps: Collect environmental data at preset periodic intervals through a multimodal sensor array and detect device status data in real time; Running the decision model optimized by federated learning to generate a control instruction set, wherein the control instruction set includes control commands for different types of control devices; After receiving the control instruction, the control device at the execution layer performs the response action according to the preset priority.

Images