Internet-of-things vegetation concrete self-adaptive maintenance monitoring method and system

Abstract

The invention discloses an internet-of-things vegetation concrete self-adaptive maintenance monitoring method and system, and belongs to the technical field of building materials and intelligent maintenance. Sensing multi-dimensional parameters in real time through a multi-modal sensing array; the local microcontroller processes, packages and transmits the data; according to a preset maintenance strategy, the local control unit adopts a multivariable decoupling control algorithm to generate adjustment instructions of moisture, pH value and nutritive salt; the micro peristaltic pump is driven by the micro-fluidic system, and liquid is precisely conveyed to the matrix through the biological activity slow release cavity; meanwhile, the system periodically encrypts the report state to the cloud platform and receives a remote instruction. And the cloud platform integrates a recurrent neural network model to realize prospective prediction maintenance. A traditional static structure is converted into a dynamic ecological complex, the problems that manual maintenance response lags behind, the resource utilization rate is low, and structural performance and ecological suitability are difficult to consider at the same time are effectively solved, and the plant survival rate and maintenance efficiency are remarkably improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of building materials and intelligent maintenance technology, and particularly relates to an Internet of Things (IoT) plant-growing concrete self-adaptive maintenance monitoring method and system. BACKGROUND

[0002] Under the macro background of accelerating urbanization and increasingly scarce land resources, the integration of building space and green ecology has become a key path to improve urban environmental quality and living experience. Especially under the promotion of the concepts of sponge city and green building, special concrete materials that can carry plant growth, i.e., plant-growing concrete, have attracted widespread attention from the industry because they can effectively realize the ecological coverage of building facades, roofs, and hard paving areas, and exhibit significant application value in vertical greening, rainwater management, and alleviating urban heat island effects.

[0003] The core design idea of the plant-growing concrete in the prior art mainly focuses on building a physical environment suitable for plant root growth and fixation. Specifically, this type of technical solution is usually achieved through two main approaches: one is the macro-porous concrete technology, i.e., by controlling the aggregate gradation and reducing the amount of fine aggregate, a large number of interconnected macro-pores are formed in the concrete to fill the nutrient soil and provide space for plant root growth; the other is the modular prefabricated lattice technology, i.e., prefabricated concrete blocks or panels with specific cavities or cells are manufactured, and after on-site assembly, soil is filled into the cavities for planting. These solutions successfully solved the basic problem of realizing vegetation coverage on traditional hard building surfaces in a specific historical period. By modifying the physical structure of concrete materials, they provided the possibility for plants to survive on non-natural substrates, constituting an important cornerstone for the development of this field. To maintain plant growth, regular maintenance by humans is usually required, including regular irrigation, seasonal fertilization, and pest control based on experience.

[0004] Therefore, how to break through the design framework of traditional plant-growing concrete as a passive physical carrier and build an intelligent system that can actively sense the multi-dimensional parameters of the internal micro-ecological environment and perform closed-loop feedback and self-adaptive adjustment based on real-time data, thereby transforming a static building structure into a dynamic ecological complex with life perception and active regulation capabilities, has become a key challenge and technical problem to be solved for those skilled in the art. SUMMARY

[0005] The present application aims to overcome the limitations inherent in the prior art of using bio- concrete as a passive physical carrier, which cannot actively sense and respond to internal micro-ecological dynamic changes. Specifically, the present application aims to solve the technical problems of inaccurate plant growth state control, delayed maintenance response, low water and fertilizer resource utilization efficiency, and difficulty in balancing structural performance and ecological suitability caused by the open-loop, low-frequency manual maintenance mode of the prior art. To achieve the above-mentioned objectives, the present application provides the following technical solutions:

[0006] According to the first aspect of the present application, the present application claims a kind of Internet of Things bio- concrete self-adaptive maintenance monitoring method, the method is based on the system formed by one or more bio- concrete unit bodies, wherein each bio- concrete unit body is embedded with a multifunctional integrated core module and interacts with a local control unit and a remote cloud platform, the method comprises the following steps:

[0007] S1, multi-dimensional parameter real-time sensing, through the multi-modal sensor array in the multifunctional integrated core module pre-embedded in the bio- concrete unit body, the multi-physical and chemical parameters of the nutrient soil substrate in the bio- concrete unit body are synchronously, in situ and continuously collected at a preset first time frequency;

[0008] S2, data local processing and protocol conversion, the microcontroller unit built-in the multifunctional integrated core module processes the original signal collected by the multi-modal sensor array, converts the original signal into standardized physical quantity data, and encapsulates the physical quantity data together with a unique unit body identification code and time stamp information into a data frame, and sends it to the local control unit through the controller area network bus interface at a preset second time frequency;

[0009] S3, closed-loop control decision based on multi-variable target interval, the local control unit receives and analyzes the data frame, compares the real-time state parameters with the preset maintenance strategy file stored internally, which contains multiple parameter target control intervals, and executes a multi-variable decoupling control algorithm to generate adjustment control instructions for moisture, pH or nutrient salt concentration;

[0010] S4, microfluidic execution, the local control unit converts the adjustment control instructions into driving signals to control the actions of the first to fourth micro- peristaltic pumps connected to the water tank, acid buffer tank, alkaline buffer tank and concentrated nutrient solution tank, respectively, to pump the corresponding liquid to the liquid distribution network inside the multifunctional integrated core module through the microfluidic pipeline;

[0011] S5, system state reporting and remote instruction response, the local control unit transmits the summary data packet containing sensor data, actuator state and control decision log to the remote cloud platform through the built-in fifth generation mobile communication module encryption at a preset third time frequency, and maintains a long connection with the remote cloud platform to receive and execute remote instructions from the remote cloud platform.

[0012] Further, the plant-growing concrete unit body is a prefabricated reinforced concrete modular component, the main structure of which is a reinforced concrete frame composed of high-strength grade cement, continuously graded aggregate and built-in steel mesh, and the frame forms a planting cavity inside for accommodating nutrient soil substrate and plant root systems;

[0013] The multifunctional integrated core module is fixed at a preset position in the mold before concrete pouring, so that after the concrete is solidified and formed, the module is permanently and integrally covered and fixed inside the plant-growing concrete unit body structure, and only a waterproof connector port for data and power transmission is exposed to the outer surface of the unit body.

[0014] The multifunctional integrated core module includes a three-dimensional grid-shaped support skeleton composed of non-metallic glass fiber reinforced composite rods, and each sensor node in the multi-modal sensor array is fixed to a designated three-dimensional coordinate node of the support skeleton to realize distributed fixed layout of the sensors in the nutrient soil substrate and obtain parameter gradient data of different depth profiles.

[0015] Further, each sensor node in the multi-modal sensor array is locally filled with epoxy resin before being installed to the support skeleton, and a porous shell made of sintered polyethylene material with a preset pore size is coated on the sensitive probe part of the sensor, which is used to provide mechanical protection and ensure sufficient moisture and ion exchange between the sensor probe and the surrounding nutrient soil substrate.

[0016] The multi-modal sensor array specifically includes:

[0017] The dielectric soil moisture sensor for measuring the water content of the substrate, the solid-state ion-selective electrode pH sensor for measuring the hydrogen ion concentration of the substrate, the four-electrode conductivity EC sensor for characterizing the total amount of soluble salt in the substrate, and the platinum resistance temperature sensor for measuring the temperature of the substrate and providing temperature compensation for the measurement results of the pH sensor and EC sensor.

[0018] Further, the liquid distribution network integrated in the multifunctional integrated core module is composed of a plurality of parallel micro-pipes made of polytetrafluoroethylene material, and the end of each micro-pipe is connected to one or more bioactive slow-release cavities, which are containers sintered from porous alumina ceramic material with a pre-set porosity and permeability coefficient. The liquid pumped by the micro-ceramic pump is injected into the cavity and slowly and uniformly permeates into the surrounding nutrient soil matrix through the porous wall;

[0019] The bioactive slow-release cavity is filled with a mixture of functional materials, which includes the following components:

[0020] Cross-linked polyacrylic acid sodium salt superabsorbent resin particles for absorbing and buffering water;

[0021] Clinoptilolite particles for adsorbing and slow-releasing ammonium ions and potassium ions in nutrient solution through ion exchange and buffering pH fluctuations;

[0022] Controlled-release compound fertilizer particles coated with biodegradable polymer film.

[0023] Further, the water balance regulation subprogram in the multivariable decoupling control algorithm in S3 adopts an incremental proportional-integral-derivative (PID) control algorithm to compare the real-time average soil moisture value with the water target control interval defined in the pre-set maintenance strategy file, calculate the control increment, and determine the water replenishment execution amount of the first micro-ceramic pump;

[0024] The proportional, integral, and derivative parameters of the PID control algorithm are remotely set by the remote cloud platform based on historical data analysis;

[0025] The acid-base dynamic balance subprogram in the multivariable decoupling control algorithm in S3 adopts a fuzzy logic-based control strategy, fuzzifies the input pH measurement value and pH change rate as input variables, performs fuzzy reasoning by querying the pre-set fuzzy rule base, and finally calculates the control output through defuzzification to determine the execution action and amount of the second or third micro-ceramic pump.

[0026] Further, the nutrient salt concentration management subprogram in the multivariable decoupling control algorithm in S3 is constrained by the water balance regulation subprogram and is configured to only generate a control instruction to start the fourth micro-ceramic pump to replenish nutrient solution when the water balance regulation subprogram determines that water replenishment is needed, and whether the real-time average conductivity value is lower than the lower limit of the conductivity target control interval is evaluated to decide whether to generate the control instruction;

[0027] The remote cloud platform is deployed with a recurrent neural network model based on a long short-term memory network, which takes historical time series data uploaded from the local control unit and external weather forecast data as input, to build a prediction model of the key parameters of the microenvironment inside the phytogenic concrete unit over time.

[0028] Further, the input data of the long short-term memory network model further specifically includes:

[0029] The time series data of all sensor readings and actuator action records uploaded from the local control unit, and the future weather forecast data corresponding to the geographical location of the phytogenic concrete unit obtained through the third-party application programming interface, the weather forecast data including air temperature, humidity, precipitation and light intensity for at least 24 hours in the future;

[0030] When the prediction result output by the long short-term memory network model indicates that at a certain time point in the future, a certain key parameter will have a high probability of deviating from the target control interval, the cloud platform automatically generates a prospective adjustment instruction set, and the instruction set is issued to the corresponding local control unit through the long connection to instruct the local control unit to perform preventive maintenance work in advance.

[0031] Further, each local control unit is matched with an independent energy module to provide continuous direct current power supply for it, and the energy module includes:

[0032] The photovoltaic panel using monocrystalline silicon technology, the solar charging controller integrated with the maximum power point tracking algorithm, and the battery pack using the lithium iron phosphate chemical system, the rated capacity of the battery pack ensures that the local control unit and all the devices driven by it can run continuously without interruption for at least a preset length of time under continuous no-light conditions.

[0033] Further, in S2, the processing performed by the microcontroller unit specifically includes:

[0034] The original analog signals collected by the multi-modal sensor array are subjected to analog-to-digital conversion, the converted digital signals are subjected to sliding average filtering to suppress noise, the real-time temperature data measured by the platinum resistance temperature sensor are used to perform temperature compensation correction on the measurement results of the conductivity and pH value, and the corrected physical quantity data, the unique unit body identification code and the time stamp information provided by the real-time clock module are encapsulated into a data frame conforming to the controller area network bus protocol.

[0035] According to the second aspect of the present application, the present application claims to protect an Internet of Things phytogenic concrete adaptive maintenance monitoring system, comprising:

[0036] One or more processors;

[0037] A memory storing one or more programs, which, when executed by one or more processors, enable the one or more processors to implement the IoT-based adaptive curing monitoring method for vegetated concrete.

[0038] This invention discloses an IoT-based adaptive curing monitoring method and system for vegetated concrete, belonging to the field of building materials and intelligent curing technology. It utilizes a multimodal sensor array to perceive multidimensional parameters in real time; a local microcontroller processes, encapsulates, and transmits the data; a local control unit generates adjustment instructions for moisture, pH, and nutrients based on a preset curing strategy using a multivariable decoupled control algorithm; a microfluidic system drives a micro-peristaltic pump to precisely deliver liquid to the substrate through a bioactive slow-release chamber; simultaneously, the system periodically and encryptedly reports its status to a cloud platform and receives remote instructions. The cloud platform integrates a recurrent neural network model to achieve proactive predictive maintenance. This invention transforms the traditional static structure into a dynamic ecological complex, effectively solving the problems of delayed response, low resource utilization, and difficulty in balancing structural performance and ecological suitability in manual curing, significantly improving plant survival rate and curing efficiency. Attached Figure Description

[0039] Figure 1 A flowchart illustrating the workflow of an IoT-based adaptive curing monitoring method for vegetated concrete, as claimed in an embodiment of the present invention.

[0040] Figure 2 This is a second flowchart of an IoT-based adaptive curing monitoring method for vegetated concrete, as claimed in an embodiment of the present invention.

[0041] Figure 3 This is a structural diagram of an IoT-based adaptive curing monitoring system for vegetated concrete, as claimed in an embodiment of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0043] The terms first, second, third, etc. are used only to describe various conditions, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Thus, the features defined with the first, second, third, etc. can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of a plurality is at least two, for example, two, three, etc., unless otherwise explicitly and specifically limited. All directional indications (such as upper, lower, left, right, front, back, etc.) in the embodiments of the present application are only used to explain the relative position relationship, movement condition, etc. between the components, and if the specific posture (as shown in the drawings) changes, the directional indications also change accordingly. In addition, the terms include and have and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to the process, method, product or device.

[0044] In this document, the term implementation means that the specific features, structures or characteristics described in conjunction with the implementation can be included in at least one implementation of the present application. The appearance of this phrase in various places in the specification does not necessarily mean the same implementation, nor is it an independent or alternative implementation that is not mutually exclusive with other implementations. Those skilled in the art explicitly and implicitly understand that the implementations described herein can be combined with other implementations.

[0045] With the continuous development of related technologies and the more stringent requirements of application scenarios on plant survival rate, landscape effect, durability and maintenance cost, etc., some inherent characteristics of the technical solutions of the existing technology at the principle level gradually show deep limitations in dealing with new challenges. The reason is that the existing green concrete system is essentially a static and passive physical base, while the plant-soil system it carries is a dynamic and multi-variable coupled complex micro-ecological environment. The healthy growth of plants depends on the maintenance of a precise dynamic balance between multiple key parameters such as water, nutrients, pH value, temperature, salt, etc., and there is strong mutual influence between these parameters. For example, excessive irrigation not only directly leads to root hypoxia, but also accelerates the leaching loss of nutrient elements and may cause a sharp fluctuation of soil pH value; and the change of environmental temperature will affect the water evaporation rate, thereby changing the concentration of soil solution and indirectly affecting the absorption efficiency of root system. The traditional green concrete structure itself does not have the ability to perceive the dynamic changes of these internal micro-environment, and its maintenance and management completely depends on external and low-frequency manual intervention. This open-loop management mode based on preset period and macroscopic observation cannot respond to the instantaneous needs of the internal ecological environment of the concrete in real time, and has significant hysteresis and uncertainty. When the maintenance personnel observe the macroscopic manifestations such as wilting and yellowing of plants, the internal ecological imbalance of the plant has often lasted for a period of time, which may have caused irreversible damage to the plant.

[0046] Further, the inherent contradiction between the static physical structure and the dynamic ecological demand leads to a series of technical bottlenecks. In order to improve the carrying capacity and durability of the structure, it is often necessary to increase the compactness of the concrete, which will directly compress the growth space of the plant root system and the air permeability and water permeability of the soil; on the contrary, if higher porosity is pursued for the benefit of plant growth, the mechanical properties of the material will be sacrificed, which limits its application in large-span and high-bearing scenarios. Similarly, the dense filling layer designed to prevent the loss of nutrient soil may hinder the normal exchange of water and gas. These mutually restrictive performance indicators make the design and maintenance of traditional green concrete fall into a dilemma of sacrificing one for the other, and it is difficult to achieve an ideal balance between structural safety, plant survival and long-term low maintenance cost.

[0047] According to the first embodiment of the present application, the present application claims a kind of Internet of Things green concrete self-adaptive maintenance monitoring method, the method is based on a system formed by one or more green concrete unit bodies, wherein each green concrete unit body is embedded with a multifunctional integrated core module, and carries out data and control instruction interaction with a local control unit and a remote cloud platform. Referring to Figure 1 , the method specifically includes the following steps:

[0048] S1, real-time multi-dimensional parameter perception, through the multi-modal sensor array embedded in the multi-functional integrated core module inside the vegetation concrete unit, the multi-physical and chemical parameters of the nutrient soil matrix in the vegetation concrete unit are synchronously, in-situ and continuously collected at a preset first time frequency;

[0049] S2, data local processing and protocol conversion, the microcontroller unit built-in the multi-functional integrated core module processes the original signals collected by the multi-modal sensor array, converts the original signals into standardized physical quantity data, encapsulates the physical quantity data together with the unique unit body identification code and time stamp information into a data frame, and sends it to the local control unit through the controller area network bus interface at a preset second time frequency;

[0050] S3, adopt closed-loop control decision based on multi-variable target interval, the local control unit receives and analyzes the data frame, compares the real-time state parameters with the preset maintenance strategy file stored internally, which contains multiple parameter target control intervals, and executes a multi-variable decoupling control algorithm to generate adjustment control instructions for moisture, pH or nutrient salt concentration;

[0051] S4, execute microfluidic, the local control unit converts the adjustment control instructions into driving signals to control the actions of the first to fourth micro peristaltic pumps connected to the water tank, the acidic buffer tank, the alkaline buffer tank and the concentrated nutrient liquid tank, respectively, and pumps the corresponding liquid to the liquid distribution network inside the multi-functional integrated core module through the microfluidic pipeline;

[0052] S5, state reporting and remote instruction response, the local control unit transmits the summary data package containing sensor data, actuator state and control decision log to the remote cloud platform through the built-in fifth generation mobile communication module at a preset third time frequency, and keeps a long connection with the remote cloud platform to receive and execute the remote instructions from the remote cloud platform.

[0053] Among them, specifically. The method further comprises the following steps in each step:

[0054] S1, Real-time multi-dimensional parameter sensing: A multi-modal sensor array embedded in a multi-functional integrated core module pre-embedded in the plant-growing concrete unit, synchronously, in-situ, continuously collects data of multiple physical and chemical parameters of the nutrient soil substrate in the unit at a preset first time frequency. The multi-modal sensor array includes at least: a dielectric soil moisture sensor for measuring the volumetric water content of the substrate, a solid-state ion-selective electrode pH sensor for measuring the hydrogen ion concentration of the substrate, a four-electrode conductivity EC sensor for characterizing the total amount of soluble salts in the substrate, and a platinum resistance temperature sensor for measuring the temperature of the substrate. The sensors in the sensor array are distributed and fixed in a three-dimensional space on the support skeleton of the multi-functional integrated core module to obtain parameter gradient data at different depth profiles.

[0055] S2, Data local processing and protocol conversion: A microcontroller unit in the multi-functional integrated core module polls the raw analog or digital signals of each sensor in the multi-modal sensor array at the first time frequency. The microcontroller unit performs analog-to-digital conversion, signal filtering, temperature compensation correction, and physical quantity calibration to convert the raw signals into standardized physical quantity data. Subsequently, the microcontroller unit encapsulates the processed parameter data, along with a unique unit identification code and time stamp information, into a standard data frame, and sends the data frame to the local control unit physically connected to the unit at a preset second time frequency through a controller area network CAN bus interface.

[0056] S3, Closed-loop control decision based on multi-variable target interval: The local control unit receives and analyzes the data frames from one or more multi-functional integrated core modules to obtain the real-time state parameters of each unit. The local control unit has a pre-set maintenance strategy file containing multiple parameter target control intervals for a specific plant stored internally. The target control interval defines the appropriate upper and lower threshold values of parameters such as moisture, pH, and conductivity. Referring to the target control interval, the local control unit executes a multi-variable decoupling control algorithm, which includes: Figure 2

[0057] Comparing the real-time average soil moisture value with the moisture target control interval. If it is lower than the lower threshold value, calculate an accurate water replenishment amount and generate a control instruction to turn on the first micro-ceramic pump; if it is higher than the upper threshold value, suppress the water replenishment instruction;

[0058] ​The real-time collected average pH value is compared with the pH target control interval. If below the lower threshold, a titration dose of acidic buffer is calculated and a control instruction to start the second micro-ceramic pump is generated; if above the upper threshold, a titration dose of alkaline buffer is calculated and a control instruction to start the third micro-ceramic pump is generated;

[0059] The real-time collected average conductivity value is compared with the conductivity target control interval. The conductivity value is evaluated synchronously only when the water balance adjustment subprogram determines the need for water replenishment. If the conductivity value is below the lower threshold, a replenishment dose of nutrient solution is calculated and a control instruction to start the fourth micro-ceramic pump is generated, which is executed in conjunction with the water replenishment instruction.

[0060] S4, microfluidic precision execution: the local control unit, through its drive circuit, converts the control instructions generated in S3 into pulse width modulation (PWM) signals for the drive motors connected to each micro-ceramic pump. The first, second, third, and fourth micro-ceramic pumps are connected to the water storage tank, acidic buffer tank, alkaline buffer tank, and concentrated nutrient solution tank, respectively. The pumps, according to the received PWM signals, pump the corresponding liquids through independent microfluidic pipelines to the liquid distribution network inside the multifunctional integrated core module with a flow rate accurate to the milliliter level, and uniformly and slowly penetrate into the plant root zone soil through the slow-release medium cavities at the end of the network.

[0061] S5, system state reporting and remote instruction response: the local control unit, at a preset third time frequency, transmits an aggregated data package containing all sensor data, actuator status, control algorithm decision logs, and device health status to the remote cloud platform server through a built-in fifth-generation mobile communication (5G) module, encrypted. At the same time, the local control unit maintains a long connection with the cloud platform and can receive and execute remote instructions from the cloud platform, including but not limited to updating maintenance strategy files, adjusting control algorithm parameters, remotely manually intervening actuators, or triggering firmware upgrade programs.

[0062] As a preferred embodiment of the present application, the plant-growing concrete unit body is a reinforced concrete modular component produced using a prefabrication process. Its main structure is a reinforced concrete frame with specific external dimensions, composed of high-strength grade Portland cement, continuously graded aggregates with controlled particle size distribution, and built-in cold-rolled ribbed steel mesh. Inside the frame is a planting cavity for accommodating nutrient soil substrate and plant root systems. The multifunctional integrated core module is accurately fixed at the preset position in the mold before concrete pouring, so that after the concrete is solidified and formed, the module is permanently and integrally covered and fixed inside the unit body structure, with only a waterproof connector port for data and power transmission exposed on the unit body surface.

[0063] Further, the structure of the multifunctional integrated core module is further refined. It contains a three-dimensional grid-shaped support skeleton composed of non-metallic, corrosion-resistant glass fiber reinforced composite material GFRP rod. Each sensor node in the multi-modal sensor array is firmly tied or clamped to the designated three-dimensional coordinate node of the support skeleton, ensuring its position stability and structural integrity during concrete pouring and vibrating. Each sensor node is locally filled with epoxy resin before installation, and is covered with a sintered polyethylene porous shell with a specific pore size. The shell provides mechanical protection while ensuring sufficient water and ion exchange between the sensor probe and the surrounding soil matrix.

[0064] As another preferred embodiment of the present application, the liquid distribution network integrated inside the multifunctional integrated core module is composed of multiple parallel polytetrafluoroethylene PTFE micro-pipes with stable chemical properties, each corresponding to a liquid pumped by a local control unit. The ends of the pipes are not directly opened in the soil, but are connected to one or more bioactive slow-release cavities. The bioactive slow-release cavity is a cylindrical container made of porous alumina ceramic material with a pre-set porosity and permeability coefficient. It is filled with a mixture of three functional materials: the first is cross-linked polyacrylic acid sodium salt superabsorbent resin SAP particles, used to absorb and buffer water, achieving slow release of water; the second is clinoptilolite particles with high cation exchange capacity, used to adsorb and slowly release ammonium and potassium ions in the nutrient solution, and to buffer the sharp fluctuations in pH value; the third is controlled-release compound fertilizer particles coated with biodegradable polymer film. The liquid pumped by the micro-ceramic pump is first injected into the cavity, and after fully interacting with the mixed functional materials inside, it slowly and uniformly penetrates into the surrounding nutrient soil matrix through the porous ceramic wall, thereby avoiding the stress damage to the plant roots caused by the local high concentration of liquid.

[0065] Further, the control algorithm executed by the local control unit is constructed as a modular multitasking real-time operating system in software implementation. The water balance regulation subprogram adopts an incremental proportional-integral-derivative (PID) control algorithm, the proportional (Kp), integral (Ki) and derivative (Kd) parameters of which can be remotely set by the cloud platform according to historical data analysis to adapt to the dynamic characteristics of water demand in different seasons and plant growth stages. The pH dynamic balance subprogram adopts a control strategy based on fuzzy logic. The input pH measurement and pH change rate are fuzzified, and the pre-set fuzzy rule base is queried, for example, if the pH value is'strong acidity' and the change rate is 'zero', the alkaline buffer pump output is 'large', fuzzy reasoning is performed, and finally the accurate pump execution time is calculated by defuzzification. Compared with the traditional PID control, the fuzzy logic control strategy shows stronger robustness and smoother regulation effect when dealing with the controlled object with strong nonlinearity and large lag characteristics such as pH.

[0066] As another preferred embodiment of the application, the functions of the cloud platform are further expanded, and it not only serves as a terminal for data storage and remote monitoring, but also undertakes the tasks of advanced data analysis and model training. The cloud platform deploys a long short-term memory (LSTM) based recurrent neural network model for constructing a prediction model of the microenvironment of the vegetation concrete. The model takes the historical time series data uploaded by a single local control unit, including all sensor readings and actuator action records, and the future 72-hour weather forecast data including temperature, humidity, precipitation and light intensity of the geographical location of the corresponding unit body obtained through a third-party API interface as model inputs. The LSTM model can output prediction curves of key parameters such as soil moisture and temperature in the next 24 hours. When the model predicts that a certain parameter will deviate from the target control interval, the cloud platform will automatically generate a prospective adjustment instruction set and issue it to the corresponding local control unit. According to the instruction, the local control unit performs preventive watering operation the night before the next day predicted to be a high-temperature hot weather, thereby upgrading the traditional passive closed-loop feedback control to a predictive and prospective active maintenance based on data driving.

[0067] Further, in order to guarantee the energy self-sufficiency of the whole system, each local control unit is equipped with an independent energy module. The energy module includes: a solar photovoltaic panel using monocrystalline silicon technology, with a photoelectric conversion efficiency of not less than 22%; a solar charging controller integrated with a maximum power point tracking (MPPT) algorithm; and a battery pack using a lithium iron phosphate (LiFePO4) chemical system, with a rated capacity of not less than 40 ampere-hours (Ah). The photovoltaic panel is installed in the best light condition of the vegetated concrete application scene, and the generated electric energy is efficiently charged to the battery pack through the MPPT controller, and provides stable and continuous direct current power for the local control unit and all sensors and actuators driven thereby. The capacity of the battery pack is accurately calculated to ensure that the whole system can still run uninterruptedly for at least 72 hours under continuous no-light conditions, thereby guaranteeing the extreme reliability of the monitoring and maintenance functions.

[0068] The core technical idea of the present application is to deeply integrate advanced sensing technology, microfluidic execution technology, embedded control algorithm, and cloud intelligent computing into the traditional vegetated concrete structure, to build a closed-loop intelligent ecological system capable of self-sensing, self-decision, and self-regulation. The system is composed of a plurality of vegetated concrete unit bodies, one or more local control units matched therewith, a remote cloud platform, and an energy module providing continuous running capability for the system. In the following, the various physical entities constituting the system, the information processing flow, and the cooperative working mechanism therebetween will be described in detail in engineering, to fully disclose the implementation mode of the present application.

[0069] The technical solution of the present application is an Internet of Things vegetated concrete self-adaptive maintenance monitoring method, and the execution subject thereof is a highly integrated system. The basic physical carrier of the system is a vegetated concrete unit body. As a preferred embodiment of the present application, the vegetated concrete unit body is produced in a factory environment using a standardized prefabrication process, to ensure the high consistency of quality and the convenience of on-site installation. The main structure thereof is a reinforced concrete frame with a specific external contour and internal volume, for example, a rectangular module with a size of 1000mm × 500mm × 400mm. The concrete base body of the frame uses high-strength Portland cement with a label of not less than C40, matched with basalt aggregate and river sand that have been strictly sieved and have continuous gradation, and the water-cement ratio is controlled between 0.35 and 0.40, to ensure the mechanical strength and durability of the building component. Inside the frame, a double-layer steel mesh welded by HRB400-grade cold-rolled ribbed steel bars with a diameter of 8mm is pre-installed, to provide excellent bending and shear resistance for the structure. The interior of the frame is designed as a planting cavity with a volume of about 150 liters, which is used to fill a scientifically proportioned nutrient soil substrate and provides physical space for the growth of plant roots.

[0070] A key technical feature of the present invention is that each phytogenic concrete unit is permanently and integrally embedded with a multifunctional integrated core module. The embedding process of the module is completed before the concrete casting. Specifically, after the steel reinforcement mesh is placed into the mold, the multifunctional integrated core module is precisely fixed at the pre-set three-dimensional coordinate position within the mold by means of a specially designed positioning fixture. The selection of this position is based on the main distribution area of plant root systems, usually at a depth of 10-30 cm from the surface, and the structural stability requirement to ensure that the module does not displace during the concrete vibrating process. After the fixation of the module is completed, the concrete casting and vibrating operations are carried out. After the concrete solidification and molding, the multifunctional integrated core module is permanently covered and anchored inside the structure of the unit, becoming an integral part of it. To realize the data and power connection with external equipment, the module only has one multi-core waterproof aviation plug connector port exposed on the outer surface of the unit, which meets the IP68 protection level.

[0071] Further, the internal structure of the multifunctional integrated core module is described in detail. It contains a three-dimensional grid-shaped support skeleton composed of glass fiber reinforced composite material (GFRP) rods as structural support. The reason for choosing GFRP material is its excellent corrosion resistance, high strength-to-weight ratio, and non-shielding properties to electromagnetic signals. The size of the skeleton is about 200mm × 150mm × 150mm, and it is distributed with multiple precisely designed nodes for mounting a multi-modal sensor array. The sensor array is responsible for in-situ and continuous monitoring of the key physical and chemical parameters of the nutrient soil substrate inside the phytogenic concrete unit.

[0072] In one specific embodiment, the multi-modal sensor array contains at least four different types of sensors, which are fixed on different nodes of the GFRP support skeleton in a distributed manner to obtain multi-point data that can reflect the gradient of soil profile parameters. Specifically, the array includes: a dielectric soil moisture sensor for measuring the matrix volumetric water content, which works on the principle of frequency domain reflectometry (FDR) and inverses the water content by measuring the change of soil dielectric constant around the probe, the probe part of which is composed of FR-4 substrate covered with an epoxy coating, the working frequency is 100 MHz, and the measurement accuracy can reach ±2% VWC. The array also includes a solid-state ion-selective electrode pH sensor for measuring the matrix hydrogen ion concentration, which uses a solid-state electrode based on a ruthenium dioxide (RuO2) sensitive film and integrates an Ag / AgCl solid-state reference electrode, which has better mechanical strength and maintenance-free characteristics compared to traditional glass electrodes, with a measurement range of pH 2 to 12 and an accuracy of ±0.1 pH. At the same time, the array integrates a four-electrode conductivity (EC) sensor for characterizing the total amount of soluble salts in the matrix, whose probe is composed of four platinum black-coated platinum electrodes with fixed spacing, and the four-electrode structure effectively overcomes the measurement error caused by electrode polarization effect. Finally, a Class A precision PT1000 platinum resistance temperature sensor for measuring the matrix temperature is integrated, which not only provides soil temperature data but also provides real-time temperature compensation for the measurement results of the pH and EC sensors. Before installation to the skeleton, the circuit part of each sensor node is treated with vacuum epoxy resin filling, and the sensitive probe part is protected by a porous shell made of sintered high-density polyethylene (HDPE) material with an average pore size of 80 microns. The shell not only provides sufficient mechanical protection but also ensures smooth water and ion exchange channels between the sensor probe and the surrounding soil matrix.

[0073] Corresponding to the sensing function, the multifunctional integrated core module also integrates a liquid distribution network for precise liquid delivery inside. The network is composed of multiple parallel polytetrafluoroethylene (PTFE) micro-pipes with an inner diameter of 1.0 mm and an outer diameter of 2.0 mm. PTFE material is selected due to its excellent chemical inertness to ensure that no reaction or degradation occurs when delivering acid, base or nutrient solution. An innovation is that the ends of these pipes are not directly opened in the soil, but are connected to one or more bioactive slow-release cavities. The cavity is a cylindrical container sintered from α-alumina ceramic powder at 1600°C, with a controllable porosity of 38% and a specific permeability coefficient. Its external dimensions are 35 mm in diameter and 60 mm in height. Its interior is filled with a precisely proportioned mixture of three functional materials: the first is cross-linked polyacrylic acid sodium salt superabsorbent polymer (SAP) particles with a water absorption ratio of up to 350 times their own weight, used to absorb and buffer water, achieving peak reduction and slow release of water; the second is natural clinoptilolite particles with a particle size of 200-300 mesh and a cation exchange capacity (CEC) of up to 1.8 meq / g, used to selectively adsorb and slowly release ammonium ions (NH4+) and potassium ions (K+) in the nutrient solution, and to buffer the sharp fluctuations in soil pH; the third is controlled-release compound fertilizer particles coated with biodegradable polylactic acid (PLA) film, with a nitrogen-phosphorus-potassium ratio of 20-10-15 and a nutrient release period designed for 90 days. The various liquids pumped from the outside are first injected into the cavity, where they undergo physical and chemical interactions with the internal mixed functional materials, and then penetrate into the surrounding nutrient soil matrix through the porous ceramic wall in a capillary manner. This design greatly avoids the stress damage to plant roots caused by excessive local liquid concentration or transient water saturation, and achieves the transition from irrigation to nourishment.

[0074] The execution flow of the method of the present application begins at S1, real-time perception of multi-dimensional parameters. The multifunctional integrated core module deployed inside each plant-growing concrete unit body performs synchronous data acquisition on all sensors in the multi-modal sensor array at a preset first time frequency, for example, T1 = 10 seconds. This means that every 10 seconds, the system can obtain a set of instantaneous state snapshots reflecting different depth profiles, including soil moisture, pH, EC and temperature.

[0075] S2 is the local processing and protocol conversion of data. This step is executed on a low-power microcontroller unit (MCU) built-in the multifunctional integrated core module. The MCU polls the raw analog or digital signals of each sensor at the same first time frequency as data acquisition through its internal 12-bit analog-to-digital converter (ADC) or serial peripheral interface (SPI / I2C). The acquired raw signals are first processed by a 5-window-width moving average filter to effectively suppress high-frequency noise. Subsequently, the MCU performs a series of correction and calibration calculations. For example, the real-time temperature T measured by the platinum resistance temperature sensor is used to temperature-compensate the EC sensor reading EC_T, and calculate the conductivity value EC_25 at the standard temperature of 25°C, with the formula: EC_25 = EC_T / [1 + a(T-25)], where a is the temperature compensation coefficient of the solution, usually taking the value of 0.0191. After completing all physical quantity calculations, the MCU encapsulates the normalized water content, pH, EC, and temperature data, together with a globally unique 64-bit unit body identification code (UID) burned in at the factory and a timestamp provided by the real-time clock (RTC) module, into an extended data frame conforming to the CAN 2.0B protocol standard. The data frame is sent to the local control unit physically connected to the unit body through a controller area network (CAN) bus transceiver at a preset second time frequency, e.g., T2 = 60 seconds. The purpose of setting T2 greater than T1 is to preliminarily aggregate and smooth the data while ensuring data freshness, and to reduce the load of the CAN bus.

[0076] S3 is the core of the entire closed-loop control, i.e., the closed-loop control decision based on the multivariate target interval, which is executed in the local control unit. The local control unit is a powerful embedded system with an industrial-grade microprocessor as its hardware core, equipped with sufficient RAM and Flash storage space, and running a real-time operating system (RTOS) that supports multi-task scheduling. The local control unit continuously listens to and receives data frames sent by all multifunctional integrated core modules within its jurisdiction through its CAN bus interface. After receiving the data frame, it first parses and stores the real-time state parameters of each unit body into the real-time database in memory. In the non-volatile memory of the local control unit, one or more care strategy files for specific plants such as Sedum lineare and Euphorbia cotinifolia are pre-stored. The file is usually in JSON format, and its structure clearly defines the ideal target control interval of each environmental parameter corresponding to different growth stages of the plant, such as the planting period, rapid growth period, and mature period. For example, the strategy for the rapid growth period of Sedum lineare may be defined as: the soil volume water content target interval is [28%, 38%], the pH target interval is [6.2, 7.0], and the EC target interval is [1.5 mS / cm, 2.2 mS / cm].

[0077] The local control unit periodically executes a sophisticated multivariable decoupling control algorithm, which is designed as three sub-programs that work independently but cooperatively under certain conditions. The first one is the water balance regulation sub-program, which compares the average soil moisture value obtained from the sensor with the water target control interval defined in the strategy file. As a preferred embodiment, this sub-program adopts a proportional-integral-derivative (PID) control algorithm. The error term e(t) is defined as the difference between the target water value, for example, the interval midpoint 33%, and the current measured value. The PID controller calculates a control increment Au(t) according to the current error e(t), the cumulative integral of the error, and the rate of change of the error, which is ultimately converted into the opening duration or the number of driving pulses of the first micro-ceramic pump. The proportional Kp, integral Ki, and derivative Kd parameters of the PID algorithm are not fixed but can be remotely online tuned by the remote cloud platform based on long-term analysis of historical data to dynamically adapt to changes in water demand characteristics due to different seasonal climates and plant growth stages.

[0078] The second one is the dynamic pH balance sub-program, which compares the real-time average pH value with the pH target control interval. Given the strong nonlinearity and large time delay characteristics of the soil pH regulation process, this sub-program preferably adopts a fuzzy logic-based control strategy. This strategy fuzzifies two input variables, the pH deviation, which is the difference between the current pH and the midpoint of the target interval, and the pH change rate, which is the difference between the current pH and the pH of the previous period. These two variables are mapped to pre-defined fuzzy sets, each defined by multiple membership functions with linguistic meanings. Then, a fuzzy inference engine infers according to a pre-defined set of expert rule base. Finally, through defuzzification calculation, an accurate control output is obtained, which determines whether to start the second micro-ceramic pump or the third micro-ceramic pump, and the respective working duration. This fuzzy control strategy, compared to traditional PID, exhibits stronger robustness and smoother regulation effect in dealing with complex objects such as pH, effectively avoiding overshoot and oscillation.

[0079] The last one is the nutrient salt concentration management sub-program, which compares the real-time average EC value with the conductivity target control interval. An important logical constraint is that this sub-program is only activated when the water balance regulation sub-program determines that water needs to be replenished. This design is to prevent the application of concentrated nutrient solution when the soil is dry, which can cause a sharp increase in soil salt concentration and harm the plant roots. When water needs to be replenished and the EC value is below the lower limit of the target interval, this sub-program calculates a nutrient solution replenishment dose and generates a control instruction to start the fourth micro-ceramic pump. This instruction will be executed in cooperation with the water replenishment instruction of the first micro-ceramic pump, and the concentrated nutrient solution will be pre-mixed with clean water in the pipeline before being pumped together to the target unit.

[0080] S4 is microfluidic precision execution. The local control unit converts the control decisions it generates in S3 into pulse width modulation (PWM) signals for the stepper motor drivers or DC motor drive circuits connected to each micro-creep pump through its internal drive circuit. The first micro-creep pump is connected to a water storage tank, the second micro-creep pump is connected to an acidic buffer solution tank containing, for example, 0.01 mol / L citric acid solution, the third micro-creep pump is connected to an alkaline buffer solution tank containing, for example, 0.01 mol / L sodium bicarbonate solution, and the fourth micro-creep pump is connected to a concentrated nutrient solution tank containing high-concentration, full-element water-soluble fertilizer. These micro-creep pumps accurately rotate according to the received PWM signals to accurately pump the corresponding liquid through independent microfluidic pipelines into the bioactive slow-release cavity inside the designated vegetated concrete unit, completing the precise maintenance work.

[0081] S5 is system state reporting and remote instruction response. The local control unit transmits a summary data packet containing all sensor data, status logs of all actuator creep pumps, decision records of the control algorithm, and device health status such as power supply voltage and internal temperature within its jurisdiction to the remote cloud platform server at a pre-set third time frequency, for example, T3 = 15 minutes, through a built-in fifth-generation mobile communication module supporting NB-IoT and 4G / 5G dual-mode, using end-to-end encryption based on TLS 1.3 protocol. At the same time, the local control unit maintains a long connection based on the MQTT protocol with the cloud platform, allowing it to receive and execute remote instructions from the cloud platform in near real time. These instructions are diverse, including but not limited to remotely updating the locally stored maintenance strategy file, online adjusting the key parameters of the PID or fuzzy control algorithm, initiating forced manual intervention of any creep pump by the remote administrator, or triggering the firmware over-the-air (FOTA) program for online upgrade of the entire local control unit.

[0082] As a further preferred embodiment of the present application, the functions of the remote cloud platform are further extended, so that it is not only a passive data storage and remote monitoring terminal, but also an active and forward-looking intelligent decision center. A recurrent neural network model based on long short-term memory network (LSTM) is deployed on the cloud platform, which is specially used to build a time series prediction model for the micro environment of the vegetated concrete. The model takes the historical time series data uploaded by the single local control unit for months, including all sensor readings and actuator action records, as the basic training data, and dynamically fuses the future 72-hour high-resolution weather forecast data for the geographical location of the unit body, including hourly temperature, relative humidity, precipitation, total cloud cover, wind speed, and total solar radiation, etc., obtained through the third-party commercial weather service API interface, as the real-time input of the model. The trained LSTM model can output the prediction curve of the key parameters such as soil moisture and temperature within the next 24 hours with high accuracy. When the model predicts that, for example, after 12 hours, due to continuous high temperature and strong sunlight, the soil moisture of a certain unit body will have more than 90% probability to drop below the lower limit of its target control interval, the cloud platform will no longer wait for passive feedback from the local, but will generate a forward-looking adjustment instruction set. The instruction set may contain specific instructions such as performing 1500 milliliter of preventive watering operation on unit body No. XYZ within the next 3 hours. The instruction is issued to the corresponding local control unit through the MQTT long connection. The local control unit performs maintenance work in advance according to the instruction, thereby successfully upgrading the traditional passive closed-loop feedback control based on after-the-fact deviation to a data-driven and proactive maintenance mode.

[0083] Further, to ensure the long-term stable operation of the whole system in various complex outdoor environments, each local control unit is equipped with an independent energy module. The core of the energy module is a solar photovoltaic panel made of A-grade monocrystalline silicon battery pieces, with a photoelectric conversion efficiency of not less than 22.5% and a rated power of 50Wp. The panel charges the battery pack through a solar charging controller integrated with a maximum power point tracking (MPPT) algorithm. The MPPT algorithm can track the maximum power output point of the photovoltaic panel in real time, ensuring that the charging efficiency is always maintained at more than 98% under different light and temperature conditions. The energy storage part uses a battery pack composed of high-safety lithium iron phosphate (LiFePO4) cells, with a rated voltage of 12.8V and a rated capacity of 40Ah, and the total energy storage capacity reaches 512Wh. The lithium iron phosphate battery is selected because it has a long life of more than 3000 cycles, excellent thermal stability, and a wider operating temperature range. After precise energy consumption budget analysis, the average standby power consumption of the entire local control unit and its attached sensors and actuators is about 1.5W, and the peak power consumption of a single maintenance operation such as pump operation for 5 minutes is about 8W. Comprehensive calculation shows that the total energy consumption of the system per day is about 40Wh. Therefore, the 512Wh battery capacity can ensure the uninterrupted normal operation of the entire monitoring and maintenance system for more than 10 days after being fully charged, even in extreme rainy weather conditions without any effective light, far exceeding the minimum 72-hour endurance standard, thereby providing a solid energy guarantee for the extreme reliability of the system.

[0084] According to the second embodiment of the present application, with reference to Figure 3 The present application claims a kind of Internet of Things plant-growing concrete adaptive maintenance monitoring system, comprising:

[0085] One or more processors;

[0086] Memory, one or more programs are stored on it, when the one or more programs are executed by the one or more processors, so that the one or more processors realize the one kind of Internet of Things plant-growing concrete adaptive maintenance monitoring method

[0087] To further verify the technical effect of the present application, we set up examples and comparative examples for comparison test.

[0088] Example:

[0089] In a vertical green wall project located on the south facade of a commercial building, 100 IoT-enabled self-adaptive maintenance monitoring units of the invented living concrete are deployed. The plants planted are Sedum lineare, which is highly adaptable to the local environment. The system is configured in the manner described above. The maintenance strategy file in the local control unit sets the target soil moisture range for Sedum lineare growth period as [25%, 35%], the target pH range as [6.0, 7.0], and the target EC range as [1.2, 1.8] mS / cm. The LSTM prediction model of the remote cloud platform is connected to the local weather service API. The test period is from June 1 to August 31, a total of 92 days, covering the typical summer high-temperature plum rain season in Shanghai.

[0090] Comparative Example:

[0091] On the same facade of the same building, adjacent to the example area, 100 traditional living concrete units with the same structure size, substrate formula, and planted plants are deployed. These units do not integrate any sensing and automatic execution devices. Their maintenance method adopts the industry standard timed manual maintenance mode, which is inspected and sprayed by professional green maintenance personnel twice a week, and liquid fertilizer is applied every half month.

[0092] After the 92-day test period, the key performance indicators of the two areas are statistically analyzed. The example using the invented method is significantly better than the comparative example using the traditional maintenance method in all indicators. The significant increase in plant survival rate and coverage rate directly proves the positive effect of the fine and adaptive growth environment provided by the invention on plant health. The total water consumption and total fertilizer consumption are reduced by 60.7% and 65.7% respectively, fully reflecting the great advantage of the invention in precise supply on demand and eliminating resource waste. The number of manual maintenance is reduced from 30 to 2, which directly reflects the great potential of the invention in reducing long-term operation and maintenance costs. The record of zero failure and zero stress in the example is in sharp contrast to the 4 serious drought stress events in the comparative example, highlighting the outstanding reliability of the invented system, especially its forward-looking maintenance capability based on data prediction, in dealing with extreme weather and ensuring plant safety.

[0093] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are merely illustrative, for example, the division of units is only a logical function division, and actual implementation can have another division manner, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0094] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit. The above is only an embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation using the content of the present application specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

[0095] The specific embodiments of the application are described in detail above, but they are only examples. The present application is not limited to the specific embodiments described above. Any equivalent modification or substitution made by those skilled in the art to the present application is also within the scope of the present application, and therefore, any equivalent transformation, modification, improvement, etc. made without departing from the spirit and principle range of the present application should be included in the scope of the present application.

Claims

1. A method for adaptive maintenance monitoring of Internet of Things (IoT) green concrete, the method being based on a system composed of one or more green concrete units, wherein each of the green concrete units is embedded with a multifunctional integrated core module and interacts with a local control unit and a remote cloud platform, characterized in that, The method comprises the following steps: S1, real-time multi-dimensional parameter sensing, through a multi-modal sensor array pre-embedded in the multifunctional integrated core module inside the phytogenic concrete unit body, to synchronously, in-situ and continuously collect data of multiple physical and chemical parameters of the nutrient soil substrate in the phytogenic concrete unit body at a preset first time frequency; S2, data local processing and protocol conversion, by a microcontroller unit built-in the multifunctional integrated core module, to process the original signals collected by the multi-modal sensor array, convert the original signals into standardized physical quantity data, encapsulate the physical quantity data together with a unique unit body identification code and time stamp information into a data frame, and send the data frame to the local control unit through a controller area network bus interface at a preset second time frequency; S3, closed-loop control decision based on multi-variable target interval, the local control unit receives and analyzes the data frame, compares the real-time state parameters with a preset maintenance strategy file stored internally and containing multiple parameter target control intervals, and executes a multi-variable decoupling control algorithm to generate adjustment control instructions for moisture, pH or nutrient salt concentration; S4, microfluidic execution, the local control unit converts the adjustment control instructions into driving signals to control the actions of the first to fourth micropump connected with the water storage tank, the acidic buffer solution tank, the alkaline buffer solution tank and the concentrated nutrient solution tank respectively, and pumps the corresponding liquid to the liquid distribution network inside the multifunctional integrated core module through the microfluidic pipeline; S5, system state reporting and remote instruction response, the local control unit transmits the summary data package containing sensor data, actuator state and control decision log to the remote cloud platform through the built-in fifth generation mobile communication module at a preset third time frequency, and maintains a long connection with the remote cloud platform to receive and execute remote instructions from the remote cloud platform. 2.The Internet of Things (IoT) vegetation-growing concrete self-adaptive maintenance monitoring method according to claim 1, wherein, The phytogenic concrete unit body is a prefabricated reinforced concrete modular component, and the main structure thereof is a reinforced concrete frame composed of high-strength grade cement, continuously graded aggregate and built-in steel mesh, and a planting cavity for accommodating nutrient soil substrate and plant root system is formed inside the frame; The multifunctional integrated core module is fixed at a preset position in the mold before concrete pouring, so that after the concrete is cured and formed, the module is permanently integrated and fixed inside the phytogenic concrete unit body structure, and only a waterproof connector port for data and power transmission is exposed on the outer surface of the unit body; The multifunctional integrated core module comprises a three-dimensional grid-shaped support skeleton composed of non-metallic glass fiber reinforced composite rods, and each sensor node in the multi-modal sensor array is fixed on a designated three-dimensional coordinate node of the support skeleton to realize distributed fixed layout of the sensors in the nutrient soil substrate and obtain parameter gradient data of different depth profiles. 3.The Internet of Things (IoT) vegetation-growing concrete self-adaptive maintenance monitoring method of claim 2, wherein, Each sensor node in the multi-modal sensor array has its circuit part locally encapsulated by epoxy resin before being mounted to the support skeleton, and its sensitive probe part is covered by a porous shell made of sintered polyethylene material with a preset pore size, which is used to provide mechanical protection and ensure sufficient moisture and ion exchange between the sensor probe and the surrounding nutrient soil matrix; The multi-modal sensor array specifically includes: a dielectric soil moisture sensor for measuring the volumetric water content of the matrix, a solid-state ion-selective electrode pH sensor for measuring the hydrogen ion concentration of the matrix, a four-electrode conductivity EC sensor for characterizing the total amount of soluble salts in the matrix, and a platinum resistance temperature sensor for measuring the temperature of the matrix and providing temperature compensation for the measurement results of the pH sensor and the EC sensor. 4.The Internet of Things (IoT) vegetation-growing concrete self-adaptive maintenance monitoring method of claim 1, wherein, The liquid distribution network integrated inside the multifunctional integrated core module is composed of multiple parallel micro-pipes made of polytetrafluoroethylene material, and the end of each micro-pipe is connected to one or more bioactive slow-release cavities, which are containers sintered from porous alumina ceramic material with a preset porosity and permeability coefficient. The liquid pumped by the micro-ceramic pump is injected into the cavity and slowly and uniformly permeates into the surrounding nutrient soil matrix through its porous wall; The bioactive slow-release cavity is filled with a mixture of functional materials, which includes the following components: cross-linked polyacrylic acid sodium salt superabsorbent resin particles for absorbing and buffering water; clinoptilolite particles for adsorbing and releasing ammonium and potassium ions in the nutrient solution through ion exchange and buffering pH fluctuations; controlled-release compound fertilizer particles coated with a biodegradable polymer film. 5.The Internet of Things (IoT) vegetation-growing concrete self-adaptive maintenance monitoring method of claim 1, wherein, The moisture balance adjustment subroutine in the multi-variable decoupling control algorithm in S3 uses an incremental proportional-integral-derivative (PID) control algorithm to compare the real-time average soil moisture value with the moisture target control interval defined in the preset maintenance strategy file, calculate the control increment, and determine the water replenishment execution amount of the first micro-ceramic pump; The proportional, integral, and derivative parameters of the PID control algorithm are remotely set by the remote cloud platform based on historical data analysis; The acid-base dynamic balance subroutine in the multi-variable decoupling control algorithm in S3 uses a fuzzy logic-based control strategy, which fuzzifies the input pH measurement value and pH change rate as input variables, queries the preset fuzzy rule base for fuzzy reasoning, and finally calculates the control output through defuzzification to determine the execution action and amount of the second or third micro-ceramic pump. 6.The Internet of Things (IoT) vegetation-growing concrete self-adaptive maintenance monitoring method of claim 1, wherein, The nutrient concentration management subroutine in the multivariate decoupling control algorithm in S3 is constrained by the water balance regulation subroutine. It is configured to only evaluate whether the real-time collected average conductivity value is lower than the lower limit of the conductivity target control range when the water balance regulation subroutine determines that water needs to be added, and decide whether to generate a control command to start the fourth micro-peristaltic pump to replenish the nutrient solution. The remote cloud platform deploys a recurrent neural network model based on a long short-term memory network. This model takes historical time-series data uploaded from the local control unit and external weather forecast data as input to construct a predictive model of the changes in key parameters of the internal microenvironment of the vegetated concrete unit over time.

7. The Internet of Things (IoT) plant-growing concrete self-adaptive maintenance monitoring method of claim 11, wherein, The input data for the Long Short-Term Memory (LSTM) network model further includes: The data includes time-series data of all sensor readings and actuator action records uploaded from the local control unit, as well as future weather forecast data corresponding to the geographical location of the vegetated concrete unit obtained through a third-party application programming interface. The weather forecast data includes temperature, humidity, precipitation and light intensity for at least the next 24 hours. When the prediction results output by the Long Short-Term Memory Network model indicate that a certain key parameter will have a high probability of deviating from the target control range at a certain point in the future, the cloud platform will proactively generate a forward-looking adjustment instruction set and send the instruction set to the corresponding local control unit through the long connection, so as to instruct the local control unit to perform preventive maintenance operations in advance. 8.The Internet of Things (IoT) vegetation-growing concrete self-adaptive maintenance monitoring method of claim 1, wherein, Each of the local control units is equipped with an independent power module to provide it with continuous DC power. The power module includes: The system comprises a photovoltaic panel using monocrystalline silicon technology, a solar charge controller integrating a maximum power point tracking algorithm, and a battery pack using a lithium iron phosphate chemistry system. The rated capacity of the battery pack ensures that the local control unit and all its driven devices can operate uninterrupted for at least a preset time under continuous no-light conditions. 9.The Internet of Things (IoT) vegetation-growing concrete self-adaptive maintenance monitoring method of claim 1, wherein, In step S2, the processing performed by the microcontroller unit specifically includes: The original analog signal acquired by the multimodal sensor array is converted from analog to digital. The converted digital signal is then filtered by moving average to suppress noise. The real-time temperature data measured by the platinum resistance temperature sensor is used to perform temperature compensation correction on the conductivity and pH measurement results. Finally, the corrected physical quantity data, along with the unique unit cell identification code and the timestamp information provided by the real-time clock module, are encapsulated into a data frame conforming to the controller area network bus protocol.

10. An Internet of Things (IoT) vegetation concrete adaptive maintenance monitoring system, comprising: include: One or more processors; A memory having stored one or more programs that, when executed by one or more processors, cause the one or more processors to implement an Internet of Things (IoT) adaptive curing monitoring method for vegetated concrete according to any one of claims 1 to 9.

Images