Automatic alarm system based on Internet of Things environment monitoring
By introducing a benchmark management module, verification module and cloud server model training module in the Internet of Things environment monitoring system, the shortcomings of the existing system in data management and alarm mechanism are solved, efficient and accurate environmental abnormality detection and rapid emergency response are achieved, and the security and work efficiency of the monitoring area are improved.
Patent Information
- Application Number
- CN202510599324.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-10
- Publication Date
- 2025-08-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing IoT environment monitoring system has shortcomings in data management and abnormal alarm mechanisms, and cannot detect environmental abnormalities in a timely and precise manner. It lacks intelligent design, resulting in insufficient prediction of monitoring blind spots and abnormal diffusion paths, which reduces environmental security guarantee capabilities.
The benchmark management module is used to store the environment benchmark data and obtain real-time monitoring data. It combines the verification module and the abnormal analysis module for consistency verification, designs first- and second-level alarm signals, and uses cloud servers to generate abnormal diffusion paths through the model training module to achieve efficient data interaction and collaborative work.
It improves the accuracy and comprehensiveness of environmental abnormality detection, improves emergency response capabilities, ensures the safety and work efficiency of the monitoring area, and realizes efficient environmental monitoring and accurate alarms.
Smart Images
Figure CN120475046A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of Internet of Things monitoring, and in particular relates to an automatic alarm system based on Internet of Things monitoring. Background Art
[0002] In the field of IoT environmental monitoring, with the continuous development of technology, the demand for real-time monitoring of environmental parameters and abnormal warning is becoming increasingly urgent. Traditional IoT environmental monitoring systems have exposed many problems in practical applications.
[0003] On the one hand, existing systems have a relatively simple approach to managing and comparing environmental baseline data and real-time monitoring data. Most systems only provide basic data collection and storage, making it difficult to efficiently verify the consistency of environmental verification data with baseline and real-time monitoring data. This results in an inability to detect environmental anomalies in a timely and accurate manner. For example, some systems cannot accurately distinguish between normal monitoring targets and abnormal data, and their ability to detect blind spots is extremely limited, often missing important anomaly information and failing to provide users with comprehensive and reliable environmental security.
[0004] On the other hand, the existing system lacks intelligent and sophisticated design in its anomaly alarm mechanism. When an anomaly occurs, the generation and transmission of alarm signals lacks hierarchy and specificity, and cannot effectively differentiate based on the severity and location of the anomaly. This makes it difficult for users to quickly understand the specific circumstances of the anomaly and hinders the timely implementation of appropriate countermeasures. Furthermore, the system's inability to predict and track the spread of anomalies makes it impossible to quickly locate the target of an anomaly after it occurs, reducing the effectiveness of risk prevention and control in the monitored area. Summary of the Invention
[0005] To this end, the present invention provides an automatic alarm system based on Internet of Things environmental monitoring.
[0006] An automatic alarm system for Internet of Things environmental monitoring, comprising:
[0007] The first terminal device is provided with:
[0008] A benchmark management module is configured to store environmental benchmark data and also configured to obtain real-time monitoring data;
[0009] a verification module configured to set a plurality of monitoring nodes and set at least one of the monitoring nodes as a reference monitoring node, each of the monitoring nodes providing data to the reference management module, and generating environmental verification data after verification by the reference monitoring node;
[0010] an abnormality analysis module configured to obtain the environmental verification data, verify the consistency of the environmental verification data with the environmental benchmark data and the real-time monitoring data, and send an alarm signal to the first terminal device or the second terminal device based on a consistency judgment result;
[0011] and the second terminal device, which is provided with:
[0012] a data management module configured to register the environmental baseline data and the monitoring parameters to the first terminal device and receive an alarm signal from the first terminal device;
[0013] And the cloud server, which is set up with:
[0014] A model training module, which obtains the environmental baseline data and the monitoring parameters and generates a training set, and obtains a preset position in the determined abnormal diffusion path after training;
[0015] The dynamic correction module is configured to match and correct the training degree of the model training module according to the alarm signal determined by the first terminal device until the response accuracy reaches a preset ratio.
[0016] As a preferred embodiment, when the abnormality analysis module performs the consistency judgment, the following steps are included:
[0017] Determining the consistency between the environmental baseline data and the monitoring parameters, and determining normal monitoring targets;
[0018] Determining whether the judgment data collected by the reference monitoring node includes abnormal data other than the normal monitoring target;
[0019] If so, an alarm signal is returned to the benchmark management module;
[0020] If not, the sampling data of the monitoring blind area that has not been sampled by the reference monitoring node is determined, and a first-level alarm signal is generated and returned to the first terminal device and the second terminal device.
[0021] As a preferred embodiment, when the abnormality analysis module performs the consistency judgment, after generating a first-level alarm signal and returning it to the first terminal device and the second terminal device, the following steps are further included:
[0022] After any monitoring node other than the reference monitoring node samples abnormal data, a secondary alarm signal is generated and returned to the first terminal device and the second terminal device.
[0023] As a preferred embodiment, when the model training module performs training, the time series data and spatial distribution data are used as the input training set, and the training is performed in the following steps:
[0024] Obtaining a first feature of the spatial distribution data of the time series;
[0025] Obtain a first sample, where the first sample is an exhaustive set of samples obtained in a time series when the first feature of the second terminal device is replaced with any other second terminal device;
[0026] A first feature within a preset threshold corresponding to multiple times under the first sample is determined, and a first response sequence is generated according to the first feature. The first response sequence includes a data set of positions of multiple adjacent monitoring nodes corresponding to any time point.
[0027] As a preferred embodiment, when the dynamic correction module performs correction, the following steps are performed:
[0028] Acquire sampling data of abnormal targets from multiple monitoring nodes, and obtain the diffusion characteristics of abnormal targets from one monitoring node to another;
[0029] The diffusion feature is written into the first response sequence to obtain a matching degree. If the matching degree is within a preset threshold, the diffusion feature is written into the first response sequence to correct the monitoring node position.
[0030] As a preferred embodiment, each of the monitoring nodes is provided with a sensor array configured by a multimodal sensing algorithm, which is used to sample corresponding sampled data in the environmental reference data and perform consistency judgment.
[0031] As a preferred embodiment, the environmental reference data includes fuzzy location information, and the fuzzy location information is used to generate all paths traversing the monitoring nodes.
[0032] As a preferred manner, the environmental reference data includes device identification data and timestamp data, and when the environmental reference data is uploaded to the first terminal device via the second terminal device, the timestamp data is uploaded at a preset period.
[0033] The above technical solution of the present invention has the following advantages over the prior art:
[0034] In terms of data management and anomaly detection, by setting up a baseline management module to store environmental baseline data and obtain real-time monitoring data, combining it with a verification module to set up multiple monitoring nodes, and an anomaly analysis module to verify the consistency of environmental verification data with environmental baseline data and real-time monitoring data, it can accurately determine normal monitoring targets and promptly detect abnormal data. This not only allows for rapid determination of anomalies but also accurately identifies monitoring blind spots, greatly improving the accuracy and comprehensiveness of environmental anomaly detection and providing strong support for ensuring environmental safety in the monitored area.
[0035] The alarm mechanism features innovatively designed level 1 and level 2 alarm signals. Level 1 signals address abnormalities in blind spots, while level 2 signals address abnormal data collected by monitoring nodes other than the baseline. This hierarchical alarm mechanism allows users to quickly understand the source and severity of anomalies, enabling timely response measures and significantly improving the system's emergency response capabilities.
[0036] From a system architecture perspective, efficient data exchange and collaboration are achieved between the first and second terminal devices and the cloud server. The second terminal device uploads environmental baseline data to the first terminal device, and parameter threshold data is uploaded at a preset interval, ensuring the timeliness and accuracy of the data. Through the model training module and dynamic correction module, the cloud server generates a training set based on the environmental baseline data and monitoring parameters, determines the preset location in the anomaly diffusion path, and continuously optimizes the training process to improve response accuracy. This enables the system to accurately track and quickly locate abnormal targets, greatly enhancing the security of the monitored area.
[0037] In terms of application scenarios, the operational processes for regional administrators, on-site inspectors, and the monitoring center have been optimized. The monitoring center can easily set up early warning information and transmit it to relevant modules. Inspectors can promptly report abnormal conditions through the on-site verification submodule. By collating abnormal and early warning information and monitoring with multimodal sensors in real time, the system can accurately determine abnormal conditions in the area and send corresponding information. The entire process is clear and efficient, improving work efficiency and ensuring the smooth implementation of environmental monitoring work. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 It is a structural block diagram of the system provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0039] In order to be able to understand the features and technical content of the embodiments of the present disclosure in more detail, the implementation of the embodiments of the present disclosure is described in detail below in conjunction with the accompanying drawings. The accompanying drawings are for reference only and are not used to limit the embodiments of the present disclosure. In the following technical description, for the sake of convenience of explanation, a full understanding of the disclosed embodiments is provided through multiple details. However, one or more embodiments can still be implemented without these details. In other cases, to simplify the drawings, well-known structures and devices can be simplified for display.
[0040] The present disclosure provides an Internet of Things environment monitoring automatic alarm system, such as Figure 1 As shown, it includes a first terminal device, a second terminal device and a cloud server.
[0041] Specifically, in the embodiment of the present disclosure, the first terminal device is configured to have a benchmark management module, a verification module, and an anomaly analysis module. The benchmark management module is configured to store environmental benchmark data and simultaneously to obtain real-time monitoring data.
[0042] In an embodiment of the present disclosure, the second terminal device is configured to have at least a data management module, which is configured to register the environmental baseline data and the monitoring parameters to the first terminal device.
[0043] Afterwards, the specific settings of other modules of the first terminal device in this embodiment are explained.
[0044] The verification module is configured to set up multiple monitoring nodes and set at least one of the monitoring nodes as a benchmark monitoring node. Each of the monitoring nodes provides data to the benchmark management module, and the benchmark monitoring node verifies and generates environmental verification data.
[0045] The abnormality analysis module is configured to obtain the environmental verification data, verify the consistency of the environmental verification data with the environmental benchmark data and the real-time monitoring data, and send an alarm signal to the first terminal device and the second terminal device according to the consistency judgment result. Specifically, when the abnormality analysis module performs the consistency judgment, it includes the following steps:
[0046] Determine the consistency between the environmental baseline data and the real-time monitoring data, and determine the normal monitoring target;
[0047] Determining whether the judgment data collected by the reference monitoring node includes abnormal data other than the normal monitoring target;
[0048] If so, an alarm signal is returned to the benchmark management module;
[0049] If not, the sampling data of the monitoring blind area that has not been sampled by the reference monitoring node is determined, and a first-level alarm signal is generated and returned to the first terminal device and the second terminal device.
[0050] As a further preferred embodiment of the present disclosure, after the abnormality analysis module performs the consistency judgment and generates the first-level alarm signal and returns it to the first terminal device and the second terminal device, it also includes the following steps: after abnormal data is sampled at any monitoring node other than the benchmark monitoring node, a second-level alarm signal is generated and returned to the first terminal device and the second terminal device.
[0051] The consistency judgment performed by the abnormality analysis module is further explained in the embodiment of the present disclosure.
[0052] In the specific implementation of generating a first-level alarm signal, the monitoring parameters and environmental benchmark data uploaded by the second terminal device can determine the corresponding normal monitoring target at the current time point through the correlation verification between the monitoring parameters and the environmental benchmark data. Therefore, a benchmark monitoring node is set, and the benchmark monitoring node is used to record whether the abnormal target is sampled at the monitoring node. If the abnormal target has been sampled at the benchmark monitoring node, the target is marked as "confirmed abnormality". It should be noted that the target marked as "confirmed abnormality" can only confirm the existence of the abnormal target. As a conventional setting method, in this application, when marking "confirmed abnormality", the environmental parameter value at the time of marking is sampled to determine whether the abnormal target exists and whether the degree of abnormality exceeds the standard. If the parameter value sampled of the target marked "confirmed abnormality" exceeds the preset threshold, it is marked as "serious abnormality". At this time, the abnormality analysis module sends the data including the data marked as "confirmed abnormality" and the data marked as "serious abnormality" to the first terminal device and the second terminal device.
[0053] In another supplementary embodiment, a plurality of verification monitoring nodes are further provided among the plurality of monitoring nodes, and the verification monitoring nodes are used to sample data of abnormal targets at set specific locations.
[0054] In the specific implementation of generating a secondary alarm signal, the difference from the specific implementation of generating a primary alarm signal is that, under this condition, based on the information that the baseline monitoring node did not sample the abnormal target, since no "confirmed abnormality" or "serious abnormality" mark was generated, the holder of the second terminal device cannot receive feedback information about the abnormal target. Therefore, if any monitoring node other than the baseline monitoring node samples the sampling data of the abnormal target, a secondary alarm signal is generated to the second terminal device.
[0055] As a supplementary implementation of the first-level alarm signal and the second-level alarm signal in the embodiment of the present disclosure, the verification monitoring node samples the sampling data of the abnormal target to be verified, specifically,
[0056] If no sampling data of an abnormal target is sampled within the first preset time range, determining whether to send a first-level alarm signal;
[0057] If so, if no sampling data of the abnormal target is sampled at the verification monitoring node within a second preset time range after the first preset time, an alarm is initiated;
[0058] If not, determine whether to send a secondary alarm signal.
[0059] If so, issue an alarm.
[0060] In this supplementary implementation, since the severity of sending a secondary alarm signal is higher than that of a primary alarm signal, if a secondary alarm signal is sent, there is no need to consider the preset time, and an alarm is directly issued to notify the holder of the second terminal device. In addition, as an indispensable technical feature in the embodiment of the present disclosure, an alarm is directly initiated after the corresponding data of the abnormal target is not sampled in any monitoring node within the preset time after the secondary alarm signal is sent, and the priority of this condition is higher than the sampling data of the abnormal target that should be verified by the verification monitoring node sampling in the supplementary implementation of the primary alarm signal and the secondary alarm signal.
[0061] It should be noted that, in the embodiment of the present disclosure, both the first-level alarm signal and the second-level alarm signal are data for a single abnormal target, and in the embodiment of the present disclosure, one second terminal device is bound to one monitoring area.
[0062] As a further preferred embodiment of the present disclosure, when the second terminal device uploads environmental baseline data to the first terminal device, the environmental baseline data includes at least location data and parameter threshold data, and the parameter threshold data is uploaded at a preset period. It should be noted that the preset period referred to in the embodiment of the present disclosure is a period set by the first terminal device. The period is set according to actual setting needs and is related to the period for performing consistency judgment in the embodiment of the present disclosure. The set preset period can be used as a period for the benchmark monitoring node to perform environmental sampling. For example, if the set preset period is one hour, the second terminal device needs to upload parameter threshold data every hour. Generally speaking, the parameter threshold data is the environmental safety threshold data of a monitoring area, and the sampling of the benchmark monitoring node is performed at one hour, the same as the set period.
[0063] As a further preferred embodiment of the present disclosure, each of the monitoring nodes is provided with a sensor array configured by a multimodal sensing algorithm, which is used to sample corresponding sampling data in the environmental reference data and perform consistency judgment.
[0064] Next, the cloud server in the embodiment of the present disclosure is described, which is specifically provided with a model training module, which obtains the environmental baseline data and the monitoring parameters and generates a training set, and obtains the preset position in the determined abnormal diffusion path after training;
[0065] The dynamic correction module is configured to match and correct the training degree of the model training module according to the alarm signal determined by the first terminal device until the response accuracy reaches a preset ratio.
[0066] When the model training module performs training, it uses time series data and spatial distribution data as input training sets and performs the following training steps:
[0067] Obtaining a first feature of the spatially distributed data of the time series; the first feature being an amount of information change encoded in the current space at a specified moment, the amount of information change including change parameters of the environmental baseline data and the real-time monitoring data in the embodiment of the present disclosure, specifically determined and marked by pixel encoding positions in the image;
[0068] Obtain a first sample, where the first sample is an exhaustive set of samples obtained in a time series when the first feature of the second terminal device is replaced with any other second terminal device;
[0069] A first feature within a preset threshold corresponding to multiple times under the first sample is determined, and a first response sequence is generated according to the first feature. The first response sequence includes a data set of positions of multiple adjacent monitoring nodes corresponding to any time point.
[0070] As a further preferred embodiment of the present disclosure, the environmental reference data includes fuzzy location information, and the fuzzy location information is used to generate all paths traversing the monitoring nodes.
[0071] In the embodiment of the present disclosure, a more detailed explanation of spatial distribution data is provided. Taking the first sample as an example, fuzzy location information is obtained from the environmental baseline data uploaded by different second terminal devices in the first sample. After generating a fuzzy path to the baseline monitoring node from each fuzzy location information, fuzzy time is sampled and the time required for each second terminal device to actually execute the fuzzy path is recorded in the cloud server to obtain sequence characteristics regarding time and space. Therefore, the response sequence includes the fuzzy location and the preset time value of the sampled target arriving at any monitoring node. Afterwards, the environmental baseline data of all second terminal devices are exhaustively replaced with any time sequence to obtain the entire range of the first sample. A preset threshold is determined based on the entire range. As a specific implementation, the preset threshold is set to the statistical distribution range of the preset time values of all fuzzy paths arriving at any monitoring node in any sampled area, thereby obtaining the theoretical statistical probability of any anomaly occurring at any monitoring node. The statistical distribution can adopt Poisson distribution or standard normal distribution, thereby obtaining the change of spatial distribution data at any moment under theoretical statistics, thereby providing data support for the alarm process in the embodiment of the present disclosure.
[0072] When the dynamic correction module performs correction, the following steps are performed:
[0073] Acquire sampling data of abnormal targets from multiple monitoring nodes, and obtain the diffusion characteristics of abnormal targets from one monitoring node to another;
[0074] The diffusion feature is written into the first response sequence to obtain a matching degree. If the matching degree is within a preset threshold, the diffusion feature is written into the first response sequence to correct the monitoring node position.
[0075] According to the implementation of the cloud server in the embodiment of the present disclosure, the theoretical statistical probability of any abnormality occurring at any location can be determined, and based on the theoretical statistical probability, when any secondary alarm signal is issued, the location of the abnormal target can be tracked in the shortest time to ensure the safety of the monitored area.
[0076] Similarly, in the embodiment of the present disclosure, since an exhaustive set of samples consisting of the first sample is collected, the position of the next monitoring node where the abnormal target appears after the position of the monitoring node where the abnormal target last appeared can be obtained, and the abnormal target can be tracked more accurately to further determine the safety of the monitoring area.
[0077] It should be noted that, in the embodiment of the present disclosure, a CNN neural network or a similar forward neural network is used for training the first response sequence. The structure of this type of neural network is a common technical means in this field. In the embodiment of the present disclosure, in addition to obtaining the first feature of the spatial distribution data of the time series as input, other technical means can refer to the existing technology. For the sake of brevity, they will not be repeated here.
[0078] Example 2
[0079] The embodiment of the present disclosure is an application embodiment based on the first embodiment. Specifically, the alarm process is further explained. The first terminal device is held by the regional administrator and the on-site inspector, and the second terminal device is held by the monitoring center.
[0080] If there is an environmental anomaly that requires an early warning, the monitoring center will operate on a dedicated console or mobile terminal. That is, after the monitoring center logs in with the registered account and password, the data management module will only display the environmental information of the corresponding area, including location coordinates, real-time parameters, and safety thresholds. When setting an early warning, the monitoring center only needs to select the corresponding monitoring node, and the monitoring parameter configuration module can organize the location, parameters, and thresholds of the area into a summary of early warning information and transmit it to the benchmark management module. The data is backed up in the benchmark management module, and the backed-up early warning information is then transmitted to the verification module. It is worth noting that the verification module also includes multiple independent on-site verification sub-modules held by inspectors. That is, after the inspector logs in to the mobile terminal with the registered account and password, the on-site verification sub-module only displays the corresponding area and node information that the inspector is responsible for. The node information includes location coordinates and real-time parameters. When the verification module receives the early warning information, it classifies it according to different areas and monitoring nodes and transmits it to the corresponding on-site verification sub-module, and sends an on-site verification reminder when the parameters are abnormal. After the inspector verifies the abnormal data on site, if it is correct, he / she enters "Confirm abnormality" in the on-site management sub-module. This information will be fed back to the verification module and then to the monitoring parameter configuration module, and then sent by the monitoring parameter configuration module to the data management module of the corresponding monitoring center to inform the monitoring center.
[0081] In the disclosed embodiment, the monitoring nodes set by the verification module are not disclosed to the inspectors, but are forwarded via the cloud server as in Example 1. Referring to Example 1, regarding the setting of the verification monitoring nodes, in the disclosed embodiment, the verification monitoring nodes are set to the key locations of the corresponding areas where the inspectors are located. Through this setting, the inspectors' verification can be used as verification monitoring nodes.
[0082] If an inspector discovers an environmental anomaly on-site, but the field verification submodule hasn't received warning information about the abnormal area, the inspector must select the corresponding node information, or abnormal information, within the corresponding field verification submodule. This abnormal information includes the area's location coordinates, real-time parameters, and safety thresholds. The abnormal parameters are directly output based on the current real-time parameters displayed within the field verification submodule. This abnormal information is input by the field verification submodule, fed back to the verification module for aggregation, and then sent to the benchmark management module.
[0083] In the benchmark management module, the abnormal information (location coordinates of the area, real-time parameters, safety thresholds) will be checked with the backed-up warning information (location coordinates of the area, parameter thresholds, warning levels). If the information is consistent, the result will not be output. This step can avoid the problem of information omission caused by failure of the benchmark management module when transmitting the warning information to the on-site verification submodule. For example, the monitoring parameter configuration module has sent the warning information to the benchmark management module, but the information is omitted when the benchmark management module is backed up and transmitted to the verification module, resulting in the inspector failing to receive the warning information of the area. At this time, the abnormal information input can be fed back to the benchmark management module and then checked with the backed-up warning information. If the information is consistent, there is no need to proceed to the next step. If the warning information and the abnormal information in the benchmark management module are inconsistent, the abnormal information of the inconsistent area (location coordinates of the area, real-time parameters, safety thresholds) is output.
[0084] Abnormal information (location coordinates, real-time parameters, and safety thresholds) for inconsistent areas output by the baseline management module is transmitted to the baseline monitoring node. Entering and exiting key locations within the abnormal environmental area requires real-time monitoring and confirmation via multimodal sensors before an alarm is triggered. Sensors collect environmental parameters and then use an algorithm to match them with baseline data in the database. If a match is successful, the abnormal state of the area is recorded and a monitoring record is generated for the day. In this embodiment, the database is pre-loaded with environmental information for all areas, including location coordinates and parameter thresholds. These monitoring records are categorized as "normal" or "abnormal." If an area has abnormal sensor readings that do not exceed the safety threshold, the corresponding area's information is marked as "warning." If an area has abnormal sensor readings that exceed the safety threshold, the corresponding area's information is marked as "severe abnormality." When abnormal information about an inconsistent area is transmitted to the baseline monitoring node, the baseline monitoring node matches the inconsistent area's location coordinates and parameter thresholds with the monitoring record to determine whether the inconsistent area is within the monitoring range. When the benchmark monitoring node determines that the area is "normal", it sends a negative message to the monitoring parameter configuration module. The negative message is "the area is abnormal but not exceeding the standard". When the benchmark monitoring node determines that the area is "abnormal", it sends a confirmation message to the emergency response module. The confirmation message is "the area is abnormal and exceeds the standard". At the same time, the location coordinates and parameter thresholds of the area pre-loaded in the benchmark monitoring node are sent.
[0085] The above description and the accompanying drawings fully illustrate the embodiments of the present disclosure so that those skilled in the art can practice them. Other embodiments may include structural, logical, electrical, process and other changes. The embodiments represent only possible variations. Unless explicitly required, individual components and functions are optional, and the order of operations may vary. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. Moreover, the words used in this application are only used to describe the embodiments and are not used to limit the claims. As used in the description of the embodiments and claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to also include plural forms. Similarly, the term "and / or" as used in this application refers to any and all possible combinations of one or more associated listings. In addition, when used in this application, the term "comprise" and its variations "comprises" and / or comprising refer to the presence of stated features, wholes, steps, operations, elements, and / or modules, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, modules and / or groups thereof. In the absence of further restrictions, an element defined by the statement "comprises a..." does not exclude the presence of other identical elements in the process, method or device that includes the element. In this article, each embodiment may focus on the differences from other embodiments, and the same and similar parts between the various embodiments can be referenced to each other. For the methods, products, etc. disclosed in the embodiments, if they correspond to the method part disclosed in the embodiments, then the relevant parts can be found in the description of the method part.
Claims
1. An automatic alarm system based on Internet of Things environmental monitoring, characterized in that: include: The first terminal device is provided with: A benchmark management module is configured to store environmental benchmark data and also configured to obtain real-time monitoring data; a verification module configured to set a plurality of monitoring nodes and set at least one of the monitoring nodes as a reference monitoring node, each of the monitoring nodes providing data to the reference management module, and generating environmental verification data after verification by the reference monitoring node; an abnormality analysis module configured to obtain the environmental verification data, verify the consistency of the environmental verification data with the environmental benchmark data and the real-time monitoring data, and send an alarm signal to the first terminal device or the second terminal device based on a consistency judgment result; and the second terminal device, which is provided with: a data management module configured to register the environmental baseline data and the monitoring parameters to the first terminal device and receive an alarm signal from the first terminal device; And the cloud server, which is set up with: A model training module, which obtains the environmental baseline data and the monitoring parameters and generates a training set, and obtains a preset position in the determined abnormal diffusion path after training; The dynamic correction module is configured to match and correct the training degree of the model training module according to the alarm signal determined by the first terminal device until the response accuracy reaches a preset ratio.
2. The automatic alarm system based on Internet of Things environmental monitoring according to claim 1 is characterized in that: When the abnormality analysis module performs the consistency judgment, the following steps are included: Determining the consistency between the environmental baseline data and the monitoring parameters, and determining normal monitoring targets; Determining whether the judgment data collected by the reference monitoring node includes abnormal data other than the normal monitoring target; If so, an alarm signal is returned to the benchmark management module; If not, the sampling data of the monitoring blind area that has not been sampled by the reference monitoring node is determined, and a first-level alarm signal is generated and returned to the first terminal device and the second terminal device.
3. The automatic alarm system based on Internet of Things environmental monitoring according to claim 2 is characterized in that: When the abnormality analysis module performs the consistency judgment, after generating a first-level alarm signal and returning it to the first terminal device and the second terminal device, the following steps are further included: After any monitoring node other than the reference monitoring node samples abnormal data, a secondary alarm signal is generated and returned to the first terminal device and the second terminal device.
4. The automatic alarm system based on Internet of Things environmental monitoring according to claim 1 is characterized in that: When the model training module performs training, it uses time series data and spatial distribution data as input training sets and performs the following training steps: Obtaining a first feature of the spatial distribution data of the time series; Obtain a first sample, where the first sample is an exhaustive set of samples obtained in a time series when the first feature of the second terminal device is replaced with any other second terminal device; A first feature within a preset threshold corresponding to multiple times under the first sample is determined, and a first response sequence is generated according to the first feature. The first response sequence includes a data set of positions of multiple adjacent monitoring nodes corresponding to any time point.
5. The automatic alarm system based on Internet of Things environmental monitoring according to claim 4 is characterized in that: When the dynamic correction module performs correction, the following steps are performed: Acquire sampling data of abnormal targets from multiple monitoring nodes, and obtain the diffusion characteristics of abnormal targets from one monitoring node to another; The diffusion feature is written into the first response sequence to obtain a matching degree. If the matching degree is within a preset threshold, the diffusion feature is written into the first response sequence to correct the monitoring node position.
6. An automatic alarm system based on Internet of Things environmental monitoring according to any one of claims 1 to 4, characterized in that: Each of the monitoring nodes is provided with a sensor array configured by a multimodal sensing algorithm, which is used to sample corresponding sampled data in the environmental reference data and perform consistency judgment.
7. The automatic alarm system based on Internet of Things environmental monitoring according to claim 1 is characterized in that: The environmental reference data includes fuzzy location information, and the fuzzy location information is used to generate all paths traversing the monitoring nodes.
8. The automatic alarm system based on Internet of Things environmental monitoring according to claim 1 is characterized in that: The environmental reference data includes device identification data and timestamp data, and when the environmental reference data is uploaded to the first terminal device via the second terminal device, the timestamp data is uploaded at a preset period.