Intelligent ecological warning system

Abstract

The invention discloses an intelligent ecological warning system, and belongs to the field of ecology, biotechnology and landscape design. The system comprises an ecological warning wall, a biosensing layer and an intelligent regulation and control system. According to the system disclosed by the invention, ppt-level pollutant detection sensitivity is realized through the graphene electrode array and the microfluidic photobacterium chip, and response time is shortened; by combining bionic vascular bundle irrigation and solar power supply, water and electricity consumption and operation and maintenance cost are reduced, and the maintenance period is prolonged; a modularized 3D printing structure and LoRa wireless transmission are adopted, so that manual intervention is reduced; meanwhile, CRISPR customized strains and fractal algorithm plant layout are integrated, so that the system has the functions of pollution monitoring, carbon sink appreciation and dynamic landscape interaction, and the precision, efficiency, sustainability and ecological benefits are comprehensively superior to those of a traditional technology.

Description

Technical Field

[0001] This invention belongs to the fields of ecology, biotechnology and landscape design, and specifically relates to an intelligent ecological warning system. Background Technology

[0002] As people's living standards improve in cities, cities are paying more and more attention to the construction of ecological warning walls, thereby adding a layer of protection to the city's ecological environment. Existing ecological warning wall technology is mainly used for slope treatment, urban greening and pollution control. Its structure often adopts a multi-layer composite design (such as concrete layer and drainage system) combined with the purification function of plants or microorganisms.

[0003] However, these technologies suffer from poor environmental adaptability, weak real-time monitoring capabilities, high maintenance costs, and limited functionality. They cannot provide real-time feedback of pollution data, and microbial activity is easily inhibited by temperature fluctuations. Therefore, there is a need for an interdisciplinary integrated ecological warning wall that can achieve highly sensitive environmental monitoring and also meet the requirements of modular, low-maintenance design (such as solar power and detachable planting units) to address the shortcomings of existing ecological warning walls. Summary of the Invention

[0004] In response to the problems mentioned in the background art, this invention proposes an intelligent ecological warning system that solves the core defects of traditional ecological warning walls, namely poor environmental adaptability and delayed monitoring response. It also specifically improves the problems of high maintenance costs and limited functionality. Through biosignal quantification, intelligent dynamic regulation, and modular scalable architecture, it systematically solves the environmental vulnerability, monitoring lag, high maintenance, and limited functionality of traditional ecological warning walls, achieving a technological leap from passive protection to active response.

[0005] Technical Solution: To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] An intelligent ecological warning system includes an ecological warning wall, a biosensor layer, and an intelligent control system;

[0007] Ecological Warning Wall: This ecological warning wall achieves highly sensitive detection and rapid response to pollutants at the ppt level through the fusion analysis of plant electrophysiological signals and microbial metabolic responses; the random forest algorithm fuses and calculates multi-source biological data to achieve accurate source tracing and risk level assessment of pollution sources; and through the color gradient of electrochromic glass and data visualization of the projection interface, the invisible pollution situation is transformed into an intuitive public warning signal; it achieves energy self-sufficiency and low-maintenance operation through solar power and biomimetic irrigation; combined with modular design to support functional expansion; and finally, through dynamic control and execution modules, it achieves a full-chain environmental governance and ecological education from pollution perception and early warning to public interaction.

[0008] Biosensing layer: This biosensing layer transforms the natural sensitivity of organisms to environmental pollution into quantifiable and transmissible digital signals; it captures microvolt-level electrophysiological fluctuations in plants under pollutant intrusion in situ using a graphene electrode array; it maintains the activity of luminescent bacteria and monitors their metabolic intensity response to environmental toxins using a microfluidic chip; and finally, it converts the differentiated biological response characteristics excited by different pollutants into a standardized data stream that can be analyzed by the upper-level intelligent system, realizing in situ, live, and continuous monitoring of complex pollution conditions.

[0009] Intelligent Control System: This intelligent control system realizes intelligent closed-loop management of the entire process from biosignals to environmental warnings; it collects and processes biosensor data of plant electrical signals and microbial luminescence intensity in real time, converting them into digital pollution indices; then, it uses data fusion and intelligent decision-making modules to comprehensively analyze multi-source information, realize the identification of pollution types, the tracing of pollution sources, and the assessment of risk levels; finally, it completes an automated intelligent cycle of perception-analysis-decision-response.

[0010] Preferably, the ecological warning wall includes a vertical support frame, a honeycomb matrix planting layer, an acrylic light guide plate and a microbial reaction unit, a display and interaction layer, and an energy and irrigation module.

[0011] The vertical support frame has a grid structure and is fixed to the wall with T-bolts;

[0012] The honeycomb substrate planting layer consists of multiple honeycomb units, which are filled with lightweight nutrient soil. Multiple honeycomb units are snapped into the grid of the vertical support frame, with their bottoms in close contact with the irrigation system and their tops having reserved holes for plant growth and connection to sensors.

[0013] The acrylic light guide plate and the microbial reaction unit constitute a detection and display system for bacterial luminescence signals;

[0014] The display and interaction layer is the direct interface between the system and environmental users. The electrochromic glass can undergo a rapid chemical color change reaction after different voltages are applied, thus gradually changing from blue to red according to the PM2.5 concentration gradient. The auxiliary laser projector receives pollution source data processed by cloud algorithms from the system and dynamically projects this information into a designated area in the form of visualized graphics and text, realizing multi-level and highly visible environmental warnings.

[0015] The energy and irrigation module includes an off-grid power supply system and an irrigation system;

[0016] The off-grid power supply system includes high-efficiency monocrystalline silicon solar panels, an MPPT controller, and energy storage batteries to provide continuous power to the entire warning wall;

[0017] The irrigation system includes a nutrient solution tank, a precision peristaltic pump, and a network of glass fiber capillaries throughout the planting layer, which automatically supplies water and nutrients to the plant roots using capillary action.

[0018] Preferably, the specific contents of the biosensing layer are as follows:

[0019] The plant electrode section of the biosensing layer is responsible for capturing physiological signals from plants. Its core is a graphene electrode array that is wrapped in a ring around the plant stem. The electrodes are connected to a centralized signal acquisition box through a flexible printed circuit. The acquisition box integrates an instrumentation amplifier circuit, which can amplify millivolt-level electrical signals and filter noise. Finally, the processed signal is transmitted to the main controller through a shielded twisted pair cable.

[0020] Preferably, the intelligent control system includes a signal processing and acquisition module, a data fusion and intelligent decision-making module, a dynamic control and execution module, and a communication and energy management module;

[0021] Signal processing and acquisition module: converts analog signals generated by biological organisms into precise digital signals that can be analyzed;

[0022] Data fusion and intelligent decision-making module: responsible for integrating, analyzing, and intelligently judging multi-source heterogeneous data;

[0023] Dynamic control and execution module: responsible for translating decision-making instructions into specific physical actions;

[0024] Communication and Energy Management Module: This is the system's support and guarantee unit, responsible for data transmission and energy supply.

[0025] Preferably, the signal processing and acquisition module contains the following features:

[0026] The millivolt-level electrical signals captured by the plant electrodes are amplified by a high-precision instrumentation amplifier, and the luminescence intensity of microorganisms is digitally sampled using a 16-bit analog-to-digital converter. At the same time, hardware filtering and software algorithms are used to eliminate environmental noise.

[0027] Preferably, the data fusion and intelligent decision-making module contains the following:

[0028] Based on machine learning algorithms, the system simultaneously analyzes plant electrophysiological fluctuations, microbial metabolic changes, and environmental parameters. Through feature extraction and pattern recognition, it achieves qualitative judgment of pollutant types and quantitative source tracing of pollution. Based on preset thresholds, it generates graded early warning instructions and control strategies to drive subsequent response execution modules.

[0029] Preferably, the specific content of the dynamic control and execution module is as follows:

[0030] The intensity of the supplementary LED and the flow rate of the nutrient solution from the peristaltic pump are adjusted in real time through a PID control algorithm to maintain the optimal activity of the biosensor. At the same time, the electrochromic glass is driven to display chromatographic warnings according to the pollution level, and the projection equipment is controlled to dynamically update the pollution map and source tracing information, so as to achieve a comprehensive response from environmental maintenance to visual warnings.

[0031] Preferably, the specific content of the communication and energy management module is as follows:

[0032] This module includes a communication section and an energy section. The communication section uses LoRa wireless transmission technology to achieve low-power, long-distance data aggregation and cloud interaction of sensor nodes. The energy section maximizes photovoltaic conversion efficiency through the MPPT solar controller and intelligently manages the charging and discharging process of the lithium energy storage battery to ensure that the system can still achieve continuous and stable operation 24 / 7 under off-grid conditions.

[0033] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0034] (1) This invention achieves ppt-level pollutant detection sensitivity through graphene electrode array and microfluidic luminescent bacteria chip, shortening the response time; combined with biomimetic vascular bundle irrigation and solar power supply, it reduces water and electricity consumption and operation and maintenance costs, and extends the maintenance cycle; adopts modular 3D printing structure and LoRa (long-range radio) wireless transmission to reduce manual intervention; at the same time, it integrates CRISPR (clustered regularly spaced short palindromic repeating sequences) customized bacteria and fractal algorithm plant layout, so that the system has pollution monitoring, carbon sink value-added and dynamic landscape interaction functions, and comprehensively surpasses traditional technologies in terms of accuracy, efficiency, sustainability and ecological benefits.

[0035] (2) This invention constructs a biosensing layer and combines it with an intelligent control system to convert biological signals into a digital pollution index. Subsequently, it uses wireless transmission and machine learning to trace pollution sources and drives electrochromic glass and a projection interface to visualize the environmental status in real time. Its advantages are:

[0036] 1. Utilizing the bioaccumulation effect to achieve ppt-level detection sensitivity (a thousand times higher than traditional ppb-level sensitivity), responding to pollutant changes within 15 minutes;

[0037] 2. Biomimetic vascular bundle irrigation and solar power reduce operation and maintenance costs; operation and maintenance costs are reduced by 62%, annual carbon sequestration is 12kg / m² and net carbon emissions are -3kgCO2 / m² / yr;

[0038] 3. Modular design dynamically optimizes the layout of the plant landscape, combining ecological restoration with aesthetic interaction;

[0039] It supports rapid replacement of CRISPR strains for extended monitoring. Technicians can directly reproduce the system based on standardized component parameters (such as peristaltic pump flow rate 0-10mL / min, PID control light intensity -50lux~50lux) without additional experiments. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the functional modules of the intelligent ecological warning system of the present invention. Detailed Implementation

[0041] The present invention will be further illustrated below with reference to specific embodiments. These embodiments are implemented based on the technical solutions of the present invention, and it should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0042] like Figure 1 As shown, the intelligent ecological warning system provided in this embodiment includes an ecological warning wall, a biosensor layer, and an intelligent control system.

[0043] Ecological Warning Wall: This ecological warning wall aims to build a comprehensive platform integrating real-time environmental perception, multi-pollutant collaborative monitoring, intelligent source tracing and early warning, and dynamic ecological display. Its core function is to achieve high-sensitivity detection of pollutants such as heavy metals, VOCs, and PM2.5 at the ppt level and rapid response within 15 minutes through the fusion analysis of plant electrophysiological signals and microbial metabolic responses. The system uses the random forest algorithm to fuse and calculate multi-source biological data, achieving accurate source tracing (accuracy >85%) and risk level assessment of pollution sources. Through the color gradient of electrochromic glass and data visualization on the projection interface, invisible pollution conditions are transformed into intuitive public warning signals.

[0044] Meanwhile, the system achieves energy self-sufficiency and low-maintenance operation through solar power and biomimetic irrigation. Combined with modular design, it supports functional expansion and ultimately achieves a full-chain environmental governance and ecological education function, from pollution perception and early warning to public interaction.

[0045] The ecological warning wall uses an anodized aluminum frame (1.5m column spacing) to support a 3D-printed honeycomb matrix (20mm side length, 65% porosity). Rye grass is implanted inside and integrated with a graphene electrode array (sheet resistance <5Ω / sq) to capture 0.1-10mV level electrophysiological signals from plants in real time. An acrylic light guide plate embeds a microfluidic chip (200μm channel width) to cultivate Vibrio fischeri to monitor pollutant metabolic responses, and a photodiode detects changes in light intensity.

[0046] The ecological warning wall includes a vertical support frame, a honeycomb matrix planting layer, an acrylic light guide plate and microbial reaction unit, a display and interaction layer, and an energy and irrigation module.

[0047] 1. Vertical support frame;

[0048] The vertical support frame is the core skeleton of the entire system, constructed from anodized aluminum profiles with a grid structure of 100mm × 50mm. This frame is securely connected to chemical anchors pre-embedded in the wall using T-bolts. The 1.5-meter spacing between columns and 0.6-meter spacing between beams ensures overall stability. The surface is coated with an anti-corrosion polyurethane coating that effectively resists weathering, and the precisely machined grooves and mounting holes on the frame provide a standardized installation base for all other functional modules.

[0049] The overall shape of the frame is a large, flat rectangular grid structure. Its overall dimensions can be flexibly customized according to the application scenario (common height 3-6 meters, width 5-20 meters).

[0050] 2. Honeycomb substrate planting layer;

[0051] The honeycomb-shaped substrate planting layer serves as the plant growth carrier, composed of numerous 3D-printed hexagonal PLA composite material units. Its high porosity of 65% optimizes gas exchange within the roots. Each 20mm-side honeycomb unit is filled with lightweight nutrient soil with specific pH and conductivity, and planted with pollutant-sensitive plants such as ryegrass. These modular units resemble a jigsaw puzzle. Figure 1 The sample is directly inserted into the grid of the vertical support frame, with its bottom in close contact with the irrigation system and its top having reserved holes for plant growth and connection to sensors.

[0052] 3. Acrylic light guide plate and microbial reaction unit;

[0053] The acrylic light guide plate and the microbial reaction unit together constitute a system for detecting and displaying bacterial luminescence signals. The highly transparent acrylic plate serves as the light guide medium, and the PDMS microfluidic chip embedded within it forms a miniature culture chamber that houses the luminescent bacteria. Silicone tubes connect the chip to an external peristaltic pump and nutrient solution tank, forming a closed circulation system to maintain bacterial activity. Meanwhile, photodiodes positioned close to the edge of the light guide plate can accurately detect the attenuation of bacterial luminescence intensity caused by contaminants.

[0054] 4. Display and interaction layer (electrochromic glass and projection);

[0055] The display and interaction layer serves as the direct interface between the system and environmental users. The electrochromic glass undergoes a rapid chemical color-changing reaction when different voltages are applied, gradually changing from blue to red according to the PM2.5 concentration gradient. The auxiliary laser projector receives pollution source tracing data processed by cloud algorithms from the system and dynamically projects this information as visualized graphics and text onto designated areas, achieving multi-layered, highly visible environmental warnings.

[0056] 5. Energy and Irrigation Module;

[0057] The energy and irrigation modules are crucial for maintaining the system's long-term autonomous operation. High-efficiency monocrystalline silicon solar panels, along with the MPPT controller and energy storage batteries, constitute an off-grid power supply system, providing continuous power to the entire warning wall. The irrigation system consists of nutrient solution tanks, precision peristaltic pumps, and a network of glass fiber capillaries distributed throughout the planting layer. This biomimetic design utilizes capillary action to automatically supply water and nutrients to the plant roots, significantly reducing maintenance frequency and water consumption.

[0058] Biosensing Layer: As the core of the entire system, this biosensing layer aims to transform the natural sensitivity of organisms to environmental pollution into quantifiable and transmissible digital signals. It achieves its function through two parallel mechanisms: first, by using a graphene electrode array to capture in situ the microvolt-level electrophysiological fluctuations (such as membrane potential changes) generated by plants under pollutant stress; and second, by using a microfluidic chip to maintain the activity of luminescent bacteria and monitor their metabolic intensity (such as light intensity decay rate) in response to environmental toxins. This layer ultimately converts the differentiated biological response characteristics elicited by different pollutants (such as heavy metals and VOCs) into a standardized data stream that can be analyzed by the upper-level intelligent system, enabling in-situ, living, and continuous monitoring of complex pollution conditions. This provides the entire warning system with a life-sensing dimension that differs from traditional sensors.

[0059] Biosensor Layer (Plant Electrodes): The plant electrode portion of the biosensor layer is responsible for capturing physiological signals from the plant. Its core is a ring-shaped array of graphene electrodes wrapped around the plant stem. These electrodes have a sheet resistance of less than 5 Ω / sq to ensure signal transmission quality. The electrodes are connected to a centralized signal acquisition box via flexible printed circuitry. The acquisition box integrates a high-performance instrumentation amplifier circuit, which amplifies weak millivolt-level electrical signals and filters noise. Finally, the processed signal is transmitted to the main controller via shielded twisted-pair cables.

[0060] Intelligent Control System: This intelligent control system aims to serve as the central nervous system of the entire ecological warning wall. Its core function is to achieve intelligent closed-loop management of the entire process from biosignals to environmental warnings. The system collects and processes real-time biosensor data such as plant electrical signals and microbial luminescence intensity, converting them into digital pollution indices. Then, it utilizes data fusion and intelligent decision-making modules (such as random forest algorithms) to comprehensively analyze multi-source information, achieving pollution type identification, pollution source tracing (accuracy >85%), and risk level assessment. Finally, through dynamic control and execution modules, it automatically adjusts parameters such as supplemental lighting and nutrient solution to maintain optimal activity of the biosensors, and drives electrochromic glass and projection equipment to achieve visualized warnings and public interaction of pollution data, thus completing an automated intelligent cycle of perception-analysis-decision-response.

[0061] Intelligent control system: Includes AD623 amplifier (1000x gain), STM32 microcontroller and LoRa wireless module (3km coverage), drives electrochromic glass (PM2.5 concentration chromatographic change) and projection interface, and uses solar power (22.5% conversion rate) and biomimetic vascular bundle irrigation (60% water saving) to maintain system operation.

[0062] When pollutants are present in the environment, plant roots absorb them, causing membrane potential fluctuations (>2mV). The graphene electrode array captures and amplifies the signal. Simultaneously, the pollutants inhibit the expression of the microbial lux operon, resulting in a 50% decrease in light intensity within 15 minutes. The signal is analyzed into a pollution index by an STM32 microcontroller. Combined with a random forest algorithm (accuracy >85%), the pollution source is traced, triggering a color change (blue → red) on the electrochromic glass and a projected warning. The system dynamically adjusts the supplemental light intensity (-50 lux ~ 50 lux) and nutrient solution flow rate (0-10 mL / min) using a PID algorithm to maintain stable biological activity under extreme conditions of -20℃ to 45℃.

[0063] The intelligent control system includes a signal processing and acquisition module, a data fusion and intelligent decision-making module, a dynamic control and execution module, and a communication and energy management module.

[0064] Signal processing and acquisition module: This module is the sensing front end of the system. Its core function is to convert the weak analog signals generated by the organism into accurate digital signals that can be analyzed.

[0065] This module amplifies the millivolt-level electrical signal captured by the plant electrode by 1000 times using a high-precision instrumentation amplifier (such as AD623), and digitally samples the luminescence intensity of microorganisms using a 16-bit analog-to-digital converter (such as ADS1115). At the same time, it uses hardware filtering and software algorithms (such as moving average filtering) to eliminate environmental noise, ensuring the reliability and accuracy of the collected data and providing a clean data source for subsequent analysis.

[0066] Data Fusion and Intelligent Decision-Making Module: This module serves as the system's computing hub, responsible for integrating, analyzing, and intelligently judging multi-source heterogeneous data.

[0067] This module uses machine learning algorithms (such as trained random forest models) to simultaneously analyze plant electrophysiological fluctuations, microbial metabolic changes, and environmental parameters. Through feature extraction and pattern recognition, it achieves qualitative judgment of pollutant types and quantitative source tracing of pollution (accuracy > 85%). Based on preset thresholds, it generates graded early warning instructions and control strategies to drive subsequent response execution modules.

[0068] The implementation of the machine learning algorithm in this invention covers the entire process of data preparation, model training, and real-time prediction:

[0069] Data preparation: First, extract characteristic values ​​such as mean, standard deviation and attenuation rate from the raw data such as plant electrical signals and microbial light intensity, and then perform standardization processing.

[0070] Model Training: Subsequently, multiple decision trees are constructed using Bootstrap sampling. During node splitting, a subset of features is randomly selected, and random forest training is completed using the minimization of Gini impurity as the criterion. The specific calculation formula is as follows:

[0071] G=1-Σpi²

[0072] In the decision trees used for classification in the Random Forest algorithm, G represents Gini impurity, which is a metric for the purity of a dataset or node. The smaller the value of G, the more singular and pure the sample categories contained in that node are.

[0073] pi represents the proportion (probability) of samples of the i-th category in the current node. Assuming a node has 10 samples, of which 6 belong to industrial pollution and 4 belong to traffic pollution, then:

[0074] p1 (proportion of industrial pollution categories) = 6 / 10 = 0.6.

[0075] p2 (proportion of traffic pollution categories) = 4 / 10 = 0.4.

[0076] Σpi² is the sum of the squares of the proportions of all categories: Σpi²=(0.6)²+(0.4)²=0.36+0.16=0.52.

[0077] Finally, G = 1 - 0.52 = 0.48. This 0.48 is the Gini impurity of that node.

[0078] In the real-time monitoring phase: the preprocessed feature vectors are input into the model, each decision tree is independently classified, and the final pollution type is determined by a majority voting mechanism. The confidence level is calculated based on the vote rate, thereby achieving pollution source tracing with an accuracy of over 85%.

[0079] The final pollution type is determined based on a voting mechanism, resulting in the following calculation formula for the final prediction:

[0080] Final prediction = mode{Tree1(X),...,TreeN(X)}

[0081] `mode` represents the mode, which is the value that appears most frequently in a set of data. In this formula, `mode{...}` means that the predictions of all decision trees within the curly braces are statistically analyzed, and the category that appears most frequently is output. This is the mathematical expression of how random forests make their final decisions through collective voting.

[0082] X represents the input feature vector. It is a structured dataset containing all sensor feature values ​​at the current moment. For example, X = [average plant voltage = 2.1mV, light intensity attenuation rate = 0.3, temperature = 25℃, humidity = 60%]. This vector is the basis for each decision tree's judgment.

[0083] TreeN(X) represents the prediction result of the Nth decision tree for the input feature vector X. When X is input into a trained decision tree, it starts from the root and, based on the values ​​of each feature in X, proceeds downwards along the tree's splitting rules (e.g., whether the light intensity attenuation rate is greater than 0.5) until it reaches a leaf node. The category represented by this leaf node (e.g., "industrial pollution") is the prediction result TreeN(X) of this tree.

[0084] Dynamic Control and Execution Module: This module is the system's execution terminal, responsible for converting decision-making instructions into specific physical actions.

[0085] This module uses a PID control algorithm to adjust the intensity of the 450nm supplemental LED in real time (accuracy range of -50 lux to 50 lux) and the nutrient solution flow rate of the peristaltic pump (0-10 mL / min) to maintain the optimal activity state of the biosensor. At the same time, it drives the electrochromic glass to display chromatographic warnings according to the pollution level and controls the projection device to dynamically update the pollution map and source tracing information, realizing a comprehensive response from environmental maintenance to visual warnings.

[0086] The system's dynamic control and execution module achieves environmental steady-state maintenance and alarm signal triggering through a closed-loop control strategy.

[0087] In terms of environmental control, a discrete PID algorithm is used to adjust the PWM duty cycle of the 450nm supplemental LED in real time, stabilizing the light intensity for microorganisms at the target value of 4950 lux to 5050 lux. The specific calculation formula is as follows:

[0088]

[0089] Output represents the controller's output value. This is the final command signal obtained after calculation using the PID algorithm. In the dynamic control and execution module of this invention, this output value is typically mapped to: the duty cycle (e.g., 0-255) of the PWM (Pulse Width Modulation) signal driving the LED fill light; or the pulse signal controlling the peristaltic pump's speed or switching duration. Function: Directly determines the intensity of the actuator's (lamp, pump) action.

[0090] Kp represents the proportional gain coefficient. It determines the strength of the controller's response to the current error. The larger the Kp value, the stronger and faster the correction of the current deviation. However, if it is too large, the system will oscillate or overshoot, becoming unstable.

[0091] e k This represents the instantaneous error at the current sampling time (the kth time).

[0092] Calculation formula: e k =Set target value - Current measurement value.

[0093] If the goal is to maintain the light intensity for microorganisms at 5000 lux, and the current photodiode reading is 4800 lux, then the current error e k = 5000-4800 = 200 lux.

[0094] K i This represents the integral gain coefficient. It is used to eliminate the steady-state error of the system (i.e., the small deviation that still exists after the system stabilizes). This is achieved by accumulating the error over all past moments (Σe). i The integral term (×Δt term) can output a continuously acting corrective force until the error is completely eliminated.

[0095] e i This represents the historical error value of the i-th sampling point from the initial time to the current time (k).

[0096] Σe i This indicates the range from e1 (first sampling error) to e k The summation of all historical error values ​​for the current error represents the total amount of error accumulated over time.

[0097] Δt represents the sampling time interval of the control system. This is a fixed time interval (e.g., 0.1 seconds or 1 second) between two reads of sensor data and calculation of the PID output by the microcontroller. It discretizes the continuous analog control process into steps that the digital system can process and serves as the time reference for integral accumulation and derivative calculation.

[0098] K d This represents the differential gain coefficient. It is based on the rate of change of the error (e... k -e k-1 The system uses Δt / Δt to predict future error trends and applies a reverse damping effect in advance. This helps suppress system oscillations, making the response process smoother and faster to stabilize at the target value. d Excessive values ​​may amplify measurement noise.

[0099] Meanwhile, the activity of the bacterial community is maintained through flow-concentration feedback control of the peristaltic pump (Q=K×ΔC) or a timed triggering mechanism.

[0100] Here, Q represents the output flow rate of the peristaltic pump (unit: typically milliliters per minute). This is the direct output of the controller and determines the volume of fresh nutrient solution added to the microbial reaction unit (microfluidic chip) per unit time. Precise control of Q is achieved by adjusting the pump's rotational speed or stepping frequency.

[0101] K represents the proportional control gain coefficient. This is a pre-set constant that amplifies or converts the concentration deviation ΔC into a corresponding flow adjustment command. The value of K determines the system's response speed and intensity to concentration deviations.

[0102] A larger K value indicates a more sensitive system response and the ability to quickly correct deviations, but it may cause flow fluctuations or overshoot.

[0103] A smaller K value indicates a smoother system response and a more stable adjustment process, but the correction speed may be slower.

[0104] Function: To link biological metabolic needs (manifested as concentration changes) with physical actions (pump flow rate).

[0105] ΔC represents the deviation between the current bacterial concentration and the target concentration.

[0106] Calculation formula: ΔC = C_target - C_current.

[0107] Wherein, C_target represents the preset target nutrient or metabolite concentration (e.g., glucose concentration or indicator pH value) for maintaining optimal activity of the luminescent bacteria. C_current represents the current concentration value monitored in real time by a microsensor (such as an electrochemical sensor or an optical sensor) integrated into the microfluidic system.

[0108] Positive deviation: ΔC>0 indicates that the current concentration is lower than the target and nutrient solution needs to be added. Q is positive.

[0109] Negative deviation or zero deviation: ΔC≤0 indicates that the concentration has reached or exceeded the target, Q is zero or very small, and the pump stops or runs at low speed.

[0110] At the level of warning display, a linear mapping relationship is established between the pollution index and the driving signal (Vout = PM). 2.5 / 25), converting PM2.5 concentration (0-75μg / m³) into a 0-3V driving voltage, precisely controlling the blue-red spectrum gradient of the electrochromic glass, and pushing the source tracing data to the projection terminal through the MQTT protocol (Message Queuing Telemetry Transport), completing the automated conversion from decision signal to physical execution.

[0111] Vout is the control signal output by the intelligent control system to the electrochromic glass drive circuit; it is an analog voltage value. Its function is to precisely control the oxidation-reduction reaction of the electrochromic material (such as tungsten trioxide WO3) inside the glass, thereby changing its optical properties (color and transparency).

[0112] Working principle: The color depth of electrochromic glass is a function of the applied voltage. In this embodiment, it is simplified to a linear control range:

[0113] When Vout=0V, the glass appears blue (indicating excellent air quality, PM2.5). 2.5 (Low concentration).

[0114] As the Vout voltage increases, the glass color gradually transitions to purple and red.

[0115] When Vout = 3V (assuming this is the maximum safe voltage for the drive circuit), the glass turns deep red (indicating severe pollution, PM2.5). 2.5 High concentration).

[0116] Formula mapping relationship: Formula Vout=PM 2.5 / 25 establishes a direct bridge between pollution data and control instructions.

[0117] Input: PM 2.5 It is the current PM2.5 mass concentration (unit: μg / m³) calculated by the decision module.

[0118] Calculation: Divide the concentration value by the constant 25.

[0119] Output: Obtain the corresponding driving voltage Vout.

[0120] Example: When the PM2.5 concentration is 75 μg / m³, the controller output voltage Vout = 75 / 25 = 3.0V, which will cause the glass to display a preset red color, representing the highest pollution warning.

[0121] Communication and Energy Management Module: This module is the system's support and guarantee unit, undertaking the key tasks of data transmission and energy supply.

[0122] This module includes a communication section and an energy section. The communication section uses LoRa wireless transmission technology (coverage radius of 3km) to achieve low-power, long-distance data aggregation and cloud interaction from sensor nodes; the energy section maximizes photovoltaic conversion efficiency (>22%) through an MPPT solar controller and intelligently manages the charging and discharging process of lithium energy storage batteries to ensure that the system can still achieve continuous and stable operation 24 / 7 under off-grid conditions.

[0123] The system uses solar power (conversion rate 22.5%) and biomimetic vascular bundle irrigation (water saving 60%). It achieves pollution source tracing through LoRa wireless transmission (3km coverage) and random forest algorithm (accuracy >85%), and drives electrochromic glass (PM2.5 concentration chromatographic change) and projection interface for dynamic warnings.

[0124] This application constructs a biosensing layer and combines it with an intelligent control system to convert biosignals into a digital pollution index. Subsequently, pollution source tracing is achieved through wireless transmission and machine learning, and electrochromic glass and a projection interface are used to visualize the environmental status in real time. Its advantages are:

[0125] 1. Utilizing the bioaccumulation effect to achieve ppt-level detection sensitivity (a thousand times higher than traditional ppb-level sensitivity), responding to pollutant changes within 15 minutes;

[0126] 2. Biomimetic vascular bundle irrigation and solar power reduce operation and maintenance costs;

[0127] 3. Modular design dynamically optimizes the layout of the plant landscape, combining ecological restoration with aesthetic interaction.

[0128] The detection sensitivity of this application system reaches the ppt level (a thousand times higher than the traditional ppb level), reduces operation and maintenance costs by 62%, has an annual carbon sequestration of 12 kg / m² and a net carbon emission of -3 kg CO2 / m² / yr, supports CRISPR strain rapid replacement extended monitoring function, and technicians can directly reproduce the system based on standardized component parameters (such as peristaltic pump flow rate 0-10 mL / min, PID control of light intensity -50 lux to 50 lux) without additional experiments.

[0129] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An intelligent ecological warning system, characterized in that: This includes an ecological warning wall, a biosensing layer, and an intelligent control system; Ecological Warning Wall: This ecological warning wall achieves highly sensitive detection and rapid response to pollutants at the ppt level through the fusion analysis of plant electrophysiological signals and microbial metabolic responses; the random forest algorithm fuses and calculates multi-source biological data to achieve accurate source tracing and risk level assessment of pollution sources; and through the color gradient of electrochromic glass and data visualization of the projection interface, the invisible pollution situation is transformed into an intuitive public warning signal. It achieves energy self-sufficiency and low-maintenance operation through solar power and biomimetic irrigation, and combined with modular design to support functional expansion, it ultimately achieves a full-chain environmental governance and ecological education from pollution perception and early warning to public interaction. Biosensing layer: This biosensing layer transforms the natural sensitivity of organisms to environmental pollution into quantifiable and transmissible digital signals; it captures microvolt-level electrophysiological fluctuations in plants under pollutant intrusion in situ using a graphene electrode array; it maintains the activity of luminescent bacteria and monitors their metabolic intensity response to environmental toxins using a microfluidic chip; and finally, it converts the differentiated biological response characteristics excited by different pollutants into a standardized data stream that can be analyzed by the upper-level intelligent system, realizing in situ, live, and continuous monitoring of complex pollution conditions. Intelligent Control System: This intelligent control system realizes intelligent closed-loop management of the entire process from biosignals to environmental warnings; it collects and processes biosensor data of plant electrical signals and microbial luminescence intensity in real time, converting them into digital pollution indices; then, it uses data fusion and intelligent decision-making modules to comprehensively analyze multi-source information, realizing the identification of pollution types, the tracing of pollution sources, and the assessment of risk levels; finally, through dynamic control and execution modules, it completes an automated intelligent cycle of perception-analysis-decision-response.

2. The intelligent ecological warning system according to claim 1, characterized in that: The ecological warning wall includes a vertical support frame, a honeycomb matrix planting layer, an acrylic light guide plate and microbial reaction unit, a display and interaction layer, and an energy and irrigation module. The vertical support frame has a grid structure and is fixed to the wall with T-bolts; The honeycomb substrate planting layer consists of multiple honeycomb units, which are filled with lightweight nutrient soil. Multiple honeycomb units are snapped into the grid of the vertical support frame, with their bottoms in close contact with the irrigation system and their tops having reserved holes for plant growth and connection to sensors. The acrylic light guide plate and the microbial reaction unit constitute a detection and display system for bacterial luminescence signals; The display and interaction layer is the direct interface between the system and environmental users. The electrochromic glass can undergo a rapid chemical color change reaction after different voltages are applied, thus gradually changing from blue to red according to the PM2.5 concentration gradient. The auxiliary laser projector receives pollution source data processed by cloud algorithms from the system and dynamically projects this information into a designated area in the form of visualized graphics and text, realizing multi-level and highly visible environmental warnings. The energy and irrigation module includes an off-grid power supply system and an irrigation system; The off-grid power supply system includes high-efficiency monocrystalline silicon solar panels, an MPPT controller, and energy storage batteries to provide continuous power to the entire warning wall; The irrigation system includes a nutrient solution tank, a precision peristaltic pump, and a network of glass fiber capillaries throughout the planting layer, which automatically supplies water and nutrients to the plant roots using capillary action.

3. The intelligent ecological warning system according to claim 1, characterized in that: The specific contents of the biosensing layer are as follows: The plant electrode section of the biosensing layer is responsible for capturing physiological signals from plants. Its core is a graphene electrode array that is wrapped in a ring around the plant stem. The electrodes are connected to a centralized signal acquisition box through a flexible printed circuit. The acquisition box integrates an instrumentation amplifier circuit, which can amplify millivolt-level electrical signals and filter noise. Finally, the processed signal is transmitted to the main controller through a shielded twisted pair cable.

4. The intelligent ecological warning system according to claim 1, characterized in that: The intelligent control system includes a signal processing and acquisition module, a data fusion and intelligent decision-making module, a dynamic control and execution module, and a communication and energy management module. Signal processing and acquisition module: converts analog signals generated by biological organisms into precise digital signals that can be analyzed; Data fusion and intelligent decision-making module: responsible for integrating, analyzing, and intelligently judging multi-source heterogeneous data; Dynamic control and execution module: responsible for translating decision-making instructions into specific physical actions; Communication and Energy Management Module: This is the system's support and guarantee unit, responsible for data transmission and energy supply.

5. The intelligent ecological warning system according to claim 4, characterized in that: The specific contents of the signal processing and acquisition module are as follows: The millivolt-level electrical signals captured by the plant electrodes are amplified by a high-precision instrumentation amplifier, and the luminescence intensity of microorganisms is digitally sampled using a 16-bit analog-to-digital converter. At the same time, hardware filtering and software algorithms are used to eliminate environmental noise.

6. The intelligent ecological warning system according to claim 4, characterized in that: The specific content of the data fusion and intelligent decision-making module is as follows: Based on machine learning algorithms, the system simultaneously analyzes plant electrophysiological fluctuations, microbial metabolic changes, and environmental parameters. Through feature extraction and pattern recognition, it achieves qualitative judgment of pollutant types and quantitative source tracing of pollution. Based on preset thresholds, it generates graded early warning instructions and control strategies to drive subsequent response execution modules.

7. The intelligent ecological warning system according to claim 4, characterized in that: The specific content of the dynamic control and execution module is as follows: The intensity of the supplementary LED and the flow rate of the nutrient solution from the peristaltic pump are adjusted in real time through a PID control algorithm to maintain the optimal activity of the biosensor. At the same time, the electrochromic glass is driven to display chromatographic warnings according to the pollution level, and the projection equipment is controlled to dynamically update the pollution map and source tracing information, so as to achieve a comprehensive response from environmental maintenance to visual warnings.

8. The intelligent ecological warning system according to claim 4, characterized in that: The specific contents of the communication and energy management module are as follows: This module includes a communication section and an energy section. The communication section uses LoRa wireless transmission technology to achieve low-power, long-distance data aggregation and cloud interaction of sensor nodes. The energy section maximizes photovoltaic conversion efficiency through the MPPT solar controller and intelligently manages the charging and discharging process of the lithium energy storage battery to ensure that the system can still achieve continuous and stable operation 24 / 7 under off-grid conditions.

Images