Iot-based large model of smart city pipe network odor emergency monitoring system, method, medium and device

Abstract

The application provides an Internet of Things large model-based smart city pipe network odor emergency supervision system, method, medium and device. The system comprises an emergency supervision management platform, which is configured to: acquire multi-source monitoring data of a target area; determine an abnormal monitoring threshold based on historical odor fingerprint data; determine an odor area based on the odor fingerprint data and the abnormal monitoring threshold; determine an odor category based on the multi-source monitoring data of the odor area; determine a recommended cleaning solution through a cleaning solution vector library according to the odor category and the multi-source monitoring data of the odor area; and generate a cleaning work order based on the odor area and the recommended cleaning solution. The system controls traffic signals through cross-system linkage to quickly isolate the dangerous area, thereby winning valuable time for professional disposal and maximizing the safety of the lives and property of citizens.

Description

Technical Field

[0001] This specification relates to the field of smart city environmental monitoring and emergency management technology, and in particular to a smart city pipeline odor emergency monitoring system, method, medium and device based on the Internet of Things big data model. Background Technology

[0002] As urban underground pipe networks continue to expand and become more complex, odor problems from these networks are gradually becoming a significant factor affecting urban environmental quality and residents' comfort. Failure to detect and address these issues promptly can lead to air pollution in the surrounding area, resident complaints, and even safety hazards.

[0003] Currently, odor monitoring in pipe networks mainly relies on fixed-point sensors and manual inspections. While these methods possess basic data collection capabilities, there is still room for improvement in real-time performance, coverage, and intelligent analysis. Faced with complex urban pipe network systems, quickly locating odor sources, accurately identifying their types, and optimizing cleaning strategies have become key industry concerns. In recent years, the development of IoT and big data technologies has provided a foundation for building intelligent monitoring systems. However, existing solutions are mostly limited to a single data dimension, lacking the ability to integrate, analyze, and predict multi-source information such as meteorological, hydrological, and business data, making it difficult to achieve closed-loop management from perception and decision-making to execution.

[0004] Therefore, there is an urgent need to build a smart city pipeline odor emergency monitoring method and system based on the Internet of Things (IoT) big data model, which can sense, identify, accurately trace the source and automatically execute cleaning tasks in real time, so as to improve the intelligence level and response efficiency of urban environmental management. Summary of the Invention

[0005] This specification provides one or more embodiments of a smart city pipeline odor emergency monitoring system based on an Internet of Things (IoT) big data model. The smart city pipeline odor emergency monitoring system based on an IoT big data model includes an emergency monitoring management platform. The emergency monitoring management platform is configured to execute a smart city pipeline odor emergency monitoring method based on an IoT big data model.

[0006] One embodiment of this specification provides a method for emergency monitoring of odors in smart city pipe networks based on an Internet of Things (IoT) big data model. The method is executed by the emergency monitoring management platform of the smart city pipe network odor emergency monitoring IoT system. The method includes: acquiring multi-source monitoring data of a target area, including odor fingerprint data, meteorological data, and pipe network hydrological data; determining anomaly monitoring thresholds based on historical odor fingerprint data; determining odor areas based on the odor fingerprint data and the anomaly monitoring thresholds; determining odor categories based on the multi-source monitoring data of the odor areas; determining recommended cleaning schemes using a cleaning scheme vector library based on the odor categories and the multi-source monitoring data of the odor areas; generating cleaning work orders based on the odor areas and the recommended cleaning schemes, and sending them to an automatic cleaning vehicle; controlling the automatic cleaning vehicle to travel to the odor area based on the cleaning work order; and adjusting the valve size of the water gun and / or the valve size of the chemical substance delivery pipe of the automatic cleaning vehicle based on the recommended cleaning scheme.

[0007] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes a smart city pipeline odor emergency monitoring method based on an Internet of Things (IoT) big data model.

[0008] This specification provides one or more embodiments of a smart city pipeline odor emergency monitoring device based on an Internet of Things (IoT) big data model. The device includes at least one processor and at least one memory; the at least one memory stores computer instructions; and the at least one processor executes at least a portion of the computer instructions to implement a smart city pipeline odor emergency monitoring method based on an IoT big data model.

[0009] One or more embodiments of this specification identify major risks such as industrial hazardous gas leaks and quickly isolate dangerous areas by controlling traffic signals across systems, thus gaining valuable time for professional response while maximizing the protection of citizens' lives and property. Attached Figure Description

[0010] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0011] Figure 1 This is a schematic diagram of the platform structure of a smart city pipeline odor emergency monitoring system based on an Internet of Things (IoT) big data model, as shown in some embodiments of this specification. Figure 2This is an exemplary flowchart of an emergency monitoring method for odor in smart city pipe networks based on an Internet of Things (IoT) big data model, as shown in some embodiments of this specification. Figure 3 These are exemplary schematic diagrams of a classification model shown according to some embodiments of this specification; Figure 4 This is a schematic diagram illustrating the determination of a recommended cleaning scheme based on some embodiments of this specification. Detailed Implementation

[0012] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0013] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0014] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0015] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0016] Figure 1 This is a schematic diagram of the platform structure of a smart city pipeline odor emergency monitoring system based on an Internet of Things (IoT) big data model, as shown in some embodiments of this specification.

[0017] In some embodiments, such as Figure 1As shown, the smart city pipeline odor emergency monitoring system 100 based on the Internet of Things big data model includes an emergency monitoring user platform 110, an emergency monitoring service platform 120, an emergency monitoring management platform 130, an emergency monitoring sensor network platform 140, and an emergency monitoring perception and control platform 150.

[0018] Emergency monitoring user platform 110 refers to a platform for interaction with users (such as regulatory personnel). In some embodiments, emergency monitoring user platform 110 includes terminal devices. For example, terminal devices may include mobile devices, tablet computers, and consoles.

[0019] Emergency monitoring service platform 120 refers to a platform used for receiving and transmitting data and / or information. In some embodiments, emergency monitoring service platform 120 is configured as a server or processor, etc. Emergency monitoring service platform 120 can interact bidirectionally with the data center 135 of emergency monitoring user platform 110 and emergency monitoring management platform 130.

[0020] The emergency monitoring and management platform 130 refers to a comprehensive management platform that manages and coordinates the connections and collaboration between multiple platforms. In some embodiments, the emergency monitoring and management platform 130 is configured as a server or processor, etc. In some embodiments, the emergency monitoring and management platform 130 includes an emergency prevention sub-platform 131, an emergency monitoring sub-platform 132, a risk prevention sub-platform 133, an emergency response sub-platform 134, and a data center 135.

[0021] Emergency Prevention Sub-Platform 131 is configured to manage emergency prevention for various sudden odor incidents.

[0022] Emergency monitoring sub-platform 132 is configured to conduct real-time emergency monitoring of various sudden odor events.

[0023] The risk prevention sub-platform 133 is configured to conduct risk assessment and prevention management for various sudden odor incidents.

[0024] Emergency Response Sub-Platform 134 is configured to handle and command various sudden odor incidents.

[0025] Data center 135 is configured to store data, run models, and provide computational support. Data center 135 includes database 136, data processing model library 137, and computing unit 138. In some embodiments, data center 135 is communicatively connected to emergency prevention sub-platform 131, emergency monitoring sub-platform 132, risk prevention sub-platform 133, and emergency response sub-platform 134, providing data storage, model support, and computational resources to each sub-platform.

[0026] In some embodiments, the emergency monitoring and management platform 130 can interact bidirectionally with the emergency monitoring sensor network platform 140 via the data center 135. For example, the emergency monitoring and management platform 130 can obtain monitoring data from the emergency monitoring perception and control platform 150 through the emergency monitoring sensor network platform 140; or, for another example, the emergency monitoring and management platform 130 can issue a cleaning work order to the emergency monitoring perception and control platform 150 through the emergency monitoring sensor network platform 140, and after receiving the cleaning work order, the emergency monitoring perception and control platform 150 can control an automatic cleaning vehicle to move to the odor area and perform cleaning based on the cleaning work order.

[0027] For more information on cleaning work orders, automatic cleaning trucks, and odorous areas, please see [link / reference]. Figure 2 And its related descriptions.

[0028] The emergency monitoring and sensing control platform 150 refers to a platform that generates sensor information and executes control commands. In some embodiments, the emergency monitoring and sensing control platform 150 includes devices such as odor sensors, automatic cleaning vehicles, ultrasonic level gauges, and flow meters deployed in the drainage pipe network. The emergency monitoring and sensing control platform 150 can interact bidirectionally with the emergency monitoring sensor network platform 140.

[0029] For more information on odor sensors, ultrasonic level gauges, and flow meters, please refer to [link / reference needed]. Figure 2 And its related descriptions.

[0030] Figure 2 This is an exemplary flowchart illustrating an emergency monitoring method for odors in smart city pipe networks based on an IoT big data model, according to some embodiments of this specification. Figure 2 As shown, process 200 includes the following steps. In some embodiments, process 200 may be executed by an emergency monitoring and management platform.

[0031] Step 210: Obtain multi-source monitoring data for the target area.

[0032] The target area refers to the area that needs to be monitored. In some embodiments, the target area is determined by user input.

[0033] Multi-source monitoring data refers to a collection of data collected from different sources and dimensions, used to comprehensively assess the status of the pipeline network.

[0034] In some embodiments, multi-source monitoring data includes odor fingerprint data, meteorological data, and pipeline hydrological data.

[0035] Odor fingerprint data refers to data used to characterize the gas composition characteristics of certain points in a target area.

[0036] In some embodiments, odor fingerprint data may include gas composition (such as the chemical composition of the gas) and concentration acquired by the sensor, sensor identification information (ID), and the time when the sensor acquired the data. By way of example only, odor fingerprint data may be characterized based on an odor fingerprint vector (timestamp, sensor ID, hydrogen sulfide and its concentration, ammonia and its concentration).

[0037] In some embodiments, odor fingerprint data can be acquired in real time by odor sensors (e.g., electronic noses) deployed in the drainage network. In some embodiments, the deployment location of the odor sensors can be determined by obtaining user input.

[0038] Meteorological data refers to data that describes the atmospheric conditions and phenomena in a target area. In some embodiments, meteorological data may include air temperature, humidity, wind speed, precipitation type and intensity, etc.

[0039] In some embodiments, meteorological data may be obtained through a third-party platform (such as a meteorological bureau).

[0040] Pipeline hydrological data refers to data describing the physical state of liquids within the drainage pipeline network of a target area. In some embodiments, pipeline hydrological data includes liquid height, fluid velocity, etc.

[0041] In some embodiments, pipeline hydrological data can be obtained by installing ultrasonic level gauges and flow meters at critical sections of the pipeline. The critical section of the pipeline refers to the pipeline section location in the drainage pipeline system that is preferentially selected for data monitoring for purposes such as monitoring, diagnosis or control. The critical section of the pipeline can be preset based on human experience.

[0042] Step 220: Determine the anomaly monitoring threshold based on historical odor fingerprint data.

[0043] Historical odor fingerprint data refers to odor fingerprint data collected before the current time point.

[0044] Anomaly monitoring thresholds are criteria used to identify whether gas concentrations deviate from normal baselines.

[0045] In some embodiments, the anomaly monitoring threshold includes the concentration threshold of gas components in the odor fingerprint data.

[0046] In some embodiments, the emergency monitoring and management platform can calculate the average concentration of each gaseous chemical component based on the odor fingerprint data of each odor sensor over a historical preset time period (e.g., the past three days), and set this average value as the abnormal monitoring threshold corresponding to that gaseous chemical component at the current time. The historical preset time period can be pre-set based on human experience. In some embodiments, the historical preset time period can be a period during which no odor has been generated.

[0047] In some embodiments, the emergency monitoring and management platform can adjust the abnormal monitoring thresholds based on meteorological data and pipeline hydrological data.

[0048] In some embodiments, the emergency monitoring and management platform can determine the abnormal monitoring threshold by querying a first preset table based on meteorological data and pipeline hydrological data. The first preset table records the correspondence between meteorological data, pipeline hydrological data, and abnormal monitoring thresholds. The first preset table can be pre-set based on human experience.

[0049] In some embodiments, when the temperature is high in meteorological data and the fluid velocity is low in pipeline hydrological data, the anomaly monitoring threshold can be lowered to improve the sensitivity of odor monitoring. When there is rainfall in the meteorological data and the fluid velocity is high in the pipeline hydrological data, the anomaly monitoring threshold can be raised to prevent false alarms and waste of resources.

[0050] In some embodiments of the specification, the abnormal monitoring threshold is dynamically fine-tuned by introducing meteorological data and pipeline hydrological data, enabling the system to intelligently sense environmental changes and improve monitoring sensitivity under conditions conducive to odor diffusion to achieve early warning. Under conditions such as heavy rain, the standards are appropriately relaxed to prevent false alarms, thereby improving the accuracy and rationality of odor anomaly identification.

[0051] Step 230: Determine the odor area based on odor fingerprint data and anomaly monitoring threshold.

[0052] An odorous area refers to an area where an odor exists. In some embodiments, an odorous area may be a sub-region of a target area.

[0053] In some embodiments, the emergency monitoring and management platform can count the number of gases in the odor fingerprint data whose concentration exceeds the corresponding abnormal monitoring threshold. When the number of gases exceeds a first preset number, the location of the sensor corresponding to the sensor identity information in the odor fingerprint data is identified as an anomaly point. All anomaly points or the area they enclose can be considered as an odor area. The first preset number can be preset based on human experience.

[0054] Step 240: Determine the odor category based on multi-source monitoring data of the odor area.

[0055] Odor categories refer to the classification of odor sources or properties. In some embodiments, odor categories include types such as food grease and grime, domestic sewage, certain industrial solvents, and waste fermentation.

[0056] In some embodiments, the emergency monitoring and management platform can determine the odor category by querying a second preset table based on the odor area and multi-source monitoring data. The second preset table records the correspondence between odor areas, multi-source monitoring data, and odor categories. The second preset table can be pre-set based on human experience. For more information on how to determine the odor category, please refer to [link to relevant documentation]. Figure 3 And its related descriptions.

[0057] Step 250: Based on the multi-source monitoring data of odor categories and odor areas, determine the recommended cleaning plan through the cleaning plan vector library.

[0058] The cleanup scheme vector library is a database storing historical cleanup cases. Each historical cleanup case consists of a feature vector and a label, and each historical cleanup case includes effective solutions. The feature vector is composed of multi-source monitoring data of historical odor categories and odor areas from the historical cleanup case, and its corresponding label is the recommended cleanup scheme corresponding to the effective scheme. The effective scheme refers to the cleanup scheme that successfully reduced the gas concentration to the abnormal monitoring threshold within a preset time period during the historical cleanup process.

[0059] A recommended cleaning plan refers to a cleaning scheme determined for the odor problem in the target area. In some embodiments, the recommended cleaning plan includes cleaning methods and cleaning parameters. Cleaning methods refer to the technical means employed to implement the cleaning, such as high-pressure flushing (e.g., water pressure exceeding a preset water pressure threshold), high-flow-rate rinsing (e.g., water flow exceeding a preset water flow threshold), or the addition of chemical agents. Cleaning parameters refer to parameters that quantify and finely adjust the cleaning method. Cleaning parameters may include the water gun pressure, the duration of the rinsing operation, and the precise dosage of chemical agents.

[0060] In some embodiments, the emergency monitoring and management platform can query a cleanup plan vector library based on multi-source monitoring data of odor categories and odor areas to determine a recommended cleanup plan. For example, the emergency monitoring and management platform can construct a first target feature vector from the real-time acquired odor categories and multi-source monitoring data, retrieve a first reference feature vector with the highest similarity to the first target feature vector from the cleanup plan vector library, and use the first reference label corresponding to the first reference feature vector as the recommended cleanup plan.

[0061] Step 260: Based on the odor area and recommended cleaning plan, generate a cleaning work order and send it to the automatic cleaning truck.

[0062] A cleaning work order is an electronic task order automatically generated and sent to the automatic cleaning vehicle by the emergency monitoring and management platform. In some embodiments, the cleaning work order may include a recommended cleaning plan, the identity information (ID) of the automatic cleaning vehicle performing the task, and a navigation route to guide the automatic cleaning vehicle to the odor area.

[0063] In some embodiments, the emergency monitoring and management platform sends cleaning work orders to the automatic cleaning vehicle. It can control the automatic cleaning vehicle to travel to the odor area based on the work order and adjust the valve size of the water gun and / or the valve size of the chemical dispensing pipe based on the recommended cleaning plan. For example, based on the cleaning method and parameters in the recommended cleaning plan, the platform controls the automatic cleaning vehicle to adjust the valve size of the corresponding water gun and / or the valve size of the chemical dispensing pipe by adjusting the water gun pressure and the precise dosage of the chemical agent.

[0064] In some embodiments, the emergency monitoring and management platform can generate a cleaning work order based on the identity information (ID) of the automatic cleaning vehicle closest to the odor area and the navigation route obtained from a third-party platform.

[0065] In some embodiments of the specification, a closed-loop monitoring system is constructed, encompassing odor perception, intelligent analysis, and automated treatment. By accurately matching cleaning plans and automatically controlling the operation of cleaning vehicles, the traditional model relying on manual inspection and experience-based judgment is changed, enabling rapid response and precise and efficient treatment of odors in the pipeline network, thereby improving the level of intelligent urban environmental management.

[0066] In some embodiments, in response to the odor being classified as an industrial hazardous gas and the duration of the odor meeting a third preset condition, the emergency monitoring and management platform generates a traffic instruction and controls the smart traffic lights on the road where the odor area is located to switch to a no-passing mode according to the traffic instruction.

[0067] Industrial hazardous gases refer to gaseous substances produced during industrial production processes that pose a direct threat to human health or the environment.

[0068] In some embodiments, industrial hazardous gases may include strongly irritating gases, flammable and explosive gases, and toxic gases. For example, industrial hazardous gases may include gases such as carbon monoxide, chlorine, and benzene compounds.

[0069] The third preset condition is a criterion used to determine whether the duration of the odor is abnormal. In some embodiments, the third preset condition may be that the duration of the odor exceeds a preset duration threshold. The preset duration threshold refers to the threshold condition for determining the maximum tolerable time for the odor to dissipate naturally, and the preset duration threshold can be preset based on human experience.

[0070] Traffic instructions refer to electronic commands generated by the emergency monitoring and management platform and sent to the traffic signal control system (such as intelligent traffic lights).

[0071] In some embodiments, traffic instructions can be pre-set based on human experience.

[0072] In some embodiments, in response to the odor being classified as an industrial hazardous gas and the duration of the odor meeting a third preset condition, the emergency monitoring and management platform may use traffic instructions pre-set based on human experience as traffic instructions.

[0073] The "No Entry" mode refers to the operating state of a smart traffic light that prohibits all or specific types of vehicles and pedestrians from entering a designated danger zone. For example, the traffic light may continuously display a red light or a specific warning signal (such as a flashing red light).

[0074] In some embodiments of the specification, by identifying major risks such as industrial hazardous gas leaks, traffic signals are controlled across systems to quickly isolate dangerous areas, thus buying valuable time for professional handling while maximizing the protection of citizens' lives and property.

[0075] Figure 3 This is an exemplary schematic diagram of a classification model shown according to some embodiments of this specification.

[0076] In some embodiments, such as Figure 3 As shown, the emergency monitoring and management platform is further configured to: determine the odor category 340 based on multi-source monitoring data 310 from the odor area using a classification model 330. The classification model is a machine learning model.

[0077] For more information on odor areas, multi-source monitoring data, and odor categories, please refer to [link / reference]. Figure 2 And its related descriptions.

[0078] In some embodiments, the classification model is a machine learning model, such as any one or a combination of a neural network (NN) model, a convolutional neural network (CNN) model, or other custom model structures.

[0079] In some embodiments, the input to the classification model may include multi-source monitoring data such as odor fingerprint data, meteorological data, and pipeline hydrological data, and the output may be odor categories.

[0080] In some embodiments, the emergency monitoring and management platform can train a classification model using supervised learning algorithms on multiple training samples with training labels. The training samples include multi-source monitoring data of the odor area over a historical period, and the training labels include the actual odor category corresponding to the odor area during that historical period. For example, the emergency monitoring and management platform can employ optimization methods such as gradient descent and backpropagation algorithms. Based on the labeled multi-source monitoring data and their corresponding actual odor category labels, the platform can iteratively adjust the classification model parameters through a training process to ultimately obtain a classification model capable of accurately identifying odor categories. The historical period can be the event cycle in which the odor occurred and the odor type was confirmed.

[0081] In some embodiments, training samples and labels can be obtained from historical data.

[0082] In some embodiments, the training process of the classification model may include: the emergency monitoring and management platform inputting multiple training samples into an initial classification model, constructing a loss function based on the training labels and the output of the initial classification model, and then iteratively updating the parameters of the initial classification model based on the loss function. When the training conditions are met, the model training is complete, and a trained classification model is obtained. The training conditions may include the convergence of the loss function or the number of iterations reaching a threshold, etc.

[0083] In some embodiments, the emergency monitoring and management platform is further configured to: determine the upstream end of the odor area based on the odor area and map data; and determine the odor category 340 based on the multi-source monitoring data 310 of the odor area and the business type data 320 of the target route using the classification model 330. The target route includes a drainage pipe route from the upstream end of the odor area to the odor area.

[0084] Map data refers to geographic data describing geospatial information and pipeline network topology. Map data may include the route of drainage pipelines, node distribution, upstream and downstream relationships, and the spatial location of surface facilities. In some embodiments, map data may be GIS maps, satellite maps, etc. Map data can be obtained from third-party organizations such as city government management departments and big data management bureaus.

[0085] The upstream end refers to the point where the abnormal odor originates in the current odor area.

[0086] In some embodiments, the emergency monitoring and management platform can trace back from the odor area along the opposite direction of water flow based on odor area and map data to determine the upstream end of the abnormal odor in the odor area.

[0087] The specific process includes: the emergency monitoring and management platform can obtain the time t when the odor area is determined to have an odor; starting from the odor area sensor S0 (the odor area sensor S0 can be any odor sensor in the odor area, or it can be the upstream odor sensor in the area), using map data, it traces along all possible upstream paths to find all directly upstream odor sensors S0 connected to different branches of the odor area sensor S0. i (i=1,2,...).

[0088] For each upstream odor sensor S i (i=1,2,...), combining pipe distance and average flow velocity, calculate the odor substances from each upstream odor sensor S. i (i=1,2,...) The time Δt required for the odor to propagate to the odor sensor S0 in the odor region. Extract the upstream odor sensor S. i Odor fingerprint data is collected at time (t-Δt), and the similarity between this odor fingerprint data and the odor fingerprint data collected by the odor region sensor S0 at time t is calculated. If the similarity is greater than or equal to the similarity threshold (e.g., 0.6), the upstream tracing continues; when a certain upstream odor sensor S0... i If the similarity between the odor fingerprint data collected at time (t-Δt) and the odor fingerprint data collected by the odor region sensor S0 at time t is less than the similarity threshold, an abnormal signal is determined to be present at the upstream odor sensor S0. i If the corresponding location is not clearly visible or has decayed to the point of being unrecognizable, then the tracing of this branch is terminated, and the currently traced upstream odor sensor S is moved to the next location. i The location of the odor sensor directly downstream of the branch is considered the "upstream end" of the odor area.

[0089] In some embodiments, the average flow velocity can be obtained from pipeline hydrological data. The similarity between odor fingerprint data can be measured using methods such as cosine similarity. The similarity threshold can be preset manually based on historical experience.

[0090] The target route includes the drainage pipe route from the upstream end of the odor area to the odor area. In some embodiments, the target route can be determined based on the pipe network topology in map data. In some embodiments, since the pipe network may have multiple path branches, an abnormal odor event may correspond to multiple upstream ends, and correspondingly, there may be multiple target routes.

[0091] Business type data refers to structured data on the types and distribution of socio-economic activities within a geographical area. For example, the composition of businesses (shopping malls, restaurants, factories, etc.) along the road surface corresponding to the target route, as well as the number of different types of businesses.

[0092] In some embodiments, the emergency monitoring and management platform can input the multi-source monitoring data of the odor area and the business data of the target route into a trained classification model based on the multi-source monitoring data of the odor area and the business data of the target route. The classification model processes the input features and outputs the corresponding odor category, thereby determining the odor category corresponding to the current abnormal odor.

[0093] In some embodiments, when the input of the classification model includes business data of the target route, the training samples are updated accordingly. Specifically, this includes: acquiring business data of the drainage pipe line from the upstream end of the odor area to the odor area itself in historical odor events; associating the business data with multi-source monitoring data of the corresponding time period and the actual odor category label to construct an expanded training sample containing multi-source monitoring data and business data of the target route; and using the expanded training sample to update and train the classification model so that the classification model learns the odor category association features under the combined effect of multi-source monitoring data and business data of the target route.

[0094] In some embodiments of this specification, by determining the upstream endpoint based on odor area and map data, and combining this with business type data of the target route to determine the odor category, the pollution source tracing results can be integrated with real environmental information for classification decisions. Using the pipeline network topology and odor sensor fingerprint similarity to trace the upstream endpoint helps to locate the potential source path of abnormal odors; using the business type data of the businesses along the drainage pipeline from the upstream endpoint to the odor area as input features of the classification model provides the model with information related to the actual polluting entity. This makes the classification process not only rely on sensor data but also combine geographic space and socio-economic attributes, enhancing the ability to distinguish different pollution types, thereby reducing classification ambiguity and improving the accuracy of odor type determination.

[0095] In some embodiments of this specification, a classification model is used to analyze multi-source monitoring data. This model can automatically learn the explicit and implicit complex correlation features between odor fingerprint data, meteorological data, pipeline hydrological data, and different odor types, effectively improving the accuracy and processing efficiency of odor category identification and achieving intelligent identification of different pollution source characteristics. The accurate odor category determination output by the classification model provides a reliable data foundation for subsequent matching of targeted recommended cleaning solutions.

[0096] Figure 4 This is a schematic diagram illustrating the determination of a recommended cleaning scheme based on some embodiments of this specification.

[0097] In some embodiments, the emergency monitoring and management platform determines the duration of an odor (420) based on the growth rate (411), odor category (412), and multi-source monitoring data (310) of the odor area using a prediction model, where the prediction model is a machine learning model. In response to the odor duration meeting a first preset condition (430), a recommended cleanup plan (440) is determined based on the odor category (412), the multi-source monitoring data (310) of the odor area, and the odor duration (420). For more information on odor categories, odor areas, multi-source monitoring data, and recommended cleanup plans, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0098] The growth rate refers to the rate of change in the concentration of various gases in odor fingerprint data per unit time. For example, the growth rate can include the rate of change in ammonia concentration, the rate of change in hydrogen sulfide concentration, etc.

[0099] Odor duration refers to the length of time required for an odor event in a target odor area to completely dissipate from the current time under natural conditions without human intervention.

[0100] In some embodiments, the emergency monitoring and management platform determines the duration of an odor based on multi-source monitoring data of growth rate, odor type, and odor area using a predictive model.

[0101] A predictive model is a model used to determine the duration of an odor. In some embodiments, the predictive model is a machine learning model, such as a neural network (NN) model.

[0102] In some embodiments, the inputs to the prediction model include multi-source monitoring data on growth rate, odor type, and odor region, and the output is the duration of odor.

[0103] In some embodiments, the prediction model is trained using a large number of second training samples with a second label. The second training samples include historical odor categories, historical growth rates, and historical multi-source monitoring data of the odor areas in the samples; the second label is the duration of the odor.

[0104] The second training sample and its labels are constructed as follows: Historical odor categories, historical growth rates, and historical multi-source monitoring data of the odor area at the time of the event are extracted from historical odor event data and used as training samples. The time span from when the odor is identified as abnormal until the concentration of all related gases remains below its abnormal monitoring threshold without human intervention is recorded as the label. The model training for the prediction model is similar to that for the classification model; please refer to the section on classification model training and related content, which will not be elaborated upon here.

[0105] The first preset condition refers to the condition used to determine whether artificial intervention is needed to account for the natural dissipation of odors. In some embodiments, the first preset condition may be that the duration of the odor exceeds a preset duration threshold. For more information on preset duration thresholds, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0106] In some embodiments, the feature vectors of historical cleaning cases in the cleaning scheme vector library also include the duration of odor.

[0107] In some embodiments, the emergency monitoring and management platform can query a cleanup solution vector library based on multi-source monitoring data of the odor type, odor area, and odor duration to determine a recommended cleanup solution. For information on how to query the cleanup solution vector library, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0108] In some embodiments of the specification, by introducing the duration of odor, the system is equipped with a forward-looking decision-making capability, which can intelligently distinguish between transient odors that can be naturally resolved and persistent problems that require human intervention. This avoids over-responding to unnecessary events, enables precise allocation of emergency resources and maximizes cost-effectiveness, and improves the level of intelligence in odor control.

[0109] In some embodiments, in response to the difference between the duration of the odor and the actual duration satisfying a second preset condition, the emergency monitoring and management platform updates the abnormal monitoring threshold or the prediction model based on the odor fingerprint data and the duration of the odor.

[0110] The second preset condition is used to determine whether it is necessary to update the anomaly monitoring threshold or the prediction model.

[0111] In some embodiments, the second preset condition may be that the difference between the duration of the odor and the actual duration exceeds a preset difference threshold. The preset difference threshold can be preset based on human experience.

[0112] In some embodiments, there are two known cases: the difference between the duration of the odor and the actual duration satisfies the second preset condition, and the difference between the duration of the odor and the actual duration does not satisfy the second preset condition. In response to the difference between the duration of the odor and the actual duration satisfying the second preset condition, and when one or more gases are present in the odor fingerprint data and their concentration exceeds a preset multiple of the corresponding preset concentration threshold, such as exceeding 2 times or 1.5 times the corresponding preset concentration threshold, the emergency monitoring and management platform will include the growth rate, odor type, and odor area multi-source monitoring data and actual duration obtained from this monitoring into the database to form a new second training sample and a second training label, and update the prediction model through the new second training sample and the second training label.

[0113] In response to the difference between the duration of the odor and the actual duration satisfying the second preset condition, and when there are one or more gases in the odor fingerprint data whose concentration exceeds the corresponding preset concentration threshold but does not reach a preset multiple of the preset concentration threshold, such as exceeding the corresponding preset concentration threshold but not reaching 2 or 1.5 times the preset concentration threshold, the abnormal monitoring threshold is reduced according to the preset reduction range. The preset reduction range and preset multiple can be preset based on human experience.

[0114] In some embodiments of the specification, an adaptive optimization closed loop of "prediction-feedback-learning" is established, enabling the system to continuously evolve. When there is a significant deviation between the duration of odor and the actual duration, the system can automatically transform abnormal cases into high-quality training samples. By iteratively updating the prediction model or dynamically correcting the anomaly monitoring threshold, the system can autonomously optimize the algorithm parameters. By dynamically adjusting the monitoring threshold, the system can maintain monitoring sensitivity when gas concentration changes, avoiding false alarms or missed alarms, thereby continuously improving the accuracy of odor identification and the timeliness of early warning during long-term operation.

[0115] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0116] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0117] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0118] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0119] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0120] For each patent, patent application, patent application publication, and other material such as articles, books, specifications, publications, and documents referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0121] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. An Internet of Things (IoT) large model-based smart city pipe network odor emergency monitoring system, characterized in that, The system includes an emergency monitoring and management platform, which is configured as follows: Acquire multi-source monitoring data of the target area, including odor fingerprint data, meteorological data and pipeline hydrological data. The odor fingerprint data includes the gas composition and concentration acquired by the sensor, the sensor's identification information and the time when the sensor acquired the data. Based on historical odor fingerprint data, anomaly monitoring thresholds are determined; Based on the odor fingerprint data and the anomaly monitoring threshold, the odor area is determined; based on the multi-source monitoring data of the odor area, the odor category is determined through a classification model, wherein the classification model is a machine learning model. Based on the multi-source monitoring data of the odor category and the odor area, a recommended cleaning plan is determined through a cleaning plan vector library; The emergency monitoring and management platform is further configured as follows: Based on the growth rate, the odor category, and the multi-source monitoring data of the odor area, the duration of the odor is determined by a prediction model, wherein the prediction model is a machine learning model, and the growth rate refers to the rate of change of the concentration of various gases in the odor fingerprint data per unit time. In response to the odor duration meeting a first preset condition, the recommended cleaning solution is determined based on the odor category, the multi-source monitoring data of the odor area, and the odor duration. The first preset condition includes the odor duration exceeding a preset duration threshold. Based on the odor area and the recommended cleaning plan, a cleaning work order is generated and sent to the automatic cleaning truck; The automatic cleaning vehicle is controlled to travel to the odor area based on the cleaning work order, and the valve size of the water gun and / or the valve size of the chemical substance delivery pipe of the automatic cleaning vehicle is adjusted based on the recommended cleaning plan.

2. The system of claim 1, wherein, The emergency monitoring and management platform is further configured as follows: In response to the odor being classified as an industrial hazardous gas and the odor duration meeting a third preset condition, a traffic order is generated, wherein the third preset condition includes the odor duration exceeding a preset duration threshold. According to traffic instructions, the intelligent traffic lights on the road where the odor area is located are switched to a no-passage mode.

3. A smart city pipe network odor emergency supervision method based on an Internet of Things large model, characterized in that, The method is executed by the emergency monitoring and management platform of the smart city pipeline odor emergency monitoring IoT system, and the method includes: Acquire multi-source monitoring data of the target area, including odor fingerprint data, meteorological data and pipeline hydrological data. The odor fingerprint data includes the gas composition and concentration acquired by the sensor, the sensor's identification information and the time when the sensor acquired the data. Based on historical odor fingerprint data, anomaly monitoring thresholds are determined; Based on the odor fingerprint data and the anomaly detection threshold, the odor area is determined; Based on the multi-source monitoring data of the odor area, the odor category is determined by a classification model, wherein the classification model is a machine learning model; Based on the multi-source monitoring data of the odor category and the odor area, a recommended cleaning plan is determined through a cleaning plan vector library; The step of determining a recommended cleaning plan based on the multi-source monitoring data of the odor category and the odor region, using a cleaning plan vector library, includes: Based on the growth rate, the odor category, and the multi-source monitoring data of the odor area, the duration of the odor is determined by a prediction model, wherein the prediction model is a machine learning model, and the growth rate refers to the rate of change of the concentration of various gases in the odor fingerprint data per unit time. In response to the odor duration meeting a first preset condition, the recommended cleaning solution is determined based on the odor category, the multi-source monitoring data of the odor area, and the odor duration. The first preset condition includes the odor duration exceeding a preset duration threshold. Based on the odor area and the recommended cleaning plan, a cleaning work order is generated and sent to the automatic cleaning truck; The automatic cleaning vehicle is controlled to travel to the odor area based on the cleaning work order, and the valve size of the water gun and / or the valve size of the chemical substance delivery pipe of the automatic cleaning vehicle is adjusted based on the recommended cleaning plan.

4. The method of claim 3, wherein, The method further includes: In response to the odor being classified as an industrial hazardous gas and the odor duration meeting a third preset condition, a traffic order is generated, wherein the third preset condition includes the odor duration exceeding a preset duration threshold. According to traffic instructions, the intelligent traffic lights on the road where the odor area is located are switched to a no-passage mode.

5. A computer readable storage medium, characterized in that, The storage medium stores computer instructions. When the computer reads the computer instructions in the storage medium, the computer executes the smart city pipeline odor emergency monitoring method based on the Internet of Things big model as described in claim 3.

6. An Internet of Things-based large model-based smart city pipe network odor emergency monitoring device, characterized in that, The device includes at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is used to execute at least some of the computer instructions to implement the smart city pipe network odor emergency monitoring method based on the Internet of Things big model as described in claim 3.

Images