Intelligent sensor abnormity management system based on genetic algorithm

Abstract

The invention provides an intelligent sensor abnormity management system based on a genetic algorithm, and the system optimizes the sensor position and parameter configuration through a genetic algorithm optimization module, and a central control and scheduling module sends an adjustment instruction to a sensor hardware module according to an optimization result, so as to adjust the layout and parameters of each sensor. Meanwhile, the sensor abnormity management module detects the abnormity of the sensor in real time and automatically selects an optimal adjustment strategy. Besides, the states and abnormal information of all the sensors can be displayed in a geographic information system through a GIS visualization module, so that layout is dynamically adjusted, abnormal points are rapidly positioned, the unmanned aerial vehicle can be matched, automatic repair is achieved, and reliability and maintenance efficiency are improved. Advanced technologies such as Internet of Things, edge computing, deep learning, GIS, genetic algorithm and reinforcement learning are fused, an intelligent and automatic sensor management and optimization platform is constructed, and efficient data acquisition, real-time optimization and fault management are achieved. By means of the modular design, the system has high expandability and flexibility and is suitable for various environment monitoring application scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental detection, and particularly to an intelligent sensor anomaly management system based on a genetic algorithm. Background Art

[0002] With the continuous progress of environmental monitoring technology, carbon emission monitoring has become an important link in environmental protection and policy-making. The ISO 14064 standard requires that carbon emission data must be of high quality, reliability, and integrity to ensure the effectiveness of policy decisions. However, traditional carbon emission monitoring systems often rely on sensors at fixed positions, which lack flexibility in layout and are easily affected by factors such as environmental changes, hardware failures, or battery aging, resulting in inaccurate monitoring data and increased errors.

[0003] Existing technologies usually adopt static or manually set sensor layout schemes and improve data accuracy through real-time calibration technology. However, when a sensor experiences an anomaly or failure, existing technologies mainly rely on manual intervention for repair. This not only increases the operation and maintenance costs but also may lead to data collection interruption or data quality degradation during sensor failures, thus affecting the overall monitoring accuracy and reliability. Especially in the field of carbon emission monitoring, ISO 14064 has strict requirements for data accuracy and integrity, and existing technologies fail to fully meet this standard.

[0004] Although real-time calibration technology can improve the accuracy of sensor data to a certain extent, existing technologies usually ignore the impact on monitoring results during the period from sensor anomaly to failure repair. In addition, existing fault detection mechanisms are often relatively single, mainly focusing on simple hardware fault detection or data threshold judgment, lacking the ability to identify complex anomaly patterns and dynamic processing. At the same time, there is a lack of systematic optimization schemes in existing technologies, and it is impossible to adjust the sensor position in real time globally or adopt data compensation strategies according to sensor anomalies. As a result, when a sensor fails or the external environment changes, the monitoring accuracy drops significantly, making it difficult to meet the high standards of data quality required by ISO 14064. Summary of the Invention

[0005] In view of the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide an intelligent sensor anomaly management system based on a genetic algorithm, which is used to solve technical problems such as unreasonable sensor layout, large data errors, and monitoring interruption caused by sensor failures in existing environmental monitoring systems.

[0006] To achieve the above and other related objectives, the present invention provides an intelligent sensor anomaly management system based on a genetic algorithm. The system includes: a sensor hardware module, a genetic algorithm optimization module, a sensor anomaly management module, a data acquisition and processing module, a central control and scheduling module, a GIS anomaly point visualization module, and a user interface and operation module. Among them, the sensor hardware module is built-in with multiple sensors at movable positions for collecting environmental data. The data acquisition and processing module is used to preprocess the environmental data collected by each sensor and real-time feedback the preprocessed environmental data to the genetic algorithm optimization module, the central control and scheduling module, and the sensor anomaly management module. The genetic algorithm optimization module is used to optimize the layout and parameter configuration of each sensor based on the genetic algorithm to obtain an optimization result and feedback it to the central control and scheduling module. The central control and scheduling module is used to send adjustment instructions to the sensor hardware module based on the optimization result feedback by the genetic algorithm optimization module to adjust the layout and parameter configuration of each sensor. The sensor anomaly management module is used to detect the abnormal state of each sensor, automatically trigger an adjustment mechanism when an anomaly is detected, and send the abnormal data and alarm information to the central control and scheduling module for corresponding control. The GIS anomaly point visualization module is used to associate the position information of each sensor with the geographic coordinate system, display the sensor distribution and its health status using a map or 3D model, and identify the sensors detected with anomalies. The user interface and operation module is used to display the system monitoring data and obtain the sensor configuration parameters input by the user.

[0007] In an embodiment of the present invention, the sensor hardware module includes: a plurality of sensor nodes. Among them, each sensor node includes: a sensor for collecting environmental data; a driving device for adjusting the position of the sensor based on an adjustment instruction; a sensor parameter adjustment interface for adjusting the sampling parameters of the sensor based on an adjustment instruction; and a battery monitoring unit for monitoring the battery status of the sensor.

[0008] In an embodiment of the present invention, the genetic algorithm optimization module includes: an initialization population unit for setting the initial configuration parameters of the sensors using a randomization method according to the initial layout of each sensor, extracting the correlation features between the initial configuration parameters and the environmental data based on a deep learning algorithm, and generating an initial population according to the correlation features; a fitness evaluation unit for evaluating the sensor layout scheme using a fitness function; a selection, crossover, and mutation unit for selecting excellent individuals based on the evaluated fitness, performing crossover and mutation operations on the selected excellent individuals to generate a new population, and obtaining the optimal sensor layout scheme; and an optimization feedback unit for adjusting the optimal sensor layout scheme based on the feedback mechanism of reinforcement learning.

[0009] In an embodiment of the present invention, the data acquisition and processing module includes: a data acquisition unit for collecting environmental data collected by each sensor in real time; a data preprocessing unit connected to the data acquisition unit for preprocessing the collected environmental data and sending the processed environmental data outwards.

[0010] In an embodiment of the present invention, the sensor anomaly management module includes: an anomaly detection unit for performing anomaly detection on each sensor based on the environmental data real-time feedback by the data acquisition and processing module and outputting the anomaly type when an anomaly is detected; wherein, the anomaly type includes: hardware failure and battery anomaly, data drift anomaly, and data missing anomaly; an automatic adjustment unit for triggering a corresponding adjustment mechanism according to the anomaly type; an alarm and notification unit for sending alarm information and anomaly data to the central control and scheduling module when an anomaly is detected.

[0011] In an embodiment of the present invention, the triggering of the corresponding adjustment mechanism according to the anomaly type includes: if the anomaly type of the sensor is hardware failure and battery anomaly, activating a standby sensor node and feeding back to the central control and scheduling module to control the genetic algorithm optimization module for optimization to adjust the layout and parameter configuration of the sensor; if the anomaly type of the sensor is data drift anomaly, using a time series algorithm to predict and compensate for the data drift; if the anomaly type of the sensor is data missing anomaly, using multi-node cooperation and deep learning algorithms to supplement the missing data.

[0012] In an embodiment of the present invention, the central control and scheduling module includes: a scheduling and control unit and a communication unit connected to the scheduling and control unit; the scheduling and control unit is used for generating an adjustment instruction based on the optimization result feedback by the genetic algorithm optimization module received through the communication unit and sending the adjustment instruction to each sensor node through the communication unit; and is also used for controlling the corresponding sensor node based on the anomaly data and alarm information received through the communication unit.

[0013] In an embodiment of the present invention, the display and operation module includes: a control panel for real-time displaying sensor data, sensor status, optimization result, anomaly data, and alarm information; a configuration management unit for obtaining sensor configuration parameters input by the user and sending them to the central control and scheduling module to control each sensor node to make adjustments.

[0014] In an embodiment of the present invention, the driving device includes: a servo motor and an electric driver for automatically adjusting the position of the corresponding sensor in cooperation with a preset track.

[0015] In an embodiment of the present invention, various wireless and wired communication protocols can be used for communication between modules; among them, the various wireless communication protocols include multiple ones among LoRa, ZigBee, MQTT, HTTP, CoAP, Bluetooth Mesh, and Wi-Fi 6.

[0016] As described above, the present invention is an intelligent sensor anomaly management system based on a genetic algorithm, having the following beneficial effects: The present invention optimizes the sensor positions and parameter configurations through a genetic algorithm optimization module. The central control and scheduling module sends adjustment instructions to the sensor hardware module according to the optimization results to adjust the layouts and parameters of the sensors. At the same time, the sensor anomaly management module detects sensor anomalies in real time and automatically selects the optimal adjustment strategy. In addition, the status and anomaly information of all sensors will be displayed in a geographic information system (GIS) through the GIS visualization module, so as to dynamically adjust the layout and quickly locate the anomaly points, cooperate with drones to achieve automatic repair, and improve reliability and maintenance efficiency. The present invention integrates advanced technologies such as the Internet of Things, edge computing, deep learning, GIS, genetic algorithm, and reinforcement learning to build an intelligent and automated sensor management and optimization platform to achieve efficient data collection, real-time optimization, and fault management. With a modular design, the system has high scalability and flexibility and is suitable for a variety of environmental monitoring application scenarios. Brief Description of the Drawings

[0017] Figure 1 It shows a schematic structural diagram of an intelligent sensor anomaly management system based on a genetic algorithm in an embodiment of the present invention.

[0018] Figure 2 It shows a schematic structural diagram of a sensor hardware module in an embodiment of the present invention.

[0019] Figure 3 It shows a schematic structural diagram of an intelligent sensor anomaly management system based on a genetic algorithm in an embodiment of the present invention. Detailed Embodiments

[0020] The following uses specific specific examples to illustrate the implementation manners of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0021] It should be noted that in the following description, with reference to the accompanying drawings, several embodiments of the present invention are described. It should be understood that other embodiments may also be used, and mechanical composition, structure, electrical, and operational changes may be made without departing from the spirit and scope of the present invention. The following detailed description should not be considered restrictive, and the scope of the embodiments of the present invention is only defined by the claims of the published patent. The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. Spatially related terms, such as "upper", "lower", "left", "right", "below", "beneath", "lower part", "above", "upper part", etc., may be used in the text to facilitate the description of the relationship between one element or feature shown in the drawings and another element or feature.

[0022] Throughout the specification, when it is said that a certain part is "connected" to another part, this includes not only the case of "direct connection" but also the case of "indirect connection" in which other elements are placed in between. Additionally, when it is said that a certain part "includes" a certain constituent element, unless there is a particularly contrary record, it does not exclude other constituent elements, but means that other constituent elements may also be included.

[0023] The first, second, and third, etc. terms mentioned therein are used to describe various parts, components, regions, layers, and / or segments, but are not limited thereto. These terms are only used to distinguish one part, component, region, layer, or segment from other parts, components, regions, layers, or segments. Therefore, the first part, component, region, layer, or segment described below may refer to the second part, component, region, layer, or segment within the scope not exceeding the present invention.

[0024] Furthermore, as used herein, the singular forms "a", "an", and "the" are intended to also include the plural forms unless the context indicates otherwise. It should be further understood that the terms "comprise", "include" indicate the presence of the stated features, operations, elements, components, items, kinds, and / or groups, but do not exclude the presence, occurrence, or addition of one or more other features, operations, elements, components, items, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or meaning any one or any combination. Thus, "A, B, or C" or "A, B, and / or C" means "any one of the following: A; B; C; A and B; A and C; B and C; A, B, and C". An exception to this definition only occurs when the combination of elements, functions, or operations is inherently mutually exclusive in some way.

[0025] The present invention provides an intelligent sensor anomaly management system based on a genetic algorithm. The genetic algorithm optimization module optimizes the sensor positions and parameter configurations. The central control and scheduling module sends adjustment instructions to the sensor hardware module according to the optimization results, thereby adjusting the layouts and parameters of the sensors. Meanwhile, the sensor anomaly management module detects sensor anomalies in real time and automatically selects the optimal adjustment strategy. In addition, the status and anomaly information of all sensors are displayed in a geographic information system (GIS) through the GIS visualization module, so as to dynamically adjust the layout and quickly locate the anomaly points. Together with drones, automatic repair is achieved, improving reliability and maintenance efficiency. The present invention integrates advanced technologies such as the Internet of Things, edge computing, deep learning, GIS, genetic algorithms, and reinforcement learning to build an intelligent and automated sensor management and optimization platform, achieving efficient data collection, real-time optimization, and fault management. With modular design, the system has high scalability and flexibility and is applicable to various environmental monitoring application scenarios.

[0026] The following will describe the embodiments of the present invention in detail with reference to the accompanying drawings, so that those skilled in the technical field of the present invention can easily implement it. The present invention can be embodied in many different forms and is not limited to the embodiments described herein.

[0027] As Figure 1 FIG. shows a schematic structural diagram of an intelligent sensor anomaly management system based on a genetic algorithm in an embodiment of the present invention.

[0028] The system includes: a sensor hardware module 1, a genetic algorithm optimization module 2, a sensor anomaly management module 3, a data collection and processing module 4, a central control and scheduling module 5, a GIS anomaly point visualization module 6, and a user interface and operation module 7;

[0029] Among them, the central control and scheduling module 5 is communicatively connected to the sensor hardware module 1, the genetic algorithm optimization module 2, the sensor anomaly management module 3, the data collection and processing module 4, the GIS anomaly point visualization module 6, and the user interface and operation module 7 respectively; the data collection and processing module 4 is also communicatively connected to the sensor hardware module 1, the genetic algorithm optimization module 2, and the sensor anomaly management module 3;

[0030] The sensor hardware module 1 is built-in with multiple sensors with movable positions for collecting environmental data, such as parameters such as carbon emissions, air quality, temperature, and humidity; the movable sensor positions enable the system to flexibly adjust the monitoring positions according to actual needs to adapt to different monitoring scenarios and environmental changes.

[0031] The data acquisition and processing module 4 is used to preprocess the environmental data collected by each sensor and feedback the preprocessed environmental data to the genetic algorithm optimization module 2, the central control and scheduling module 5, the sensor anomaly management module 3, and the GIS anomaly point visualization module 6 in real time;

[0032] The genetic algorithm optimization module 2 is used to optimize the layout and parameter configuration of each sensor based on the genetic algorithm to obtain an optimization result and feedback it to the central control and scheduling module 5; The genetic algorithm is an optimization algorithm that simulates natural selection and genetic mechanisms. By continuously iterating, it searches for the optimal sensor layout and parameter combination to improve the monitoring efficiency and accuracy of the sensor system. After that, the optimal sensor layout and parameter combination will be feedback to the central control and scheduling module 5.

[0033] The central control and scheduling module 5, as the core of the system, is responsible for coordinating the work between various modules; Based on the optimization result feedback by the genetic algorithm optimization module, it sends adjustment instructions to the sensor hardware module 1 to adjust the layout and parameter configuration of each sensor; At the same time, it receives the anomaly data and alarm information sent by the sensor anomaly management module 3 and performs corresponding processing.

[0034] The sensor anomaly management module 3 is used to detect the anomaly status of each sensor, automatically trigger an adjustment mechanism when an anomaly is detected, and send the anomaly data and alarm information to the central control and scheduling module 5;

[0035] The GIS anomaly point visualization module 6 is used to associate the position information of each sensor with the geographic coordinate system, display the sensor distribution and its health status using a map or 3D model, and identify the sensors detected with anomalies;

[0036] The user interface and operation module 7 is used to display the system monitoring data and obtain the sensor configuration parameters input by the user. These configuration instructions are transmitted through the central control and scheduling module 5 to the corresponding sensor hardware module 1 for sensor configuration.

[0037] The control flow of the system includes: After the Central Control and Scheduling Module 5 receives the optimization results from the Genetic Algorithm Optimization Module 2, it sends adjustment instructions to the Sensor Hardware Module 1 based on these results. These instructions can change parameters such as the position, sampling frequency, and sensitivity of the sensors to optimize the sensor layout and performance. When the Sensor Anomaly Management Module 3 detects a sensor anomaly, it sends an alarm message to the Central Control and Scheduling Module 5 and triggers the automatic adjustment mechanism. The Central Control and Scheduling Module 5 may require the Genetic Algorithm Optimization Module 2 to re-evaluate and adjust according to the anomaly situation to restore the normal operation of the system. The user can manually input sensor configuration parameters through the User Interface and Operation Module 7. These configuration instructions are passed through the Central Control and Scheduling Module 5 to the corresponding Sensor Hardware Module 1 or Genetic Algorithm Optimization Module 2 to achieve manual control and adjustment of the system by the user.

[0038] The data flow of the system includes: The Sensor Hardware Module 1 collects environmental data and transmits this data to the Data Acquisition and Processing Module 4 through a wireless communication protocol (such as LoRa, ZigBee, etc.). The Data Acquisition and Processing Module 4 preprocesses and analyzes the data, removing noise and error data to improve data quality. The preprocessed data is provided to the Genetic Algorithm Optimization Module 2, which optimizes the sensor layout and configuration based on this data and feeds back the optimization results to the Central Control and Scheduling Module 5. The Central Control and Scheduling Module 5 passes the data to the GIS Anomaly Point Visualization Module 6, which combines geographical information and presents the anomaly monitoring points to the user in the form of a map or 3D model, enabling the user to intuitively understand the distribution and working status of the sensors.

[0039] In one embodiment, as Figure 2 , the Sensor Hardware Module 1 includes: a plurality of sensor nodes; each sensor node includes:

[0040] Sensors, using high-precision environmental monitoring sensors, supporting the measurement of key environmental parameters such as carbon emissions, air quality, temperature and humidity. The sensors need to have highly integrated hardware and be able to support remote configuration and adjustment.

[0041] Drive device, which adjusts the position of the sensor according to the adjustment instructions; uses mechanical devices such as servo motors and stepper motors to control the position adjustment of the sensor within a preset track or range to ensure that the sensor can be in the optimal data acquisition position.

[0042] Sensor parameter adjustment interface, used to adjust the sampling parameters of the sensor based on the adjustment instructions; through hardware interfaces such as PWM modulation, analog signal adjustment, digital control signals, etc., to achieve real-time adjustment of sampling parameters such as the sampling frequency, sensitivity, and range of the sensor.

[0043] The battery monitoring unit is used to monitor the battery status of the sensor. Through the built-in Battery Management System (BMS), it monitors indicators such as battery voltage, temperature, and remaining power. Based on these indicators, the health status of the battery can be fed back in real time. Once abnormal battery conditions are detected, such as too low battery power or abnormal temperature, an alarm will be triggered immediately to remind relevant personnel to handle it in time, avoiding affecting the normal operation of the sensor due to battery problems.

[0044] In one embodiment, as Figure 3 , the genetic algorithm optimization module 2 includes:

[0045] The initial population unit is used to set the initial configuration parameters of the sensor, such as initial position, sensitivity, and sampling frequency, according to the initial layout of each sensor using a randomization method. And using deep learning algorithms, such as Convolutional Neural Network (CNN) or Long Short-Term Memory Network (LSTM), to extract the correlation features between the initial configuration parameters and environmental data (such as temperature, pressure, flow rate, etc.). Based on the correlation features, the population size and gene coding format are set to generate the initial population. The initial population is composed of a series of individuals, and each individual represents a possible sensor configuration scheme. The formula X init = f DL (E, S) is a deep learning model that can generate a meaningful population initialization configuration based on environmental and sensor data.

[0046] The fitness evaluation unit is used to evaluate the fitness of the sensor layout scheme using the fitness function, and comprehensively evaluate the effect of each sensor layout scheme using the fitness function. The fitness function consists of multiple sub-functions, and each sub-function corresponds to a performance metric (such as measurement accuracy, coverage rate, failure rate, etc.).

[0047] Specifically, obtain the representation vector X of each individual from the initial population or the current population. This vector contains various parameters related to the sensor, such as sensor position, environmental temperature, environmental pressure, measurement accuracy, coverage rate, failure rate, data acquisition frequency, flow rate, etc. These parameters comprehensively describe the specific characteristics of a sensor layout scheme.

[0048] X = [X1, X2, X3, X4, X5, X6, X7, X8, X9, X 10 ; (1)

[0049] Collect and prepare the specific data related to the environment and sensor measurements, and calculate the values of each sub-function respectively according to the composition of the fitness function. Each sub-function corresponds to a performance metric:

[0050]

[0051] The sub-function MP(X) represents the measurement accuracy, which is related to the ambient temperature, pressure, flow rate, and component concentration. Among them, T env is the ambient temperature, and T sensor is the temperature measured by the sensor, P env is the ambient pressure, and P sensor is the pressure measured by the sensor, F env is the ambient flow rate, and F sensor is the flow rate measured by the sensor, C env is the ambient temperature, and C sensor is the concentration measured by the sensor, σ temp is the standard deviation of temperature, and σ pressure is the standard deviation of pressure, and σ flow is the standard deviation of flow rate, and σ conc is the standard deviation of concentration; this sub-function measures the closeness between the sensor measurement value and the actual ambient value. It calculates the measurement errors of the ambient temperature, pressure, flow rate, and component concentration (in the form of relative errors and normalized by their respective standard deviations), and then takes the reciprocal to obtain a value. The smaller the error, the larger the value of the measurement accuracy sub-function, indicating higher measurement accuracy of the sensor.

[0052]

[0053] The sub-function C(X) represents the coverage rate, A overlap is the overlapping area of the sensor, and A total is the total area of the target area, α flow is the influence factor of the flow rate on the coverage range, and F flow is the ambient flow rate. This sub-function is used to evaluate the coverage of the sensor in the target area, and the coverage rate C is affected by the flowing gas. When the gas flows faster, the sensor may need to cover a larger area.

[0054]

[0055] The sub-function FR(X) represents the failure rate, FS is the number of faulty sensors, and TS is the total number of sensors, reflecting the failure rate caused by hardware damage or data anomalies of the sensors.

[0056]

[0057] The sub-function TE(X) represents the influence of the ambient temperature, T max is the maximum temperature range, and T min is the minimum temperature range. The influence of temperature on sensor measurement is no longer limited to the ambient temperature, and the temperature of gas flow also needs to be considered.

[0058]

[0059] The sub-function PE(X) represents the influence of environmental pressure, where P max is the maximum pressure range, and P min is the minimum pressure range. The influence of environmental pressure can be adjusted by the pressure of gas flow.

[0060]

[0061] The sub-function FRE(X) represents the influence of environmental flow rate, where F env represents the environmental flow rate, and F max is the maximum flow rate range. The flow rate directly affects the measurement effect of the sensor. When the flow rate is large, the sensor may require higher response speed and accuracy.

[0062]

[0063] The sub-function CE(X) is the influence of component concentration, where C max is the maximum concentration range. The component concentration directly affects the measurement result of the sensor. At high concentrations, the sensor may become saturated or have errors.

[0064] Φ(X) = α·MP(X)+β·C(X)-γ·FR(X)-δ·TE(X)-η·PE(X)+ξSF(X)+κ·FRE(X)+λ·CE(X); (9)

[0065] This function is the fitness function, which is composed of the sum of sub-functions with different weights. Among them, α, β, γ, δ, η, ξ, κ, λ are pre-set weight coefficients, which are used to adjust the importance of each sub-function in the comprehensive evaluation.

[0066] The selection, crossover and mutation units are used to select excellent individuals based on the evaluated fitness, and perform crossover and mutation operations on the selected excellent individuals to generate a new population, so as to obtain the optimal sensor layout scheme;

[0067] Specifically, the core purpose of the selection operation is to select individuals with higher fitness from the current population, so that they have a greater chance to participate in subsequent crossover and mutation operations, thereby passing on excellent genes to the next generation and promoting the population to evolve in a better direction. In the scenario of sensor layout optimization, it is to screen out those individuals whose sensor layout schemes have better comprehensive performance in terms of measurement accuracy, coverage rate, failure rate and other performance indicators. Specifically, the roulette wheel selection and tournament selection methods can be used.

[0068] The crossover operation simulates the gene recombination process in biological evolution. By exchanging genes of the selected excellent individuals, new individuals (offspring) are generated. In this way, the excellent genes of different individuals can be combined together, which may produce individuals with higher fitness, thus promoting the continuous evolution of the population to find a better sensor layout scheme. Specific methods include: in the sensor layout optimization problem, the gene encoding of an individual usually includes parameters such as the position, sensitivity, and sampling frequency of the sensor. The crossover operation can exchange these parameters to generate new offspring.

[0069] The main purpose of the mutation operation is to increase the diversity of the population and avoid the algorithm falling into a local optimal solution. During the evolution process, even if some individuals with higher fitness are found through the selection and crossover operations, these individuals may only be local optimal solutions. Through the mutation operation, one or more individuals in the population are randomly selected, and their positions or parameters in the gene encoding are randomly changed. By making random small changes to the genes of the individuals, it is possible to generate completely new gene combinations, thus jumping out of the local optimum and finding a better global optimum.

[0070] During the evolution process, a termination condition is needed to determine whether the algorithm has converged to a satisfactory solution or whether the computational resource limit has been reached. For example, a maximum number of iteration generations is set, such as 100 generations. When the algorithm reaches the 100th generation, regardless of whether the current fitness meets the requirements, the iteration stops. Or a fitness threshold is preset in advance. When the fitness value of the individual with the highest fitness in the population reaches or exceeds this threshold, it is considered that the algorithm has found a satisfactory solution and the iteration stops. If the termination condition is met, the iteration stops, and the individual with the highest fitness in the current population is output as the optimal solution; otherwise, the next round of selection, crossover, and mutation operations continue.

[0071] The optimization feedback unit is used to adjust the optimal sensor layout scheme based on the feedback mechanism of reinforcement learning. The optimization result of the genetic algorithm will be continuously adjusted according to the feedback mechanism of reinforcement learning. The genetic algorithm is good at global search in a large solution space and can quickly find potential excellent configurations; while the policy optimization method in reinforcement learning will be used to determine the probability and method of crossover and mutation, making the optimization process more adaptive and capable of dynamically adjusting the evolutionary strategy according to the system state.

[0072] The feedback mechanism based on reinforcement learning uses the reward function R(St,at) to evaluate the pros and cons of different actions, and then learns how to adjust the sensor configuration parameters by updating the Q value to achieve further optimization of the scheme.

[0073] R(S t ,a t )=ω1·C(S t ,a t) + ω2MA(S t ,a t ) - ω3FR(S t ,a t ); (10)

[0074] R is the reward function evaluated according to the current configuration, S t is the current state, a t is the action taken, C(S t ,a t ) may represent the coverage rate, MA(S t ,a t ) may represent the measurement accuracy, R(S t ,a t ) may represent the failure rate. ω1, ω2, and ω3 are the corresponding weight coefficients, and this function is used to adjust the importance of each indicator in the reward evaluation. This function is used to evaluate the reward obtained by taking the action a t under the current state S t . The reward function provides a quantitative way for reinforcement learning to evaluate the pros and cons of each action. When the genetic algorithm finds a potentially excellent configuration, reinforcement learning determines whether the action helps improve the performance of the solution by calculating the reward values of this configuration under different actions.

[0075] Q(S t ,a t ) = R(S t ,a t ) + γmaxQ(S t ,a′); (11)

[0076] Q is the Q - value of choosing a certain action under the current state, which reflects the long - term expected reward of taking this action in this state. γ is the discount factor, used to weigh the importance of current rewards and future rewards. When the genetic algorithm finds a potentially excellent configuration in a certain generation, reinforcement learning further analyzes the performance of this configuration in actual operations and updates the Q - value. By continuously updating the Q - value, reinforcement learning can learn which action to take in different states to obtain the maximum long - term reward. In the optimization of the sensor placement scheme, the update of the Q - value can help determine how to adjust the configuration parameters of the sensors (such as position, sensitivity, etc.) to improve the overall performance of the scheme.

[0077] In one embodiment, as Figure 3 ,the data acquisition and processing module 4 includes:

[0078] A data acquisition unit, used to collect the environmental data collected by each sensor in real - time; collect sensor data through the Internet of Things (IoT) protocol (such as LoRa, ZigBee, Wi - Fi6, 5G, etc.).

[0079] A data preprocessing unit, connected to the data acquisition unit, is used to preprocess the collected environmental data and send the processed environmental data outward. Specifically, preprocessing techniques such as data cleaning, filtering, and noise reduction are used to ensure the quality of the collected data. After preprocessing, the quality of the environmental data is improved, and it can be sent outward to the genetic algorithm optimization module, the central control and scheduling module, and the sensor anomaly management module for further analysis and processing.

[0080] In one embodiment, as Figure 3 , the sensor anomaly management module 3 includes:

[0081] An anomaly detection unit, which is used to perform anomaly detection on each sensor based on the environmental data real-time feedback by the data acquisition and processing module, and output the anomaly type when an anomaly is detected; among them, the anomaly type includes: hardware failure and battery anomaly, data drift anomaly, and data missing anomaly; specifically, with the help of the battery monitoring data and environmental data transmitted by the sensor, the fluctuation of the measurement data is dynamically monitored. Based on the statistical analysis method, the range of normal data can be set by calculating statistical quantities such as the mean, variance, and standard deviation of the data. When the data exceeds this range, it is determined as an anomaly. Machine learning techniques such as anomaly detection algorithms and decision trees can also be used. The anomaly detection algorithm can learn the pattern of normal data, and once it finds data that does not conform to this pattern, it will identify it as an anomaly. The decision tree can classify according to different features and conditions to judge the anomaly type. For example, through the decision tree, the size of the data can be compared with a preset threshold to distinguish between hardware failure and battery anomaly, data drift anomaly, and data missing anomaly. Specifically, if the data is greater than the threshold of hardware failure, it is identified as hardware failure and battery anomaly; if the data is less than the threshold of hardware failure and greater than the data anomaly threshold, it will be further analyzed according to the specific data and identified as data drift anomaly or data missing anomaly.

[0082] An automatic adjustment unit, which is used to trigger the corresponding adjustment mechanism according to the anomaly type;

[0083] An alarm and notification unit, which is used to send the alarm information and anomaly data to the central control and scheduling module when an anomaly is detected; when the sensor status is abnormal, an alarm is sent to the central control and scheduling module through a wireless communication or wired communication protocol (such as MQTT, LoRa, ZigBee, etc.), and detailed information about the anomaly type and location is provided.

[0084] In one embodiment, the triggering of the corresponding adjustment mechanism according to the anomaly type includes:

[0085] If the abnormal types of the sensors are hardware failures and battery anomalies, activate the backup sensor nodes and feedback to the central control and scheduling module to control the genetic algorithm optimization module for optimization, so as to adjust the layout and parameter configuration of the sensors;

[0086] If the abnormal type of the sensor is data drift anomaly, use the time series algorithm to predict and compensate for the data drift. The time series algorithm can analyze the change trend and law of the historical data of the sensor, so as to predict the direction and degree of the data drift, compensate the current data, and ensure the data continuity and accuracy;

[0087] If the abnormal type of the sensor is data missing anomaly, utilize multi-node cooperation and deep learning algorithms to supplement the missing data. When a data missing anomaly occurs, the system first resorts to the multi-node cooperation method. The missing data is supplemented by the data of adjacent sensors. The data collected by these adjacent sensors can reflect partial environmental information of this area. The system uses the spatial interpolation (such as Kriging interpolation) method to estimate the approximate value of the missing data according to the positions and data values of the adjacent sensors. At the same time, combined with the multi-sensor data weighted fusion technology, the data obtained through spatial interpolation is further optimized. Different adjacent sensors have different influence degrees on the missing data points. According to factors such as their distances from the missing data points and data quality, different weights are assigned to the data of each adjacent sensor. The data of the sensors closer in distance and with higher data reliability have larger weights, and vice versa. The data of these adjacent sensors are fused according to the weights to obtain a more accurate missing data compensation value. This method ensures efficient compensation and provides more accurate data in case of sensor failures or data missing.

[0088] In an embodiment, such as Figure 3 , the central control and scheduling module 5 includes: a scheduling and control unit and a communication unit connected to the scheduling and control unit;

[0089] The communication unit is used to communicate with each module using a unified communication protocol; the communication protocols such as MQTT, HTTP, CoAP, etc., and integrate Bluetooth Mesh and Wi-Fi 6 technologies; the communication unit is responsible for receiving feedback information from each module. Receive the environmental data from the data acquisition and processing module 4, receive the optimization results from the genetic algorithm optimization module 2, receive the abnormal data and alarm information from the sensor anomaly management module 3; the communication unit also sends instructions to each module to achieve the control and management of the system. Send adjustment instructions to the sensor hardware module 1, and send sensor data to the GIS abnormal point visualization module 6 and the user interface and operation module 7 for display.

[0090] The scheduling and control unit is implemented through a central control platform, such as a cloud platform or an edge computing platform. The scheduling and control unit is used to generate adjustment instructions based on the optimization results fed back by the genetic algorithm optimization module 2 received through the communication unit. The genetic algorithm optimization module 2 obtains a better solution through the optimization of sensor layout and parameter configuration. The scheduling and control unit generates specific adjustment instructions according to these optimization results, such as adjusting the position coordinates of the sensors, changing the sensitivity settings of the sensors, etc., and sends these instructions to each sensor node through the communication unit, so that the sensors can work according to the new configuration, thereby improving the monitoring effect and performance of the system.

[0091] The scheduling and control unit also controls the corresponding sensor nodes through the abnormal data and alarm information received through the communication unit. When abnormal data is received, the scheduling and control unit will take corresponding measures according to the type of abnormality. If the received abnormal data is hardware failure and battery abnormality, the scheduling and control unit needs to control the genetic algorithm optimization module 2 to perform optimization to adjust the sensor layout and parameter configuration. This is because hardware failure or battery abnormality may affect the normal operation of the sensors. Through the optimization of the genetic algorithm optimization module, the sensor layout can be re-planned and the parameter configuration can be adjusted to make up for the impact of the faulty sensors and improve the overall stability and reliability of the system. For other types of abnormal data, such as data drift abnormality or data missing abnormality, the data compensated by the sensor abnormality management module 3 is sent to the GIS abnormal point visualization module 6 and the user interface and operation module 7 for display.

[0092] In one embodiment, the GIS abnormal point visualization module 6 first closely associates the position information of the sensors with the geographic coordinate system. Each sensor has its specific installation position in the actual physical environment. By obtaining the precise position data of the sensors (such as longitude and latitude coordinates, etc.) and matching them with the corresponding positions in the geographic coordinate system, the corresponding relationship between the sensor positions and the geographical space is established. Maps or 3D models are used to visually display the distribution of the sensors and their health status. The map can clearly present the layout of the sensors in different regions, and users can clearly see information such as which regions have sensors installed and the density distribution of the sensors at a glance. The 3D model can provide a more three-dimensional and realistic display effect. For some complex geographical environments or scenarios with special spatial structures, the 3D model can more accurately reflect the actual installation position of the sensors and the relationship with the surrounding environment. A front-end framework is used to implement the real-time display function. These front-end frameworks have powerful map drawing and interaction functions. Front-end frameworks such as Leaflet, OpenLayers or ArcGIS API are used; when a sensor fails, this module will identify the position of the abnormal sensor on the map or 3D model. The identification methods can be various. For example, different colored icons can be used to represent different types of abnormalities, or the position of the abnormal sensor can be highlighted by flashing, highlighting, etc., so that users can quickly discover and locate the problem sensor and take corresponding measures for processing in a timely manner.

[0093] In one embodiment, as Figure 3 , the display and operation module 7 includes:

[0094] A control panel, whose core function is to display various key data in real time, including sensor data (such as various environmental parameters such as temperature, humidity, pressure, flow, etc.), sensor status (normal, faulty, abnormal, etc.), optimization results (the optimized sensor layout and parameter configuration plan obtained by the genetic algorithm optimization module), abnormal data (relevant data characteristics when the sensor has problems), and alarm information (such as abnormal type, abnormal sensor position, occurrence time, etc.). By displaying this information in real time, operators can comprehensively and timely understand the operation status of the system and provide a basis for decision-making. A Web or mobile-based control panel is used and displayed with the help of modern front-end frameworks (such as React, Vue or Angular).

[0095] A configuration management unit is used to obtain the sensor configuration parameters input by the user and send them to the central control and scheduling module to control each sensor node to make adjustments. Users can manually input various configuration parameters of the sensor according to actual needs, such as working mode (continuous monitoring, timed monitoring, etc.), location (adjust the installation coordinates of the sensor), sampling frequency (set the time interval for data acquisition), etc. The configuration management unit accurately transmits these parameters to the central control and scheduling module 5, and the central control and scheduling module 5 further controls each sensor node to make corresponding adjustments. The configuration management unit allows users to manually intervene in the system settings, providing users with the ability to flexibly adjust the system. To facilitate user operation, this module provides an intuitive configuration interface. In terms of interface design, a simple and clear layout and interaction method are adopted, enabling users to easily find the parameters that need to be set and make corresponding modifications. For example, through controls such as sliders, drop-down menus, and text input boxes, users can intuitively set the sampling frequency of the sensor, select the working mode, etc., reducing the operation difficulty of users and improving the configurability and usability of the system.

[0096] In one embodiment, the modules can communicate with each other through a variety of wireless and wired communication protocols; among them, the variety of wireless communication protocols include: multiple of LoRa, ZigBee, MQTT, HTTP, CoAP, Bluetooth Mesh, and Wi-Fi 6. Among them, Bluetooth Mesh supports local area network communication for low-power devices and is suitable for resource-constrained devices, while Wi-Fi 6 provides the advantages of high bandwidth and low latency, meeting high data transmission requirements, especially in harsh climate and remote environments, ensuring that the data transmission of sensors is not interfered, and improving the overall system stability. The variety of wired standard communication protocols include: I2C, SPI, CAN bus, etc., realizing real-time interaction between hardware control and sensors, and supporting remote adjustment and data transmission.

[0097] To better describe the above intelligent sensor anomaly management system based on the genetic algorithm, the following specific embodiments are now combined for illustration.

[0098] Embodiment 1: An intelligent sensor anomaly management system based on the genetic algorithm.

[0099] The system includes: a sensor hardware module, a genetic algorithm optimization module, a sensor anomaly management module, a data acquisition and processing module, a central control and scheduling module, a GIS anomaly point visualization module, and a user interface and operation module;

[0100] The working process of the system includes:

[0101] At system startup, the Central Control and Scheduling Module communicates with the Sensor Hardware Module. Based on the target coverage range of the sensors and environmental characteristics, through a preset algorithm or a preliminary configuration specified manually, it determines the initial positions of the sensors at multiple key positions in the preset orbit network, and sets initial parameters such as the initial sampling frequency and sensitivity. Subsequently, the Central Control and Scheduling Module transmits the preliminary configuration results to the Sensor Hardware Module via wireless communication (such as Wi-Fi or LoRa) or wired communication (such as industrial Ethernet), completing the preliminary deployment of the sensors.

[0102] After the preliminary deployment is completed, the Genetic Algorithm Optimization Module starts the optimization calculation. Through fitness evaluation, selection, crossover, and mutation operations, it continuously adjusts and optimizes the positions and configuration parameters of the sensors according to real-time data and environmental changes, and transmits the optimal deployment plan to the Central Control and Scheduling Module. The Central Control and Scheduling Module then sends adjustment instructions to the Sensor Hardware Module to achieve dynamic adjustment. The Sensor Hardware Module adjusts the positions or parameters of the sensors through servo motors or electric drives according to the received instructions to ensure the optimal deployment. The optimized configuration results are transmitted to the Sensor Hardware Module via wireless communication (such as Wi-Fi) or a wired network (such as RS485).

[0103] During system operation, the Sensor Anomaly Management Module monitors the health status of the sensors in real time. Once a hardware fault (such as low battery power, hardware damage) or data anomaly (such as drift, missing) is detected, this module will automatically trigger an adjustment mechanism. For hardware faults, it triggers a position adjustment; for data anomalies, it triggers an interpolation algorithm or parameter reconfiguration to ensure the normal operation of the sensors.

[0104] The status and anomaly information of all sensors are displayed in a Geographic Information System (GIS) through the GIS Anomaly Point Visualization Module. The GIS Visualization Module receives the real-time data, positions, and health status information of the sensors via wireless communication (such as Wi-Fi) or a wired network (such as Ethernet) and displays them in real time on the map. This enables operation and maintenance personnel to intuitively view the operating status and coverage area of each sensor. Especially when an anomaly occurs, they can quickly identify the location of the problem sensor with the help of map positioning, facilitating timely repair or adjustment in cooperation with drones.

[0105] Through the combination of the preliminary deployment and subsequent optimization, supplemented by GIS visualization, the system can ensure that the sensors cover key areas, and at the same time continuously optimize the layout of the sensors with the help of the genetic algorithm, thereby improving the measurement accuracy, fault troubleshooting efficiency, and system reliability. The optimization results and sensor status are transmitted between modules through a wireless or wired communication network, ensuring the high-efficiency operation and dynamic adjustment capabilities of the system.

[0106] The present invention has the following advantages compared with the prior art:

[0107] 1. Improve measurement accuracy and reliability

[0108] The present invention uses a genetic algorithm to optimize the position and sampling parameters of sensors, which can ensure that the sensor layout achieves the best coverage effect, reduce errors in data acquisition, and significantly improve measurement accuracy. In addition, automatically adjusting parameters such as sensitivity and frequency enables the sensors to adapt to different environmental conditions, thereby enhancing the reliability of the system and the quality of data, and ensuring compliance with the high-quality requirements for carbon emission data in ISO14064. The adjustable hardware configuration of the sensors (such as position, sensitivity, and range) combined with the optimization results of the genetic algorithm can automatically adapt to environmental changes, avoid hysteresis problems caused by human intervention, and enhance system reliability.

[0109] 2. Fault tolerance and enhanced system robustness

[0110] When a sensor in the present invention fails (such as abnormal data, low battery power, hardware damage, etc.), it automatically makes adjustments to avoid measurement deviations and data loss caused by faulty sensors. This self-repair ability significantly improves the fault tolerance of the system, ensures that the system can still operate continuously when a sensor fails, reduces the need for manual intervention and repair, and meets the requirement of ISO 14064 for the automated repair ability of the exception management module. When a sensor fails, the system can automatically adjust its position or parameter settings, avoiding long-term error accumulation caused by waiting for manual repair, and further improving the robustness of the system and the continuity of data.

[0111] 3. Reduce operation and maintenance costs and improve efficiency

[0112] With the help of the genetic algorithm and intelligent exception management in the present invention, the system realizes automated layout, adjustment, and fault repair, without frequent manual intervention, significantly reducing maintenance costs. When the system detects sensor anomalies or low battery power, it will automatically trigger an alarm to notify maintenance personnel for timely inspection, which can not only avoid human errors but also greatly reduce the cost and workload of manual inspections, improving the overall operation and maintenance efficiency.

[0113] 4. Optimize resource allocation and strong adaptability

[0114] The present invention adapts to environmental changes such as weather and seasons by dynamically adjusting the sensor position, overcomes the limitations of fixed sensor layouts, and ensures optimal measurement effects under complex conditions. The system can respond in real time to external conditions such as environmental changes and equipment status changes, and adjust the working state of the sensors according to the feedback. This enables the system to operate stably for a long time and maintain high accuracy even in complex working environments or abnormal situations.

[0115] 5. Enhance system visualization and operation convenience

[0116] The present invention visualizes the location and status information of sensors through GIS technology, enabling intuitive viewing of the sensor layout and fault points, and providing immediate system health status and fault location information for operators. This function not only improves the problem diagnosis efficiency but also helps maintenance personnel quickly locate and handle sensor faults, reducing the blindness in the traditional operation and maintenance mode. By providing an easy-to-operate control panel, users can conveniently view the sensor status, optimization results, and abnormal alarms, further simplifying the system management and operation processes.

[0117] 6. Improve the reliability and quality of data

[0118] The present invention processes the collected data through the data acquisition and preprocessing module, such as cleaning and noise reduction, to ensure that the quality of the uploaded data meets the predetermined standards. At the same time, the optimization feedback mechanism of the genetic algorithm can dynamically adjust the sensor configuration, further improving the reliability and accuracy of the data. Through data fusion technology, the data from multiple sensors are integrated to improve the reliability and coverage of the data, avoiding data loss or deviation caused by the failure of a single sensor, and ensuring stable and accurate measurement results obtained by the system through the cooperation of multiple sensors.

[0119] 7. System scalability and the ability to adapt to different scenarios

[0120] The independence and scalability of each module of the system of the present invention enable the system to flexibly adapt to different application scenarios. Whether it is in carbon emission monitoring, environmental pollution monitoring, or in the fields of agriculture, industry, etc., the system can flexibly adjust the configuration according to the needs to meet the requirements of different monitoring tasks. The system supports multiple wireless and wired communication protocols (such as LoRa, ZigBee, MQTT, Bluetooth Mesh, and Wi-Fi 6, etc.), enabling it to adapt to the application requirements of different regions and scenarios. Under conditions such as remote monitoring, harsh environments, and complex terrains, the system can ensure stable data transmission by selecting the appropriate communication protocol. At the same time, combining the multi-hop characteristics of Bluetooth Mesh and the high bandwidth and high concurrency capabilities of Wi-Fi 6, the system can still ensure signal stability in complex environments, meeting the needs of large-scale device interconnection and efficient data acquisition. The system can dynamically adjust the configuration of each module and automatically optimize communication and data acquisition parameters according to the actual environmental changes, such as the working frequency and sampling accuracy of sensors. By combining deep learning and reinforcement learning algorithms, the system can optimize the sensor layout and communication strategy in real time according to the environmental data, achieving a more efficient scenario adaptation ability.

[0121] In summary, for the intelligent sensor anomaly management system based on the genetic algorithm of the present invention, the genetic algorithm optimization module optimizes the sensor positions and parameter configurations. The central control and scheduling module sends adjustment instructions to the sensor hardware module according to the optimization results to adjust the layouts and parameters of the sensors. At the same time, the sensor anomaly management module detects sensor anomalies in real time and automatically selects the optimal adjustment strategy. In addition, the status and anomaly information of all sensors will be displayed in the geographic information system (GIS) through the GIS visualization module, so as to dynamically adjust the layout and quickly locate the anomaly points, cooperate with the unmanned aerial vehicle to achieve automatic repair, and improve the reliability and maintenance efficiency. The present invention integrates advanced technologies such as the Internet of Things, edge computing, deep learning, GIS, genetic algorithm, and reinforcement learning to build an intelligent and automated sensor management and optimization platform to achieve efficient data collection, real-time optimization, and fault management. With the modular design, the system has high scalability and flexibility and is applicable to various environmental monitoring application scenarios. Therefore, the present invention effectively overcomes various disadvantages in the prior art and has high industrial utilization value.

[0122] The above embodiments merely exemplarily illustrate the principles and effects of the present invention, rather than limiting the present invention. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes completed by those with ordinary knowledge in the technical field without departing from the spirit and technical ideas disclosed by the present invention should still be covered by the claims of the present invention.

Claims

1. An intelligent sensor anomaly management system based on a genetic algorithm, characterized in that, The system includes: a sensor hardware module, a genetic algorithm optimization module, a sensor anomaly management module, a data acquisition and processing module, a central control and scheduling module, a GIS anomaly point visualization module, and a user interface and operation module; among which, The sensor hardware module is built-in with multiple sensors at movable positions, and is used for collecting environmental data; The data acquisition and processing module is used for preprocessing the environmental data collected by each sensor, and real-time feedback the preprocessed environmental data to the genetic algorithm optimization module, the central control and scheduling module, and the sensor anomaly management module; The genetic algorithm optimization module is used for optimizing the layout and parameter configuration of each sensor based on the genetic algorithm to obtain an optimization result, and feedback it to the central control and scheduling module; The central control and scheduling module is used for sending adjustment instructions to the sensor hardware module based on the optimization result feedback by the genetic algorithm optimization module to adjust the layout and parameter configuration of each sensor; The sensor anomaly management module is used for detecting the anomaly state of each sensor, automatically triggering an adjustment mechanism when an anomaly is detected, and sending the anomaly data and alarm information to the central control and scheduling module for corresponding control; The GIS anomaly point visualization module is used for associating the position information of each sensor with the geographic coordinate system, displaying the sensor distribution and its health status using a map or a 3D model, and identifying the sensors detected with anomalies; The user interface and operation module is used for displaying the system monitoring data and obtaining the sensor configuration parameters input by the user.

2. The intelligent sensor anomaly management system based on a genetic algorithm according to claim 1, wherein The sensor hardware module includes: multiple sensor nodes; among which, each sensor node includes: A sensor, which is used for collecting environmental data; A driving device, which is used for adjusting the position of the sensor based on an adjustment instruction; A sensor parameter adjustment interface, which is used for adjusting the sampling parameters of the sensor based on an adjustment instruction; A battery monitoring unit, which is used for monitoring the battery state of the sensor.

3. The intelligent sensor anomaly management system based on a genetic algorithm according to claim 1, wherein The genetic algorithm optimization module includes: An initial population unit, which is used for setting the initial configuration parameters of the sensor using a randomization method according to the initial layout of each sensor, extracting the correlation features between the initial configuration parameters and the environmental data based on a deep learning algorithm, and generating an initial population according to the correlation features; A fitness evaluation unit, which is used for evaluating the fitness of the sensor layout scheme using a fitness function; A selection, crossover and mutation unit, which is used for selecting excellent individuals based on the evaluated fitness, performing crossover and mutation operations on the selected excellent individuals to generate a new population, and obtaining the optimal sensor layout scheme; An optimization feedback unit, which is used for adjusting the optimal sensor layout scheme based on the feedback mechanism of reinforcement learning.

4. The intelligent sensor anomaly management system based on the genetic algorithm according to claim 1, wherein The data acquisition and processing module includes: A data acquisition unit, which is used for collecting the environmental data collected by each sensor in real time; A data preprocessing unit, connected to the data acquisition unit, which is used for preprocessing the collected environmental data and sending the processed environmental data outward.

5. The intelligent sensor anomaly management system based on a genetic algorithm according to claim 1, wherein The sensor anomaly management module includes: Anomaly detection unit, used to perform anomaly detection on each sensor based on the environmental data real-time feedback by the data acquisition and processing module, and output the anomaly type when an anomaly is detected; among them, the anomaly types include: hardware failure and battery anomaly, data drift anomaly, and data missing anomaly; Automatic adjustment unit, used to trigger the corresponding adjustment mechanism according to the anomaly type; Alarm and notification unit, used to send the alarm information and anomaly data to the central control and scheduling module when an anomaly is detected.

6. The intelligent sensor anomaly management system based on the genetic algorithm according to claim 5, characterized in that, The triggering of the corresponding adjustment mechanism according to the anomaly type includes: If the anomaly type of the sensor is hardware failure and battery anomaly, activate the standby sensor node and feedback to the central control and scheduling module to control the genetic algorithm optimization module for optimization, so as to adjust the layout and parameter configuration of the sensor; If the anomaly type of the sensor is data drift anomaly, use the time series algorithm to predict and compensate for the data drift; If the anomaly type of the sensor is data missing anomaly, use multi-node cooperation and deep learning algorithm to supplement the missing data.

7. The intelligent sensor anomaly management system based on a genetic algorithm according to claim 1, wherein The central control and scheduling module includes: a scheduling and control unit and a communication unit connected to the scheduling and control unit; The scheduling and control unit is used to generate adjustment instructions based on the optimization results feedback by the genetic algorithm optimization module received through the communication unit, and send the adjustment instructions to each sensor node through the communication unit; it is also used to control the corresponding sensor node based on the anomaly data and alarm information received through the communication unit.

8. The intelligent sensor anomaly management system based on a genetic algorithm according to claim 1, characterized in that The display and operation module includes: Control panel, used to display the sensor data, sensor status, optimization results, anomaly data, and alarm information in real time; Configuration management unit, used to obtain the sensor configuration parameters input by the user and send them to the central control and scheduling module to control each sensor node for adjustment.

9. The intelligent sensor anomaly management system based on the genetic algorithm according to claim 2, wherein The driving device includes: a servo motor and an electric driver, which cooperate with a preset track to automatically adjust the position of the corresponding sensor.

10. The intelligent sensor anomaly management system based on a genetic algorithm according to claim 1, wherein Each module can communicate through a variety of wireless and wired communication protocols; among them, the variety of wireless communication protocols include: multiple of LoRa, ZigBee, MQTT, HTTP, CoAP, Bluetooth Mesh, and Wi-Fi 6.

Images