Internet-of-things-driven intelligent playground equipment state real-time monitoring method and system

Abstract

The invention discloses an Internet of Things-driven intelligent playground equipment state real-time monitoring method and system, and mainly relates to the technical field of Internet of Things monitoring, and the method comprises the steps: enabling a target intelligent playground equipment set to access an Internet of Things gateway through a wireless communication module, and obtaining an equipment collection data sequence set; obtaining an equipment operation state index sequence set; determining a target intelligent playground equipment data feedback coefficient set through multiple rounds of information interaction, complementation and completion; adjacent state iterative memory interaction is carried out, and an equipment operation state feedback coefficient set is determined; and carrying out equipment state real-time monitoring on the target intelligent playground equipment set based on the optimized monitoring resource allocation scheme. The beneficial effects of the invention are that the technical problem of low monitoring response timeliness caused by lack of real-time monitoring of the state of the intelligent playground equipment in the prior art is solved, and the technical effect of improving the monitoring response timeliness and reliability is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things monitoring, in particular to an Internet of Things driven real-time monitoring method and system for intelligent playground equipment state. BACKGROUND

[0002] With the development of smart campus and intelligent sports facilities, intelligent playground equipment has gradually become popular, such as intelligent treadmills, intelligent jump ropes, basketball counters, etc. These devices usually collect data and monitor the state through Internet of Things technology to record and analyze students' sports behavior and equipment running state. In the prior art, most of the data collection and monitoring of intelligent playground equipment relies on single data upload and statistical analysis, mainly on fixed interval data collection, and simple state judgment and alarm prompt through the center platform. However, the real-time, accuracy and adaptive monitoring ability of the device running state and data anomaly are still obviously insufficient. When only relying on threshold or basic statistical method to analyze the device data, it is impossible to reliably identify the abnormal reason, and then to report or miss the abnormal device state.

[0003] There is a technical problem of lacking real-time monitoring of the state of intelligent playground equipment in the prior art, resulting in low timeliness of monitoring response. SUMMARY

[0004] The present application provides an Internet of Things driven real-time monitoring method and system for intelligent playground equipment state, which is used to solve the technical problem of lacking real-time monitoring of the state of intelligent playground equipment in the prior art, resulting in low timeliness of monitoring response.

[0005] In view of the above problems, the present application provides an Internet of Things driven real-time monitoring method and system for intelligent playground equipment state.

[0006] In a first aspect of the present application, an Internet of Things driven real-time monitoring method for intelligent playground equipment state is provided, which comprises: accessing a target intelligent playground equipment set to an Internet gateway through a wireless communication module, transmitting the data collected by the target intelligent playground equipment set to a center Internet platform through a wireless communication module, and obtaining a device collection data sequence set; calling a cloud analysis engine of the center Internet platform, performing running state analysis on the target intelligent playground equipment set according to a preset monitoring resource allocation scheme and a preset index, and obtaining a device running state index sequence set; based on the device collection data sequence set, performing data anomaly active reasoning analysis, and through multiple rounds of information interaction, determining a target intelligent playground equipment data feedback coefficient set; The device running state index sequence set is subjected to adjacent state iterative memory interaction to determine a device running state feedback coefficient set; The preset monitoring resource allocation scheme is subjected to device adaptive correction according to the target smart playground device data feedback coefficient set and the device running state feedback coefficient set, with the callable resources of the central Internet of Things platform as constraints, to determine an optimized monitoring resource allocation scheme, and the target smart playground device set is subjected to device state real-time monitoring based on the optimized monitoring resource allocation scheme.

[0007] In a second aspect of the present application, an Internet of Things driven smart playground device state real-time monitoring system is provided, which comprises: A device collected data sequence set obtaining module is configured to connect a target smart playground device set to an Internet of Things gateway through a wireless communication module, and transmit data collected by the target smart playground device set to a central Internet of Things platform through a wireless communication module to obtain a device collected data sequence set. A running state analysis module is configured to call a cloud analysis engine of the central Internet of Things platform, and analyze the running state of the target smart playground device set according to a preset monitoring resource allocation scheme and a preset index to obtain a device running state index sequence set. A data feedback coefficient set determining module is configured to perform data anomaly active reasoning analysis based on the device collected data sequence set, and complete through multiple rounds of information interaction to determine a target smart playground device data feedback coefficient set. A device running state feedback coefficient set determining module is configured to perform adjacent state iterative memory interaction on the device running state index sequence set to determine a device running state feedback coefficient set. A device state real-time monitoring module is configured to perform device adaptive correction on the preset monitoring resource allocation scheme according to the target smart playground device data feedback coefficient set and the device running state feedback coefficient set, with the callable resources of the central Internet of Things platform as constraints, to determine an optimized monitoring resource allocation scheme, and the target smart playground device set is subjected to device state real-time monitoring based on the optimized monitoring resource allocation scheme.

[0008] One or more technical solutions provided in the present application have at least the following technical effects or advantages: The application obtains the device collection data sequence set through the target smart playground equipment set access through the wireless communication module to the Internet gateway, transmits the data collected by the target smart playground equipment set to the center Internet platform through the wireless communication module, calls the cloud analysis engine of the center Internet platform, analyzes the running state of the target smart playground equipment set according to the preset monitoring resource allocation scheme and the preset index, obtains the device running state index sequence set, and then performs data anomaly active reasoning analysis based on the device collection data sequence set, and determines the target smart playground equipment data feedback coefficient set through multi-round information interaction completion, then performs adjacent state iteration memory interaction on the device running state index sequence set, determines the device running state feedback coefficient set, takes the callable resources of the center Internet platform as a constraint, and performs device adaptive correction on the preset monitoring resource allocation scheme according to the target smart playground equipment data feedback coefficient set and the device running state feedback coefficient set, determines the optimized monitoring resource allocation scheme, and performs device state real-time monitoring on the target smart playground equipment set based on the optimized monitoring resource allocation scheme. The technical effects of fitting the actual allocation of the Internet of Things monitoring resources, efficiently and real-timely monitoring the state of the smart playground equipment, reducing the monitoring delay and improving the accuracy are achieved. BRIEF DESCRIPTION OF DRAWINGS

[0009] FIG. 1 is a flowchart of an Internet of Things driven smart playground equipment state real-time monitoring method provided by an embodiment of the application. Figure 1 FIG. 2 is a schematic diagram of an Internet of Things driven smart playground equipment state real-time monitoring system structure provided by an embodiment of the application.

[0010] FIG. 3 is a schematic diagram of an Internet of Things driven smart playground equipment state real-time monitoring system structure provided by an embodiment of the application. Figure 2 FIG. 4 is a schematic diagram of an Internet of Things driven smart playground equipment state real-time monitoring system structure provided by an embodiment of the application.

[0011] Reference signs shown in the drawings: The device collection data sequence set obtaining module 11, the running state analysis module 12, the data feedback coefficient set determining module 13, the device running state feedback coefficient set determining module 14, and the device state real-time monitoring module 15. DETAILED DESCRIPTION

[0012] The application will be further described below in connection with specific embodiments. It should be understood that these embodiments are only used to illustrate the application and not intended to limit the scope of the application. Furthermore, it should be understood that those skilled in the art can make various modifications or changes to the application after reading the content of the application, and these equivalent forms also fall within the scope defined by the appended claims. It should be noted that the terms "comprise" and "have" are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server comprising a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices.

[0013] It should be noted in advance that the acquisition and processing of all information or data in the application are carried out in compliance with the corresponding data protection regulations and policies, and with the authorization given by the owner of the corresponding device.

[0014] Embodiment one, as shown in the accompanying Figure 1 The application provides an Internet of Things driven intelligent playground equipment state real-time monitoring method, wherein the method comprises: Step S100: connecting a target intelligent playground equipment set to an Internet of Things gateway through a wireless communication module, transmitting data collected by the target intelligent playground equipment set to a central Internet of Things platform through the wireless communication module, and obtaining a device collected data sequence set; In one possible embodiment, the target intelligent playground equipment set refers to various intelligent devices monitored or managed in the playground, such as intelligent treadmills, skipping rope counters, heart rate monitoring devices, etc. The wireless communication module is a communication unit built in the device, which is used for data transmission with the Internet of Things gateway through a wireless protocol, wherein the wireless protocol includes Wi-Fi, LoRa, NB-IoT, etc. The central Internet of Things platform is a platform for cloud centralized processing and storage of data, which can call an analysis engine, a database and a monitoring application. The device collected data sequence set reflects the real-time running condition of the target intelligent playground equipment set.

[0015] Preferably, all target intelligent playground equipment sets in the playground are connected to the Internet of Things gateway through the wireless communication module, completing the connection of the devices and the network. In the running state, the target intelligent playground equipment set continuously collects its own state data (such as the number of uses, sensor readings, heart rate, signal transmission log, etc.), and uploads it to the Internet of Things gateway in real time through the wireless communication module. The Internet of Things gateway forwards the collected data to the central Internet of Things platform, forming a structured device collected data sequence set at the platform end. This achieves the technical effect of providing an original data basis for subsequent running state analysis, abnormal monitoring and active reasoning.

[0016] Step S200: calling a cloud analysis engine of the central Internet of Things platform, performing running state analysis on the target smart playground equipment set according to a preset monitoring resource allocation scheme and a preset index, and obtaining a device running state index sequence set; Further, the preset index at least includes device accuracy, response speed, signal strength and signal quality.

[0017] In an embodiment of the present application, the cloud analysis engine is used to perform running state analysis on the target smart playground equipment set according to the preset monitoring resource allocation scheme of the central Internet of Things platform and the preset index reflecting the device state, and obtain the device running state index sequence set. The preset monitoring resource allocation scheme includes allocation of data collection frequency of different smart playground equipment capable of reflecting the device state running, allocation of analysis calculation capability, and allocation of upload data network bandwidth, etc.

[0018] The preset index at least includes device accuracy, response speed, signal strength and signal quality. The device accuracy reflects the closeness of the device collected data to the true value, for example, the error range of the heart rate monitoring device measurement value to the true heart rate. The response speed is the reaction time of the smart playground equipment to the operation or environmental change, such as the delay of the treadmill sensor in detecting user action. The signal strength is the device wireless signal receiving capability, which affects the data transmission stability. The signal quality includes data integrity, bit error rate, etc., reflecting the transmission reliability. The device running state index sequence set is a set of index data recorded in time sequence generated by the cloud analysis engine during the running process of the device, such as signal strength, response speed, etc.

[0019] Preferably, the device collected data sequence set obtained in step S100 is input into the cloud analysis engine of the central Internet of Things platform. The cloud analysis engine performs comprehensive running state analysis on each device according to the preset monitoring resource allocation scheme, such as high-frequency sampling for high-activity area equipment and low-frequency sampling for low-activity area equipment, and the preset index. The high-activity area equipment can include the skipping rope area and the basketball area, which are frequently used and have high motion intensity, and require high-frequency sampling (such as 1-5 times per second) to capture rapidly changing data. The low-activity area equipment can include the leisure area and the walking area, which have low activity frequency, and can use low-frequency sampling (such as 1 time per minute) to save transmission bandwidth and computing resources.

[0020] Optionally, the specific operational status analysis operation involves calculating index values ​​for the collected data of each smart playground device according to a time series, and then comparing them with preset standards to determine operational performance. For example, if the sampling value deviation of the device's heart rate sensor exceeds ±5%, the marking accuracy decreases; if the signal strength is below -80 dBm, communication is considered unstable; and if the response time exceeds 1 second, the response speed is insufficient. Furthermore, the analysis results are sequentially sorted according to time to generate a set of device operational status index sequences.

[0021] Step S300: Based on the data sequence set collected by the device, perform proactive reasoning analysis of data anomalies, and complete the data feedback coefficient set of the target smart playground device through multiple rounds of information interaction; Furthermore, based on the data sequence set collected by the device, proactive reasoning analysis of data anomalies is performed, and through multiple rounds of information interaction and supplementation, the set of data feedback coefficients for the target smart playground device is determined. In this embodiment, step S300 further includes: The set of data sequences collected by the device is traversed to identify abnormal data and determine the set of data sequences collected by abnormal devices. According to the preset inference association constraints, the abnormal device data collection sequence set is traversed to perform inference association data collection on the device data collection sequence set to obtain the abnormal device data collection inference association neighborhood sequence set. Based on the set of data sequences collected by the abnormal device and the set of neighboring sequences related to the data inference of the abnormal device, active reasoning of the mapping data abnormality is performed to obtain a set of active reasoning question sequences and a set of active reasoning result sequences. Based on the set of active reasoning question sequences and the set of active reasoning result sequences, multiple rounds of information interaction and completion are performed to determine the set of data feedback coefficients for the target smart playground equipment.

[0022] In one possible embodiment, the target smart playground equipment data feedback coefficient set is used to analyze the data collected during the operation of the target smart playground equipment, such as real-time operating data like heart rate, height, and number of rope skips, reflecting the equipment status. The data sequence collected by the equipment is used to identify abnormal data using a threshold method, such as setting upper and lower limits for each type of sensor or collected indicator: heart rate <30 or >220 beats / minute, rope skipping count <0 or >100 skips / minute. For example, for the rope skipping count sensor data sequence [12, 15, 13, -1, 14], the threshold method directly identifies -1 as an abnormal data point, and can identify continuously changing abnormal points as the collected data is continuously recorded. Then, the abnormal data is grouped into an abnormal equipment collected data sequence set: [-1].

[0023] After obtaining the set of data sequences collected by the abnormal devices, related data within the set are collected according to preset inference association constraints. These constraints include adding data collected by devices belonging to the same user as the abnormal device data to the abnormal device data inference association neighborhood, thus providing data support for subsequent inference. Since the abnormal device data inference association neighborhood belongs to the same user, this provides data support for subsequent analysis of whether the anomalies in the collected data are due to problems with the smart playground equipment itself or individual user differences.

[0024] Based on the set of data sequences collected by the abnormal devices, proactive reasoning questions are automatically generated. Using the set of related neighboring sequences from the abnormal device data as the corresponding reasoning data material, the proactive reasoning questions are analyzed to identify possible causes. This provides data support for subsequent analysis of the smart playground's equipment status from a working data perspective. For example, an abnormal jump rope count of -1, after neighboring data analysis reveals that the preceding and following data are normal and other devices are counting normally synchronously, the proactive reasoning question is the cause of the abnormal jump rope count. The proactive reasoning result is either a momentary sensor failure or a user operation error.

[0025] Furthermore, by performing multiple rounds of information interaction and completion on the obtained set of active reasoning question sequences and the set of active reasoning result sequences, the operational status of the smart playground equipment is analyzed from a holistic perspective, determining the corresponding set of data feedback coefficients for the target smart playground equipment. The set of data feedback coefficients for the target smart playground equipment reflects the status of the equipment. A larger data feedback coefficient indicates a better operational status, and the required monitoring resources can be reduced accordingly. Conversely, a smaller data feedback coefficient indicates a higher probability of abnormal operational status, and the required monitoring resources can be increased accordingly.

[0026] Furthermore, based on the set of data sequences collected by the abnormal device and the set of neighboring sequences related to the data collected by the abnormal device, proactive reasoning of the mapped data abnormalities is performed to obtain a set of proactive reasoning question sequences and a set of proactive reasoning result sequences. In this embodiment, step S300 further includes: The active reasoning analyzer is used to identify the anomaly type of the data sequence set collected by the abnormal device, and an active reasoning question sequence set is generated. Using the set of neighboring sequences of data collected by the abnormal device as the reasoning analysis data, the active reasoning question parser is used to perform abnormal active reasoning on the corresponding neighboring sequences of data collected by the abnormal device in the active reasoning question sequence set and the reasoning analysis data, respectively, to obtain the set of active reasoning result sequences.

[0027] In one possible embodiment, the active inference analyzer is used to identify the anomaly type based on the data collected by the abnormal device, and generate a corresponding active inference question based on the identification result. Preferably, multiple samples of abnormal device data and corresponding multiple samples of active inference questions are acquired as training data. The framework built on a feedforward neural network is supervised and trained using the training data until convergence is achieved, obtaining a set of trained active inference question sequences. For example, based on the abnormal count data of the device jump rope, the corresponding active inference question is determined to be whether the abnormal jump rope count is due to sensor failure or abnormal user wearing.

[0028] Furthermore, since the inference association neighborhood of the abnormal device data includes user data associated with the abnormal device data, the set of sequences of inference association neighborhoods of abnormal device data is used as inference analysis data. An abnormal reasoning problem parser, capable of performing problem analysis based on proactive reasoning questions and inference analysis data, is used to perform proactive abnormal reasoning on the corresponding abnormal device data inference association neighborhoods in the proactive reasoning question sequence set and the inference analysis data, respectively, to obtain the set of proactive reasoning result sequences. For example, regarding the proactive reasoning question of whether the abnormal jump rope count is due to sensor failure or user wearing abnormalities, if the abnormal jump rope count corresponds to a normal device and surrounding devices in the abnormal device data inference association neighborhood, the reasoning result tends to be a single device failure; if the abnormal heart rate is synchronously abnormal with the data in the abnormal device data inference association neighborhood, the reasoning result may be caused by high-intensity user exercise. Transforming abnormal data from a single-point anomaly into an explainable problem and reasoning result facilitates subsequent multi-round information interaction and data feedback coefficient calculation.

[0029] Preferably, multiple sample abnormal device collection data inference association neighborhoods and multiple sample active reasoning questions, along with corresponding multiple sample active reasoning results, are acquired as parser training data. This parser training data is divided into a training set and a validation set. The training set is used to train a framework built on a convolutional neural network, and the validation set is used to supervise the framework using the multiple sample abnormal device collection data inference association neighborhoods and multiple sample active reasoning questions, obtaining multiple validation active reasoning results. Then, the multiple validation active reasoning results are compared with the multiple sample active reasoning results. When the probability of matching is greater than or equal to a preset probability known to those skilled in the art, the validation is successful, and the trained active reasoning question parser is obtained.

[0030] Furthermore, based on the set of active reasoning question sequences and the set of active reasoning result sequences, multiple rounds of information interaction and completion are performed to determine the set of data feedback coefficients for the target smart playground equipment. In this embodiment, step S300 further includes: Extract the first active reasoning question sequence and the first active reasoning result sequence from the active reasoning question sequence set and the active reasoning result sequence set; A mapping interaction log is generated by processing the first active reasoning question sequence and the first active reasoning result sequence to obtain the first active reasoning mapping interaction log sequence. The first round of information interaction completion is performed on the first active reasoning mapping interaction log sequence to obtain the first round of information interaction vector; The first round of information interaction vector is used to complete the information interaction of the second active reasoning mapping interaction log in the first active reasoning mapping interaction log sequence, thereby generating the second round of information interaction vector, and so on, to obtain multiple rounds of information interaction vectors. The device status feedback analysis is performed on the multi-round information interaction vectors to determine the first smart playground device data feedback coefficient, and the first smart playground device data feedback coefficient is added to the target smart playground device data feedback coefficient set.

[0031] Furthermore, the first round of information interaction completion is performed on the first active reasoning mapping interaction log sequence to obtain the first round of information interaction vector. In this embodiment, step S300 further includes: Extract the first active reasoning mapping interaction log from the first active reasoning mapping interaction log sequence, and perform first-order keyword recognition on the first active reasoning mapping interaction log according to preset device status keywords to obtain the first first-order keyword recognition result. According to the preset nearest neighbor similarity threshold, the first-order keyword identification result is supplemented by second-order derivatives to obtain the first-order second-order derivative keyword supplementation result; The first-order keyword recognition result, the first-order second-order derived keyword supplementation result, and the preset device status keyword are added to the initially empty vector to generate the first round of information interaction vector.

[0032] In one possible embodiment, a first active reasoning question sequence and a first active reasoning result sequence are extracted from the active reasoning question sequence set and the active reasoning result sequence set, respectively. The first active reasoning question sequence and the first active reasoning result sequence are each any one active reasoning question and first active reasoning result from the active reasoning question sequence set and the active reasoning result sequence set, respectively. The question and result are mapped and logged to generate a first active reasoning mapping interaction log sequence. Preferably, the first active reasoning mapping interaction log includes a question description, reasoning conclusion, timestamp, and exception type, etc.

[0033] The first active reasoning mapping interaction log is extracted from the first active reasoning mapping interaction log sequence. Then, according to preset device status keywords associated with device status pre-defined by those skilled in the art, keywords in the first active reasoning mapping interaction log whose cosine similarity to the preset device status keywords meets a preset similarity threshold are extracted to obtain the first-order keyword recognition result. For example, the first-order keyword recognition result may not directly reflect key feature words of the device status, such as "sensor failure," "weak signal," or "abnormal jump rope count."

[0034] According to a preset nearest neighbor similarity threshold set by those skilled in the art, words whose similarity to the first-order first-level keyword recognition result meets the preset nearest neighbor similarity threshold are added to the first-order second-order derived keyword supplementary result, thereby expanding the keywords and improving the accuracy of recognition. For example, "sensor failure" may lead to "device restart" or "abnormal data fluctuation".

[0035] Then, the first-order keyword recognition result, the first-order second-order derived keyword supplementation result, and the preset device status keywords are added to the initially empty vector to generate the first-round information interaction vector. This first-round information interaction vector is used for subsequent information exchange regarding the overall operation of the smart playground equipment, converting text information into vector space data.

[0036] Furthermore, based on the same principle as obtaining the first round of interaction vector, the second active reasoning mapping interaction log in the first active reasoning mapping interaction log sequence is subjected to first-order keyword recognition and second-order derivation supplementation based on the first round of information interaction vector, and the obtained results are added to the first round of interaction vector for updating and supplementation to obtain the second round of information interaction vector. In this way, multiple rounds of interaction are performed on the first active reasoning mapping interaction log sequence to obtain the multiple rounds of information interaction vector.

[0037] Preferably, by utilizing a device data feedback coefficient recognizer to analyze the multi-round information interaction vectors, the first smart playground device data feedback coefficient is determined, and then added to the target smart playground device data feedback coefficient set. Optionally, by acquiring multiple sample multi-round information interaction vectors and multiple sample smart playground device data feedback coefficients, the device data feedback coefficient recognizer is constructed based on the same principle as obtaining an active reasoning problem parser. By transforming a single abnormal reasoning result into a multi-dimensional, multi-round information interaction result, a comprehensive quantitative analysis of the device's abnormal state is achieved, thereby improving the reliability of understanding the device's state and ultimately enhancing monitoring reliability.

[0038] Step S400: Perform adjacency state iterative memory interaction on the set of equipment operating status index sequences to determine the set of equipment operating status feedback coefficients; Furthermore, the device operating status index sequence set is subjected to adjacency state iterative memory interaction to determine the device operating status feedback coefficient set. In this embodiment, step S400 further includes: The first and second equipment operating status indicators of each equipment operating status indicator sequence in the equipment operating status indicator sequence set are extracted to obtain the first equipment operating status indicator set and the corresponding second equipment operating status indicator set. Similarity identification is performed on the first equipment operating status indicator set and the corresponding second equipment operating status indicator set, and an adjacency state matrix is ​​constructed on the similarity identification result set to obtain the first adjacency state matrix set. The first adjacency state matrix set is used to perform memory interaction on the second device operation status index set to obtain the second interactive device operation status index set. Based on the second set of interactive device operation status indicators, the adjacency state iterative memory interaction is performed on the third set of device operation status indicators in the set of device operation status indicator sequences, and so on, until the iterative memory interaction set of device operation status indicators is obtained. Based on the set of operating status indicators of the iterative memory interaction device, the device operating status is analyzed to determine the set of device operating status feedback coefficients.

[0039] In one embodiment of this application, a first device operating status indicator and a second device operating status indicator are extracted from each device operating status indicator sequence in the device operating status indicator sequence set. Using the cosine similarity calculation formula, the similarity of the same type of indicators between the first and second device operating status indicators is calculated to obtain a corresponding similarity recognition result set. The similarity recognition result set is then normalized within the set, and the normalized results are filled into an initially empty matrix to obtain a first adjacency state matrix set. This first adjacency state matrix set reflects the association relationship of the same type of indicators between each first and second device operating status indicator. Furthermore, graph convolution operations are performed on the second device operating status indicator set using the first adjacency state matrix set to obtain a second interactive device operating status indicator set. This second interactive device operating status indicator set implicitly contains the data of the corresponding first interactive device operating status indicators.

[0040] Furthermore, based on the second set of interactive device operating status indicators, an adjacency state iterative memory interaction is performed on the third set of device operating status indicators in the set of device operating status indicator sequences. Based on the same principle, this process is repeated until the end of each device operating status indicator sequence is reached, obtaining an iterative memory interaction set of device operating status indicators. Using the same principle as obtaining the device data feedback coefficient recognizer, a device operating status analyzer is constructed. The iterative memory interaction set of device operating status indicators is input into the device operating status analyzer for analysis, obtaining a set of device operating status feedback coefficients. The device operating status feedback coefficient is a coefficient that quantifies the reliability and anomaly risk of each device's operating status after comprehensive iterative memory interaction of device indicator information. The larger the device operating status feedback coefficient, the better the status of the corresponding playground equipment, and the lower the required monitoring resources. The smaller the device operating status feedback coefficient, the higher the probability of anomalies in the corresponding playground equipment, and the higher the required monitoring resources. By utilizing adjacency state iteration and memory interaction, single-point, isolated device status information is expanded to a global perspective, achieving the technical effect of improving anomaly identification capabilities and providing reliable data support for subsequent adaptive monitoring resource allocation.

[0041] Step S500: Using the available resources of the central IoT platform as constraints, the preset monitoring resource allocation scheme is adaptively modified according to the data feedback coefficient set and the equipment operation status feedback coefficient set of the target smart playground equipment to determine an optimized monitoring resource allocation scheme, and the equipment status of the target smart playground equipment set is monitored in real time based on the optimized monitoring resource allocation scheme.

[0042] In one embodiment, the callable resources of the central IoT platform refer to the computing resources, storage resources, bandwidth, and sensor scheduling capabilities within the central IoT platform that can be used for device status monitoring. The number of callable resources is limited, therefore optimization is necessary when allocating monitoring tasks.

[0043] According to weights pre-set by those skilled in the art, the set of data feedback coefficients and the set of equipment operation status feedback coefficients of the target smart playground equipment are weighted and calculated to obtain a set of equipment resource feedback coefficients. The sum of each equipment resource feedback coefficient in the set of equipment resource feedback coefficients and the total coefficient of the set of equipment resource feedback coefficients is calculated, and the difference between the calculated result and 1 is used as a resource allocation coefficient to obtain a set of resource allocation coefficients. The larger the resource allocation coefficient, the higher the corresponding resource allocation. The available resources are multiplied by the resource allocation coefficients in the resource allocation coefficient set to obtain the resources allocated to each smart playground equipment, thereby correcting the preset monitoring resource allocation scheme to obtain the optimized monitoring resource allocation scheme. Resources are allocated to the central IoT platform according to the optimized monitoring resource allocation scheme, and then the allocated central IoT platform is used to monitor the equipment status of the target smart playground equipment set. By converting feedback coefficients into actual operational resource allocation strategies, closed-loop control is achieved. Furthermore, through adaptive monitoring and real-time analysis, the central IoT platform can intelligently schedule resources based on equipment status and data quality, significantly improving monitoring efficiency and anomaly response capabilities.

[0044] In summary, the embodiments of this application have at least the following technical effects: This application connects a set of target smart playground equipment to an IoT gateway via a wireless communication module. The wireless communication module transmits the data collected by the target smart playground equipment to a central IoT platform, obtaining a set of equipment data sequences. Then, the cloud-based analysis engine of the central IoT platform is invoked to analyze the operational status of the target smart playground equipment according to a preset monitoring resource allocation scheme and preset indicators, obtaining a set of equipment operational status indicator sequences. Based on the equipment data sequence sets, proactive anomaly reasoning analysis is performed, and through multiple rounds of information interaction and supplementation, a set of data feedback coefficients for the target smart playground equipment is determined. Then, the set of equipment operational status indicator sequences undergoes adjacency state iterative memory interaction to determine the set of equipment operational status feedback coefficients. Constrained by the available resources of the central IoT platform, the preset monitoring resource allocation scheme is adaptively adjusted based on the set of target smart playground equipment data feedback coefficients and the set of equipment operational status feedback coefficients to determine an optimized monitoring resource allocation scheme. Based on the optimized monitoring resource allocation scheme, the equipment status of the target smart playground equipment is monitored in real time. This achieves the technical effects of realistically allocating IoT monitoring resources, efficiently and in real-time monitoring of the smart playground equipment status, reducing monitoring latency, and improving accuracy.

[0045] Example 2, based on the same inventive concept as the IoT-driven real-time monitoring method for smart playground equipment status in the foregoing examples, as shown in the appendix. Figure 2As shown, this application provides an IoT-driven smart playground equipment status real-time monitoring system. The system and method embodiments in this application are based on the same inventive concept. The system includes: The device data sequence set acquisition module 11 is used to connect the target smart playground device set to the Internet of Things gateway through the wireless communication module, and transmit the data collected by the target smart playground device set to the central Internet of Things platform through the wireless communication module to obtain the device data sequence set; The operation status analysis module 12 is used to call the cloud analysis engine of the central Internet of Things platform to perform operation status analysis on the target smart playground equipment set according to the preset monitoring resource allocation scheme and preset indicators, and obtain a set of equipment operation status indicator sequences. The data feedback coefficient set determination module 13 is used to perform proactive reasoning analysis of data anomalies based on the data sequence set collected by the device, and to determine the data feedback coefficient set of the target smart playground device through multiple rounds of information interaction and supplementation. The equipment operation status feedback coefficient set determination module 14 is used to perform adjacent state iterative memory interaction on the equipment operation status index sequence set to determine the equipment operation status feedback coefficient set. The real-time equipment status monitoring module 15 is used to perform equipment adaptive correction on the preset monitoring resource allocation scheme based on the data feedback coefficient set and the equipment operation status feedback coefficient set of the target smart playground equipment, constrained by the available resources of the central IoT platform, to determine an optimized monitoring resource allocation scheme, and to perform real-time monitoring of the equipment status of the target smart playground equipment set based on the optimized monitoring resource allocation scheme.

[0046] Furthermore, the preset indicators include at least device accuracy, response speed, signal strength, and signal quality.

[0047] Furthermore, the data feedback coefficient set determination module 13 is used to perform the following steps: The set of data sequences collected by the device is traversed to identify abnormal data and determine the set of data sequences collected by abnormal devices. According to the preset inference association constraints, the abnormal device data collection sequence set is traversed to perform inference association data collection on the device data collection sequence set to obtain the abnormal device data collection inference association neighborhood sequence set. Based on the set of data sequences collected by the abnormal device and the set of neighboring sequences related to the data inference of the abnormal device, active reasoning of the mapping data abnormality is performed to obtain a set of active reasoning question sequences and a set of active reasoning result sequences. Based on the set of active reasoning question sequences and the set of active reasoning result sequences, multiple rounds of information interaction and completion are performed to determine the set of data feedback coefficients for the target smart playground equipment.

[0048] Furthermore, the data feedback coefficient set determination module 13 is used to perform the following steps: The active reasoning analyzer is used to identify the anomaly type of the data sequence set collected by the abnormal device, and an active reasoning question sequence set is generated. Using the set of neighboring sequences of data collected by the abnormal device as the reasoning analysis data, the active reasoning question parser is used to perform abnormal active reasoning on the corresponding neighboring sequences of data collected by the abnormal device in the active reasoning question sequence set and the reasoning analysis data, respectively, to obtain the set of active reasoning result sequences.

[0049] Furthermore, the data feedback coefficient set determination module 13 is used to perform the following steps: Extract the first active reasoning question sequence and the first active reasoning result sequence from the active reasoning question sequence set and the active reasoning result sequence set; A mapping interaction log is generated by processing the first active reasoning question sequence and the first active reasoning result sequence to obtain the first active reasoning mapping interaction log sequence. The first round of information interaction completion is performed on the first active reasoning mapping interaction log sequence to obtain the first round of information interaction vector; The first round of information interaction vector is used to complete the information interaction of the second active reasoning mapping interaction log in the first active reasoning mapping interaction log sequence, thereby generating the second round of information interaction vector, and so on, to obtain multiple rounds of information interaction vectors. The device status feedback analysis is performed on the multi-round information interaction vectors to determine the first smart playground device data feedback coefficient, and the first smart playground device data feedback coefficient is added to the target smart playground device data feedback coefficient set.

[0050] Furthermore, the data feedback coefficient set determination module 13 is used to perform the following steps: Extract the first active reasoning mapping interaction log from the first active reasoning mapping interaction log sequence, and perform first-order keyword recognition on the first active reasoning mapping interaction log according to preset device status keywords to obtain the first first-order keyword recognition result. According to the preset nearest neighbor similarity threshold, the first-order keyword identification result is supplemented by second-order derivatives to obtain the first-order second-order derivative keyword supplementation result; The first-order keyword recognition result, the first-order second-order derived keyword supplementation result, and the preset device status keyword are added to the initially empty vector to generate the first round of information interaction vector.

[0051] Furthermore, the device operating status feedback coefficient set determination module 14 is used to perform the following steps: The first and second equipment operating status indicators of each equipment operating status indicator sequence in the equipment operating status indicator sequence set are extracted to obtain the first equipment operating status indicator set and the corresponding second equipment operating status indicator set. Similarity identification is performed on the first equipment operating status indicator set and the corresponding second equipment operating status indicator set, and an adjacency state matrix is ​​constructed on the similarity identification result set to obtain the first adjacency state matrix set. The first adjacency state matrix set is used to perform memory interaction on the second device operation status index set to obtain the second interactive device operation status index set. Based on the second set of interactive device operation status indicators, the adjacency state iterative memory interaction is performed on the third set of device operation status indicators in the set of device operation status indicator sequences, and so on, until the iterative memory interaction set of device operation status indicators is obtained. Based on the set of operating status indicators of the iterative memory interaction device, the device operating status is analyzed to determine the set of device operating status feedback coefficients.

[0052] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0053] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0054] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for real-time monitoring of the state of smart playground equipment driven by the Internet of Things, characterized in that, The method comprises: Accessing the target smart playground equipment set through a wireless communication module to an Internet gateway, transmitting data collected by the target smart playground equipment set to a central Internet platform through a wireless communication module, and obtaining a device collection data sequence set; Calling a cloud analysis engine of the central Internet platform, performing running state analysis on the target smart playground equipment set according to a preset monitoring resource allocation scheme and a preset index, and obtaining a device running state index sequence set; Based on the device collection data sequence set, data anomaly active reasoning analysis is performed, and through multiple rounds of information interaction completion, a target smart playground equipment data feedback coefficient set is determined; Adjacent state iterative memory interaction is performed on the device running state index sequence set to determine a device running state feedback coefficient set; With the callable resources of the central Internet platform as a constraint, the preset monitoring resource allocation scheme is adaptively corrected according to the target smart playground equipment data feedback coefficient set and the device running state feedback coefficient set to determine an optimized monitoring resource allocation scheme, and the target smart playground equipment set is monitored in real time based on the optimized monitoring resource allocation scheme.

2. The IoT-driven smart playground equipment state real-time monitoring method of claim 1, wherein, The preset index at least includes device accuracy, response speed, signal strength and signal quality.

3. The IoT-driven smart playground equipment state real-time monitoring method of claim 1, wherein, Based on the device collection data sequence set, data anomaly active reasoning analysis is performed, and through multiple rounds of information interaction completion, a target smart playground equipment data feedback coefficient set is determined, comprising: Iterate through the device collection data sequence set to identify abnormal data and determine an abnormal device collection data sequence set; According to the preset reasoning association constraint, iterate through the abnormal device collection data sequence set to perform reasoning association data collection on the device collection data sequence set to obtain an abnormal device collection data reasoning association neighborhood sequence set; Based on the abnormal device collection data sequence set and the abnormal device collection data reasoning association neighborhood sequence set, mapping data anomaly active reasoning is performed to obtain an active reasoning problem sequence set and an active reasoning result sequence set; According to the active reasoning problem sequence set and the active reasoning result sequence set, multiple rounds of information interaction completion are performed to determine the target smart playground equipment data feedback coefficient set.

4. The IoT-driven smart playground equipment state real-time monitoring method of claim 3, wherein, Based on the abnormal device collection data sequence set and the abnormal device collection data reasoning association neighborhood sequence set, mapping data anomaly active reasoning is performed to obtain an active reasoning problem sequence set and an active reasoning result sequence set, comprising: An active reasoning analyzer is used to identify the abnormal type of the abnormal device collection data sequence set to generate an active reasoning problem sequence set; The abnormal device collection data reasoning association neighborhood sequence set is used as reasoning analysis data, and an active reasoning problem solver is used to perform active reasoning on the corresponding abnormal device collection data reasoning association neighborhood in the active reasoning problem sequence set and the reasoning analysis data to obtain the active reasoning result sequence set.

5. The IoT-driven smart playground equipment status real-time monitoring method of claim 4, wherein, According to the active reasoning question sequence set and the active reasoning result sequence set, a plurality of rounds of information interaction completion is performed to determine the target smart playground equipment data feedback coefficient set, comprising: extracting a first active reasoning question sequence and a first active reasoning result sequence in the active reasoning question sequence set and the active reasoning result sequence set; mapping interaction record log generation is performed on the first active reasoning question sequence and the first active reasoning result sequence to obtain a first active reasoning mapping interaction log sequence; first round information interaction completion is performed on the first active reasoning mapping interaction log sequence to obtain a first round information interaction vector; information interaction completion is performed on a second active reasoning mapping interaction log in the first active reasoning mapping interaction log sequence by using the first round information interaction vector to generate a second round information interaction vector, and the same is repeated to obtain a plurality of rounds of information interaction vectors; equipment state feedback analysis is performed on the plurality of rounds of information interaction vectors to determine a first smart playground equipment data feedback coefficient, and the first smart playground equipment data feedback coefficient is added to the target smart playground equipment data feedback coefficient set.

6. The IoT-driven smart playground equipment status real-time monitoring method of claim 5, wherein, The first round information interaction completion is performed on the first active reasoning mapping interaction log sequence to obtain a first round information interaction vector, comprising: extracting a first active reasoning mapping interaction log in the first active reasoning mapping interaction log sequence, and performing first-order keyword recognition on the first active reasoning mapping interaction log according to a preset equipment state keyword to obtain a first-order keyword recognition result; second-order derivative supplement is performed on the first-order keyword recognition result according to a preset near neighbor similarity threshold to obtain a first-order derivative keyword supplement result; the first-order keyword recognition result, the first-order derivative keyword supplement result, and the preset equipment state keyword are added to an initially empty vector to generate a first round information interaction vector.

7. The IoT-driven smart playground equipment status real-time monitoring method of claim 1, wherein, Adjacent state iteration memory interaction is performed on the equipment running state indicator sequence set to determine an equipment running state feedback coefficient set, comprising: extracting a first equipment running state indicator and a second equipment running state indicator of each equipment running state indicator sequence in the equipment running state indicator sequence set to obtain a first equipment running state indicator set and a corresponding second equipment running state indicator set, performing similarity recognition on the first equipment running state indicator set and the corresponding second equipment running state indicator set, and constructing an adjacent state matrix set according to the similarity recognition result set to obtain a first adjacent state matrix set; memory interaction is performed on the second equipment running state indicator set by using the first adjacent state matrix set to obtain a second interaction equipment running state indicator set; based on the second interaction equipment running state indicator set, adjacent state iteration memory interaction is performed on a third equipment running state indicator set in the equipment running state indicator sequence set, and the same is repeated until an iteration memory interaction equipment running state indicator set is obtained; based on the iteration memory interaction equipment running state indicator set, equipment running state analysis is performed to determine an equipment running state feedback coefficient set.

8. The Internet of Things driven smart playground equipment state real-time monitoring system is characterized in that, The system is used for implementing the Internet of Things driven intelligent playground equipment state real-time monitoring method of any one of claims 1-7, and the system comprises: A device collected data sequence set obtaining module is configured to access a target intelligent playground equipment set to an Internet gateway through a wireless communication module, and transmit collected data of the target intelligent playground equipment set to a central Internet platform through the wireless communication module to obtain a device collected data sequence set; An operating state analysis module is configured to call a cloud analysis engine of the central Internet platform, perform operating state analysis on the target intelligent playground equipment set according to a preset monitoring resource allocation scheme and a preset index, and obtain a device operating state index sequence set; A data feedback coefficient set determining module is configured to perform data anomaly active reasoning analysis based on the device collected data sequence set, and determine a target intelligent playground equipment data feedback coefficient set through multiple rounds of information interaction completion; A device operating state feedback coefficient set determining module is configured to perform adjacent state iteration memory interaction on the device operating state index sequence set to determine a device operating state feedback coefficient set; A device state real-time monitoring module is configured to take the callable resources of the central Internet platform as a constraint, perform device adaptive correction on the preset monitoring resource allocation scheme according to the target intelligent playground equipment data feedback coefficient set and the device operating state feedback coefficient set, determine an optimized monitoring resource allocation scheme, and perform device state real-time monitoring on the target intelligent playground equipment set based on the optimized monitoring resource allocation scheme.

Images