Port environment detection system based on Internet of Things technology

Abstract

The invention relates to the technical field of port environment detection, in particular to a port environment detection system based on the Internet of Things technology, and the system comprises a wind current driving type environment sensing modeling module which collects data such as wind speed, air pressure and particulate matter concentration, and constructs a wind disturbance correction atlas; the dynamic disturbance perception and correction module identifies and corrects an abnormal region based on a wind disturbance mapping function, and outputs a dynamic environment state set containing trust weights and risk factors; and the energy-sensing fusion scheduling decision-making module is combined with the wind power energy storage state to generate a dynamic regulation and control strategy of the sampling frequency, the power supply priority and the communication window so as to realize energy-sensing coupling scheduling. According to the invention, through wind disturbance correction modeling, dynamic data correction and energy-sensing fusion scheduling, high-credibility collection, intelligent restoration and resource optimization configuration of port environment monitoring data are realized, and the monitoring precision and adaptive ability of the system are significantly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of port environment detection, and particularly relates to a port environment detection system based on Internet of Things technology. BACKGROUND

[0002] In the process of port micro-powder loading operation, a large amount of fine particles are easy to diffuse under the disturbance of wind force, forming a local high-concentration suspended area, which poses a significant threat to the port operation environment and personnel health. In order to reduce the pollution risk brought by dust diffusion, an environmental sensor is usually deployed at the air outlet of the dust removal system to monitor key parameters such as particle concentration, wind speed, air pressure, and the like, and a wind energy recovery technology is used to support the power supply and lighting system of the equipment. How to realize high-trust collection of sensing data and efficient scheduling of wind power resources has become a key problem to ensure the stable operation of the system.

[0003] In the existing dust removal system, environmental sensing and wind power driving are usually designed independently, without fully considering the interference of wind flow disturbance on environmental sensing data, resulting in frequent local abnormal monitoring values. At the same time, the energy supply of the wind power system is usually operated with a fixed strategy, which cannot be dynamically adjusted according to the real-time energy storage state and the importance of environmental sensing tasks, and is prone to problems such as insufficient power supply in high-risk areas or waste of resources in low-risk areas. In addition, the traditional method lacks a coupling modeling mechanism between wind energy utilization and environmental data, making it difficult to support refined and partitioned resource regulation strategies. SUMMARY

[0004] The present application provides a port environment detection system based on Internet of Things technology, which corrects monitoring data anomalies by constructing a wind disturbance correction map, realizes high-confidence dynamic environmental state sensing, further combines the energy storage state of the wind power system and the priority of the sensing task, constructs a multi-objective scheduling strategy set, and realizes optimal resource allocation between sensing reliability and energy consumption limitation, thereby improving the environmental sensing reliability and wind power resource utilization efficiency of the port dust removal system.

[0005] The port environment detection system based on Internet of Things technology comprises a wind flow driven environmental sensing module, a dynamic disturbance sensing and environmental data correction module, and an energy sensing integrated scheduling decision module, wherein: The wind flow driven environmental sensing modeling module collects environmental data near the air outlet of the micro-powder loading dust removal system during operation, including wind speed, air pressure, suspended particle concentration, and wind energy utilization state, and constructs a wind disturbance correction map reflecting the influence degree of wind flow disturbance on environmental data based on a spatial disturbance sensing model; The dynamic disturbance perception and environment data correction module receives the wind disturbance correction map and real-time collected environment data, identifies the local monitoring abnormal area caused by the operation of the wind energy recovery device by constructing a wind flow disturbance mapping function, and uses a disturbance compensation correction algorithm to regionally correct the target monitoring parameter, and outputs a dynamic environment state set including a trust weight and a risk factor; The disturbance response mapping function is constructed based on the wind disturbance correction map and the real-time environment data at the current moment, the disturbance intensity of the wind disturbance correction map and the disturbance response mapping function between the environment data are constructed through the establishment of the space-time joint distribution relationship, the disturbance abnormality degree of the monitoring points in the port operation space is analyzed by receiving the result of the disturbance response mapping function, the area affected by the operation of the wind energy device is identified, the abnormal response area dominated by the wind disturbance is extracted through the calculation of the dynamic residual field of the monitoring value deviating from the disturbance expected value, and the spatial continuity constraint is combined, and the local abnormal area label, the residual intensity and the time sequence change trend are output.

[0006] Optionally, the wind flow driven environment perception modeling module comprises: Multi-source environment data synchronous collection: in the operation process of the micro-powder loading and unloading dust removal system, distributed multi-type sensing nodes are arranged downstream of the air outlet to collect environment data, including local wind speed , local air pressure , suspended particulate matter concentration , wind energy utilization state , the position coordinates of each sensing node are set as , and the corresponding time slice is recorded, and a multi-time and space environment data set is constructed; Space disturbance factor field construction: the collected environment data is mapped to a local space area, the disturbance factor of each sampling point is calculated, and the disturbance factor field of each node is output; Wind disturbance correction map construction: based on the disturbance factor field , the spatial disturbance degree is interpolated and expanded to form a wind disturbance correction map under continuous space .

[0007] Optionally, the dynamic disturbance perception and environment data correction module comprises: Disturbance response mapping construction: based on the wind disturbance correction map and the real-time environment data at the current moment, the disturbance response mapping function between the disturbance intensity of the wind disturbance correction map and the environment data is constructed through the establishment of the space-time joint distribution relationship; Local abnormality perception area identification: receiving the result of the disturbance response mapping function, the disturbance abnormality degree of the monitoring points in the port operation space is analyzed, the area affected by the operation of the wind energy device is identified, the abnormal response area dominated by the wind disturbance is extracted through the calculation of the dynamic residual field of the monitoring value deviating from the disturbance expected value, and the spatial continuity constraint is combined, and the local abnormal area label, the residual intensity and the time sequence change trend are output. Disturbance compensation and state reconstruction: Regional correction and state reconstruction are performed on environmental data within the identified local anomaly areas. Using the wind disturbance correction map as the weight basis and combined with the dynamic residual fitting results, a disturbance compensation and repair algorithm is executed to perform noise reduction, interpolation, or confidence scoring on pollution values. The final output dynamic environmental dataset includes the corrected environmental data for each monitoring point, the corresponding trust weight factor, and the disturbance risk factor.

[0008] Optionally, the disturbance response mapping construction includes: Spatiotemporally aligned feature sample construction: Extract the current time from all sensor nodes in the same port operation area. The spatial position below And obtain nodes from the wind disturbance correction map. Location disturbance intensity Obtain nodes from real-time environmental data monitoring values and construct a sample set ; Construction of the disturbance response mapping function: for each monitoring dimension in the sample set Kernel-weighted regression was used to construct a perturbation strength For input, environmental data The output is a perturbation response mapping function, and finally, a set of multidimensional perturbation response mapping functions. .

[0009] Optionally, the identification of the local anomaly sensing region includes: Perturbation prediction generation and bias estimation: utilizing a set of multidimensional perturbation response mapping functions from the output. The intensity of the disturbance to each current sensing node Input the data to obtain the predicted value for each monitoring dimension. And combined with the actual monitoring values ​​of the nodes Calculate its disturbance deviation residual ; Constructing a dynamic residual field and spatial anomaly measurement: Calculate the sum of squared disturbance residual intensities of all nodes across all monitoring dimensions, and construct a node-level comprehensive disturbance anomaly value. Based on node location Will Spatial mapping into dynamic residual field ; Anomaly response region extraction and label output: in dynamic residual field Set the disturbance anomaly threshold above Identify all that meet the requirements The spatial location constitutes a set of potential anomalies. Based on spatial continuity constraints Perform connected component clustering to extract multiple local anomaly response regions. and for each local anomaly response patch Output the corresponding tag ID Average residual strength The residual trend curve over continuous time intervals Ultimately, this forms a regional set that includes anomaly response information from multiple areas. .

[0010] Optionally, the disturbance compensation and state reconstruction include: Weighted modeling of disturbance effects at anomaly monitoring points: For each identified local anomaly response region Extract each monitoring point Raw environmental observations Furthermore, a repair reference function based on disturbance intensity weighting is constructed to calculate the disturbance compensation reconstruction value; Dynamic residual scoring and trust weight generation: Calculate the residual deviation for each monitoring point. Assign a confidence value to the monitoring point, and combine the confidence levels of each monitoring point across all dimensions into a disturbance risk factor. ; Output dynamic environment state set: for each monitoring point Output the corrected dynamic environment state vector Finally, the reconstructed dynamic environment dataset of the entire anomaly response area is output. The reconstruction results of all abnormal response areas are then merged into a corrected environmental state set. .

[0011] Optionally, the sensing fusion scheduling decision module includes: Wind power-environmental state fusion modeling: Receive dynamic environmental state set and current energy storage state data of wind power system, extract trust weight factor and disturbance risk factor of each monitoring point, and combine the state of charge, voltage level and remaining available power of wind power battery to construct energy-sensing coupled state matrix. According to the environmental state level and wind power output capacity, calculate the current allocable energy resources and sensing task priority relationship to form the basis of dynamic resource-task adaptation under wind power drive. Multi-objective dynamic control strategy generation: Based on the energy-sensing coupled state matrix, a multi-objective scheduling function is constructed that simultaneously considers the data perception value and energy consumption constraints, generating adaptive control strategies, including sampling frequency scheduling, power supply priority ranking, and communication window allocation. The adaptive control strategies are dynamically adjusted according to the current operating strategy level and sudden environmental disturbances to achieve optimized allocation of different functional resources. Strategy execution and command output: The generated adaptive control strategy is transformed into issued device control commands, which include sensor sampling control, lighting equipment power supply switching, and communication unit activation plan. These commands are then synchronously distributed to edge nodes via IoT communication protocols (MQTT / LoRaWAN).

[0012] Optionally, the wind power-environmental state fusion modeling includes; State data synchronization and feature extraction: from corrected environment state sets Extract each monitoring point Corrected environmental data vector Trust weight factor vector Disturbance risk factors Simultaneously, the battery state of charge is obtained from the current energy storage state data of the wind power system. Battery terminal voltage Remaining available power Formation time Multi-source state sample set ; Energy-sensory coupled state matrix construction: Based on the confidence level, risk factor, and wind power availability of each monitoring point, calculate the state matrix for each node. Perception priority indicators Define the share of allocable power in the current wind power system. And construct the sensing coupling state matrix ; Dynamic resource-task adaptation calculation: based on the energy-sensory coupling state matrix Calculate the adaptation coefficient between the node's sensing task and the available energy resources at the current moment. And according to the environmental status level This forms the basis for dynamic resource-task adaptation. .

[0013] Optionally, the generation of the multi-objective dynamic control strategy includes: Construct a set of node perception performance indicators: for each monitoring point Based on its environmental status level Trust factor Disturbance risk factors and current battery status Construct its perceived effectiveness indicators With energy tolerance factor ; Constructing a multi-objective scheduling function: Calculating monitoring points using a multi-objective scheduling function. The scheduling priority score is used to achieve the dual objectives of maximizing perceived value and minimizing energy consumption costs. Control strategy generation: Based on the scheduling priority score of each monitoring point, an adaptive control strategy is generated, including sampling frequency scheduling (reducing the frequency in areas with high confidence), power supply priority ranking (prioritizing power supply to high-risk areas), and communication window allocation (adjusting the data reporting cycle). Policy level and disturbance adaptive update: based on the current running policy level and sudden disturbance factors , dynamically adjust adaptive regulation strategies.

[0014] Optionally, the strategy execution and instruction output include: Strategy instruction structured encoding: Map each control parameter in the adaptive control strategy to standardized equipment instruction fields to construct the instruction data packet structure; Control command allocation and packaging: based on the perception node ID and the current operating policy level. Construct a scheduling instruction mapping table and complete the packaging of device instructions; Protocol selection and command issuance: Select the IoT communication protocol based on the node's communication capabilities and the environmental topology to complete the broadcasting or targeted issuance of control commands.

[0015] The beneficial effects of this invention are: This invention, by constructing a wind-driven environmental perception modeling module, can fully utilize the structural information generated by wind disturbance during the operation of the micro-powder loading and dust removal system. It enables dynamic acquisition of multi-dimensional environmental elements such as wind speed, air pressure, particulate matter concentration, and wind energy utilization status, as well as the generation of wind disturbance correction maps. This effectively solves the problems of sensor reading distortion, data anomalies, and spatial discontinuities caused by operational disturbances in traditional port environmental monitoring, thereby improving the spatiotemporal accuracy and reliability of environmental monitoring data.

[0016] This invention introduces a dynamic disturbance perception and environmental data correction module, constructs a wind disturbance response mapping function and a dynamic residual field, identifies local abnormal areas and realizes regional data repair and state reconstruction. Thus, without relying on a large amount of manual screening and redundant equipment deployment, it can achieve intelligent identification, noise reduction and completion of sensor errors, and produce a dynamic environmental state set with trust weights and risk factors.

[0017] This invention constructs an energy-sensing fusion scheduling and decision-making module that integrates energy information such as the state of charge, voltage level, and remaining power of the wind power system. By combining the risk factors, trust weights, and sensing task priorities of the current sensing network, it achieves multi-objective adaptive strategy generation and control command issuance for sampling frequency, power supply sequencing, and communication windows. This ensures that, under the background of dynamic fluctuations in wind power resources, priority monitoring of high-risk areas, compensatory data collection of low-reliability points, and energy efficiency optimization and sensing quality coupled scheduling of the entire system are achieved, thereby enhancing the responsiveness, robustness, and operational intelligence level of the port environmental monitoring system. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Fig. 1 This is a schematic diagram of the system functional modules according to an embodiment of the present invention; Fig. 2 This is a schematic diagram of the dynamic disturbance sensing and environmental data correction module in an embodiment of the present invention. Detailed Implementation

[0020] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Those skilled in the art may employ other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0021] like Figs. 1-2 As shown, the port environmental monitoring system based on Internet of Things (IoT) technology includes a wind-driven environmental sensing module, a dynamic disturbance sensing and environmental data correction module, and an energy-sensing fusion scheduling and decision-making module, wherein; The wind-driven environmental perception modeling module collects environmental data near the air outlet during the operation of the micro powder loading and dust removal system, including wind speed, air pressure, suspended particulate matter concentration and wind energy utilization status. Based on the spatial disturbance perception model, it constructs a wind disturbance correction map that reflects the degree of influence of wind flow disturbance on environmental data. The dynamic disturbance perception and environmental data correction module receives wind disturbance correction maps and real-time environmental data. By constructing a wind flow disturbance mapping function, it identifies local monitoring anomalies caused by the operation of the wind energy recovery device, and uses a disturbance compensation correction algorithm to regionally correct the target monitoring parameters, outputting a dynamic environmental state set including trust weights and risk factors. The energy sensing fusion scheduling decision module receives dynamic environmental state sets and current wind power system energy storage state data. Based on wind power load adaptation rules and environmental priority level system, it constructs a dynamic control strategy set for scheduling sampling frequency, power supply priority and communication window, realizing coupled scheduling between wind power status and environmental sensing quality.

[0022] The wind-driven environmental perception modeling module includes: Multi-source environmental data synchronous acquisition: During the operation of the micro-powder loading and dust removal system, distributed multi-type sensor nodes are deployed downstream of the air outlet to collect environmental data, including local wind speed. Local air pressure Suspended particulate matter concentration Wind energy utilization status The position coordinates of each sensor node are set as follows: And record the corresponding time segment. Constructing a multi-temporal and spatiotemporal environment dataset , is represented as: ; in, This represents the number of sensor nodes. Spatial perturbation factor field construction: The collected environmental data is mapped to a local spatial region, and the perturbation factor at each sampling point is calculated. This is used to measure the degree of interference of wind flow on the stability of sensing data and outputs the perturbation factor field for each node. , is represented as: ; in, For nodes The spatial gradient of wind speed in the vicinity reflects the degree of drastic change in local wind flow. For nodes The spatial pressure gradient in the vicinity, For nodes The rate of change of suspended particulate matter concentration over time , , These are the corresponding perturbation weight coefficients; Construction of wind disturbance correction map: based on disturbance factor field The spatial disturbance level is interpolated and extended to form a wind disturbance correction map in continuous space. , is represented as:

[0023] in, For any three-dimensional spatial location, For spatial interpolation weights, The kernel function (Gaussian kernel) is used for smooth weighting. For nodes With the target point The Euclidean distance.

[0024] The dynamic disturbance sensing and environmental data correction module includes: Disturbance response mapping construction: Based on the wind disturbance correction map and the real-time environmental data at the current moment, a disturbance response mapping function between the disturbance intensity of the wind disturbance correction map and the environmental data is constructed by establishing a spatial-temporal joint distribution relationship; Local anomaly perception area identification: Receive the results of the disturbance response mapping function, analyze the degree of disturbance anomaly of the monitoring points in the port operation space, identify the areas affected by the operation of wind power devices, calculate the dynamic residual field of the monitoring value deviating from the expected disturbance value, combine spatial continuity constraints, extract the wind-dominated anomaly response area, and output the local anomaly area label, residual intensity and temporal change trend. Disturbance compensation and state reconstruction: Regional correction and state reconstruction are performed on environmental data within the identified local anomaly areas. Using the wind disturbance correction map as the weight basis and combined with the dynamic residual fitting results, a disturbance compensation and repair algorithm is executed to perform noise reduction, interpolation, or confidence scoring on pollution values. The final output dynamic environmental dataset includes the corrected environmental data for each monitoring point, the corresponding trust weight factor, and the disturbance risk factor.

[0025] The construction of the disturbance response map includes: Spatiotemporally aligned feature sample construction: Extract the current time from all sensor nodes in the same port operation area. The spatial position below And obtain nodes from the wind disturbance correction map. Location disturbance intensity Obtain nodes from real-time environmental data monitoring values ,in, Indicates the first The monitoring values ​​of each monitoring dimension (wind speed, air pressure, suspended particulate matter concentration, and wind energy utilization status) were collected, and a sample set was constructed. ; Construction of the disturbance response mapping function: for each monitoring dimension in the sample set Kernel-weighted regression was used to construct a perturbation strength For input, environmental data The output is a perturbation response mapping function, and finally, a set of multidimensional perturbation response mapping functions. , is represented as: ; ; in, For a given disturbance strength Predictive environmental data at that time For kernel function (using , Dimension (core bandwidth) These are the local regression coefficients of the kernel regression.

[0026] Local anomaly detection region identification includes: Perturbation prediction generation and bias estimation: utilizing a set of multidimensional perturbation response mapping functions from the output. The intensity of the disturbance to each current sensing node Input the data to obtain the predicted value for each monitoring dimension. And combined with the actual monitoring values ​​of the nodes Calculate its disturbance deviation residual , is represented as: ; ; in, For sensor node indexing; Constructing a dynamic residual field and spatial anomaly measurement: Calculate the sum of squared disturbance residual intensities of all nodes across all monitoring dimensions, and construct a node-level comprehensive disturbance anomaly value. Based on node location Will Spatial mapping into dynamic residual field , is represented as: ; ; in, This is a spatial interpolation weighting function based on a Gaussian weight kernel. For spatial smoothing parameters; Anomaly response region extraction and label output: in dynamic residual field Set the disturbance anomaly threshold above Identify all that meet the requirements The spatial location constitutes a set of potential anomalies. Based on spatial continuity constraints Perform connected component clustering to extract multiple local anomaly response regions. and for each local anomaly response patch Output the corresponding tag ID Average residual strength The residual trend curve over continuous time intervals Ultimately, this forms a regional set that includes anomaly response information from multiple areas. , is represented as: ; ; in, This represents the number of local anomaly response regions; ; in, The average value of the residual field across all sensing nodes. The corresponding standard deviation, This is the abnormal intensity factor.

[0027] Disturbance compensation and state reconstruction include: Weighted modeling of disturbance effects at anomaly monitoring points: For each identified local anomaly response region Extract each monitoring point Raw environmental observations A repair reference function based on disturbance intensity weighting is constructed, and the disturbance compensation reconstruction value is calculated, expressed as: ; in, For monitoring points In dimensions Disturbance compensation reconstruction value This is the perturbation weighting function constructed based on the wind disturbance correction map. , These represent the disturbance intensity values ​​of the corresponding points in the wind disturbance correction map. These are the perturbation weight control parameters; Dynamic residual scoring and trust weight generation: Calculate the residual deviation for each monitoring point. Assign a confidence value to the monitoring point, and combine the confidence levels of each monitoring point across all dimensions into a disturbance risk factor. The larger the value, the higher the risk, expressed as: ; ; ; in, For the first Each monitoring indicator at the point Trust weight, The residual tolerance coefficient is used to control the degree of confidence decay. Total number of monitored dimensions; Output dynamic environment state set: for each monitoring point Output the corrected dynamic environment state vector Finally, the reconstructed dynamic environment dataset of the entire anomaly response area is output. The reconstruction results of all abnormal response areas are then merged into a corrected environmental state set. .

[0028] The sensing fusion scheduling decision module includes: Wind power-environmental state fusion modeling: Receive dynamic environmental state set and current energy storage state data of wind power system, extract trust weight factor and disturbance risk factor of each monitoring point, and combine the state of charge, voltage level and remaining available power of wind power battery to construct energy-sensing coupled state matrix. According to the environmental state level and wind power output capacity, calculate the current allocable energy resources and sensing task priority relationship to form the basis of dynamic resource-task adaptation under wind power drive. Multi-objective dynamic control strategy generation: Based on the energy-sensing coupled state matrix, a multi-objective scheduling function is constructed that simultaneously considers the data perception value and energy consumption constraints, generating adaptive control strategies, including sampling frequency scheduling, power supply priority ranking, and communication window allocation. The adaptive control strategies are dynamically adjusted according to the current operating strategy level and sudden environmental disturbances to achieve optimized allocation of different functional resources. Strategy execution and command output: The generated adaptive control strategy is transformed into issued device control commands, which include sensor sampling control, lighting equipment power supply switching, and communication unit activation plan. These commands are then synchronously distributed to edge nodes via IoT communication protocols (MQTT / LoRaWAN).

[0029] Wind power-environmental condition fusion modeling includes; State data synchronization and feature extraction: from corrected environment state sets Extract each monitoring point Corrected environmental data vector Trust weight factor vector Disturbance risk factors Simultaneously, the battery state of charge is obtained from the current energy storage state data of the wind power system. (0-1) Battery terminal voltage Remaining available power Formation time Multi-source state sample set , is represented as: ; in, This represents the total number of currently monitored nodes. Energy-sensory coupled state matrix construction: Based on the confidence level, risk factor, and wind power availability of each monitoring point, calculate the state matrix for each node. Perception priority indicators Define the share of allocable power in the current wind power system. And construct the sensing coupling state matrix , is represented as: ; ; ; in, For nodes In dimensions Trust weight, For environmental dimensions Importance weights The trust factor is the risk gain term after normalizing the disturbance risk factor. This is the reference voltage (calibrated value). For power adjustment factor, The sensing coupling state matrix In the middle, the corresponding monitoring point location The entire row vector information, that is, the joint state description of the monitoring point across all coupled attribute dimensions (such as trust weight factor, disturbance risk factor, state of charge, voltage level, remaining power supply capacity, etc.); Dynamic resource-task adaptation calculation: based on the energy-sensory coupling state matrix Calculate the adaptation coefficient between the node's sensing task and the available energy resources at the current moment. And according to the environmental status level This forms the basis for dynamic resource-task adaptation. , is represented as: ; Environmental Status Level Represented as: ; in, For monitoring points The perturbation residual, The global maximum disturbance residual, , , These are the corresponding weight parameters.

[0030] The generation of multi-objective dynamic control strategies includes: Construct a set of node perception performance indicators: for each monitoring point Based on its environmental status level Trust factor Disturbance risk factors and current battery status Construct its perceived effectiveness indicators With energy tolerance factor , is represented as: ; ; in, , , These are the corresponding weight coefficients. The battery is at full capacity. Constructing a multi-objective scheduling function: Calculating monitoring points using a multi-objective scheduling function. The scheduling priority score, in order to achieve the dual objectives of maximizing perceived value and minimizing energy consumption cost, is expressed as: ; in, For monitoring points The scheduling priority score is assigned, with higher values ​​indicating higher scheduling priority. , These are the corresponding target weight parameters; Control strategy generation: Based on the scheduling priority score of each monitoring point, an adaptive control strategy is generated, including sampling frequency scheduling (reducing the frequency in high-confidence areas), power supply priority ranking (prioritizing power supply to high-risk areas), and communication window allocation (adjusting the data reporting cycle). Specifically, this includes: (1) Sampling frequency scheduling : ; in, This is the system's basic sampling frequency. Confidence adjustment coefficient (e.g.) In high-confidence regions, the sampling frequency is reduced; (2) Power supply priority ranking : ; in, Indicates all The values ​​are sorted in descending order, and power is supplied to areas with high sensing efficiency first. (3) Communication window allocation : ; in, For the minimum communication cycle, To adjust the factors, high-risk areas ( (High) Shorten the communication window and increase the reporting frequency; Policy level and disturbance adaptive update: based on the current running policy level and sudden disturbance factors Dynamically adjust adaptive control strategies, specifically including: (1) Operational strategy level Division: ; ; in, The score is the rating of the operational strategy level. This represents the average environmental status level across all monitoring points. This represents the average of the disturbance risk factors across all monitoring points. This represents the average energy consumption tolerance factor for all monitoring points. , , These are the weight coefficients of the corresponding scoring functions. The high threshold for strategy level classification, The low threshold for classifying strategy levels; (2) Adjusting adaptive regulation strategies: If in energy-saving mode Then increase Strengthen energy consumption penalties; If the intensity of the disturbance increases ( ), then shorten Increase sampling frequency ,in, For monitoring points The intensity of the disturbance. The threshold for determining the intensity of the disturbance; Disturbance intensity determination threshold Represented as: ; in, , These represent the average and standard deviation of the disturbance intensity at the current monitoring point, respectively. The sensitivity coefficient is denoted as .

[0031] Strategy execution and command output include: Strategy instruction structured encoding: Mapping each control parameter in the adaptive control strategy to standardized device instruction fields, constructing the instruction data packet structure, specifically including: (1) Sampling frequency control strategy Mapped to Field; (2) Prioritize power supply Mapped to Control commands; (3) Set the communication window duration Mapped to parameter; (4) For each monitoring node Construct a set of control instructions: ; Control command allocation and packaging: based on the perception node ID and the current operating policy level. Construct a scheduling instruction mapping table and complete the packaging of device instructions, specifically including: (1) For each node Bind its unique address ; (2) Construct the packaging structure: ,in, For nodes The content of the control instructions to be executed. The timestamp generated for the instruction packet; (3) If the system is in a high-level strategy Then, high-risk region instruction packets are encapsulated first, among which, This represents the policy level threshold. Protocol selection and command issuance: Based on the node's communication capabilities and the environmental topology, an IoT communication protocol is selected to complete the broadcasting or targeted issuance of control commands, specifically including: (1) Communication protocol adaptation: Low-power wide-area deployment → Adopting LoRaWAN; High-speed LAN requirements → Use MQTT; (2) Execute the distribution operation: Use a unified scheduling platform to issue control command packets to the edge control gateway in batches; The gateway routes the instructions to the corresponding sensing node via the selected protocol.

[0032] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0033] 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. A port environmental monitoring system based on Internet of Things (IoT) technology, characterized in that: It includes a wind-driven environmental perception module, a dynamic disturbance perception and environmental data correction module, and an energy-sensing fusion scheduling and decision-making module, among which; The wind-driven environmental perception modeling module collects environmental data near the air outlet during the operation of the micro powder loading and dust removal system, including wind speed, air pressure, suspended particulate matter concentration and wind energy utilization status, and constructs a wind disturbance correction map based on the spatial disturbance perception model to reflect the degree of influence of wind flow disturbance on environmental data. The dynamic disturbance perception and environmental data correction module receives the wind disturbance correction map and real-time collected environmental data. By constructing a wind flow disturbance mapping function, it identifies local monitoring anomaly areas caused by the operation of the wind energy recovery device, and uses a disturbance compensation correction algorithm to regionally correct the target monitoring parameters, outputting a dynamic environmental state set including trust weights and risk factors. The energy sensing fusion scheduling decision module receives the dynamic environmental state set and the current wind power system energy storage state data. Based on the wind power load adaptation rules and the environmental priority level system, it constructs a dynamic control strategy set for scheduling sampling frequency, power supply priority and communication window, so as to realize the coupled scheduling between wind power status and environmental sensing quality.

2. The port environmental monitoring system based on Internet of Things technology according to claim 1, characterized in that, The wind-driven environmental perception modeling module includes: Multi-source environmental data synchronous acquisition: During the operation of the micro-powder loading and dust removal system, distributed multi-type sensor nodes are deployed downstream of the air outlet to collect environmental data, including local wind speed. Local air pressure Suspended particulate matter concentration Wind energy utilization status The position coordinates of each sensor node are set as follows: And record the corresponding time segment. Constructing a multi-temporal and spatiotemporal environment dataset ; Spatial perturbation factor field construction: The collected environmental data is mapped to a local spatial region, and the perturbation factor at each sampling point is calculated. It outputs the perturbation factor field for each node. ; Construction of wind disturbance correction map: based on disturbance factor field The spatial disturbance level is interpolated and extended to form a wind disturbance correction map in continuous space. .

3. The port environmental monitoring system based on Internet of Things technology according to claim 2, characterized in that, The dynamic disturbance sensing and environmental data correction module includes: Disturbance response mapping construction: Based on the wind disturbance correction map and the real-time environmental data at the current moment, a disturbance response mapping function between the disturbance intensity of the wind disturbance correction map and the environmental data is constructed by establishing a spatial-temporal joint distribution relationship; Local anomaly perception area identification: Receive the results of the disturbance response mapping function, analyze the degree of disturbance anomaly of the monitoring points in the port operation space, identify the areas affected by the operation of wind power devices, calculate the dynamic residual field of the monitoring value deviating from the expected disturbance value, combine spatial continuity constraints, extract the wind-dominated anomaly response area, and output the local anomaly area label, residual intensity and temporal change trend. Disturbance compensation and state reconstruction: Regional correction and state reconstruction are performed on environmental data within the identified local anomaly areas. Using the wind disturbance correction map as the weight basis and combined with the dynamic residual fitting results, a disturbance compensation and repair algorithm is executed to perform noise reduction, interpolation, or confidence scoring on pollution values. The final output dynamic environmental dataset includes the corrected environmental data for each monitoring point, the corresponding trust weight factor, and the disturbance risk factor.

4. The port environmental monitoring system based on Internet of Things technology according to claim 3, characterized in that, The perturbation response mapping construction includes: Spatiotemporally aligned feature sample construction: Extract the current time from all sensor nodes in the same port operation area. The spatial position below And obtain nodes from the wind disturbance correction map. Location disturbance intensity Obtain nodes from real-time environmental data monitoring values and construct a sample set ; Construction of the disturbance response mapping function: for each monitoring dimension in the sample set Kernel-weighted regression was used to construct a perturbation strength For input, environmental data The output is a perturbation response mapping function, and finally, a set of multidimensional perturbation response mapping functions. .

5. The port environmental monitoring system based on Internet of Things technology according to claim 4, characterized in that, The identification of the local anomaly sensing region includes: Perturbation prediction generation and bias estimation: utilizing a set of multidimensional perturbation response mapping functions from the output. The intensity of the disturbance to each current sensing node Input the data to obtain the predicted value for each monitoring dimension. And combined with the actual monitoring values ​​of the nodes Calculate its disturbance deviation residual ; Constructing a dynamic residual field and spatial anomaly measurement: Calculate the sum of squared disturbance residual intensities of all nodes across all monitoring dimensions, and construct a node-level comprehensive disturbance anomaly value. Based on node location Will Spatial mapping into dynamic residual field ; Anomaly response region extraction and label output: in dynamic residual field Set the disturbance anomaly threshold above Identify all that meet the requirements The spatial location constitutes a set of potential anomalies. Based on spatial continuity constraints Perform connected component clustering to extract multiple local anomaly response regions. and for each local anomaly response patch Output the corresponding tag ID Average residual strength The residual trend curve over continuous time intervals Ultimately, this forms a regional set that includes anomaly response information from multiple areas. .

6. The port environmental monitoring system based on Internet of Things technology according to claim 5, characterized in that, The disturbance compensation and state reconstruction include: Weighted modeling of disturbance effects at anomaly monitoring points: For each identified local anomaly response region Extract each monitoring point Raw environmental observations Furthermore, a repair reference function based on disturbance intensity weighting is constructed to calculate the disturbance compensation reconstruction value; Dynamic residual scoring and trust weight generation: Calculate the residual deviation for each monitoring point. Assign a confidence value to the monitoring point, and combine the confidence levels of each monitoring point across all dimensions into a disturbance risk factor. ; Output dynamic environment state set: for each monitoring point Output the corrected dynamic environment state vector Finally, the reconstructed dynamic environment dataset of the entire anomaly response area is output. The reconstruction results of all abnormal response areas are then merged into a corrected environmental state set. .

7. The port environmental monitoring system based on Internet of Things technology according to claim 6, characterized in that, The energy fusion scheduling decision module includes: Wind power-environmental state fusion modeling: Receive dynamic environmental state set and current energy storage state data of wind power system, extract trust weight factor and disturbance risk factor of each monitoring point, and combine the state of charge, voltage level and remaining available power of wind power battery to construct energy-sensing coupled state matrix. According to the environmental state level and wind power output capacity, calculate the current allocable energy resources and sensing task priority relationship to form the basis of dynamic resource-task adaptation under wind power drive. Multi-objective dynamic control strategy generation: Based on the energy-sensing coupled state matrix, a multi-objective scheduling function is constructed that simultaneously considers the data perception value and energy consumption constraints, generating adaptive control strategies, including sampling frequency scheduling, power supply priority ranking, and communication window allocation. The adaptive control strategies are dynamically adjusted according to the current operating strategy level and sudden environmental disturbances to achieve optimized allocation of different functional resources. Strategy execution and command output: The generated adaptive control strategy is transformed into issued device control commands, which include sensor sampling control, lighting equipment power supply switching, and communication unit activation plans. These commands are then synchronously distributed to edge nodes via the Internet of Things communication protocol.

8. The port environmental monitoring system based on Internet of Things technology according to claim 7, characterized in that, The wind power-environmental state fusion modeling includes: State data synchronization and feature extraction: from corrected environment state sets Extract each monitoring point Corrected environmental data vector Trust weight factor vector Disturbance risk factors Simultaneously, the battery state of charge is obtained from the current energy storage state data of the wind power system. Battery terminal voltage Remaining available power Formation time Multi-source state sample set ; Energy-sensory coupled state matrix construction: Based on the confidence level, risk factor, and wind power availability of each monitoring point, calculate the state matrix for each node. Perception priority indicators Define the share of allocable power in the current wind power system. And construct the sensing coupling state matrix ; Dynamic resource-task adaptation calculation: based on the energy-sensory coupling state matrix Calculate the adaptation coefficient between the node's sensing task and the available energy resources at the current moment. And according to the environmental status level This forms the basis for dynamic resource-task adaptation. .

9. The port environmental monitoring system based on Internet of Things technology according to claim 8, characterized in that, The generation of the multi-objective dynamic control strategy includes: Construct a set of node perception performance indicators: for each monitoring point Based on its environmental status level Trust factor Disturbance risk factors and current battery status Construct its perceived effectiveness indicators With energy consumption tolerance factor ; Constructing a multi-objective scheduling function: Calculating monitoring points using a multi-objective scheduling function. The scheduling priority score is used to achieve the dual objectives of maximizing perceived value and minimizing energy consumption costs. Control strategy generation: Based on the scheduling priority score of each monitoring point, an adaptive control strategy is generated, including sampling frequency scheduling, power supply priority ranking, and communication window allocation; Policy level and disturbance adaptive update: based on the current running policy level and sudden disturbance factors , dynamically adjust adaptive regulation strategies.

10. The port environment monitoring system based on Internet of Things technology according to claim 9, characterized in that, The strategy execution and instruction output include: Strategy instruction structured encoding: Map each control parameter in the adaptive control strategy to standardized equipment instruction fields to construct the instruction data packet structure; Control command allocation and packaging: based on the perception node ID and the current operating policy level. Construct a scheduling instruction mapping table and complete the packaging of device instructions; Protocol selection and command issuance: Select the IoT communication protocol based on the node's communication capabilities and the environmental topology to complete the broadcasting or targeted issuance of control commands.

Images