Internet of Things equipment management platform based on big data

Abstract

The invention discloses an Internet of Things equipment management platform based on big data, and relates to the technical field of equipment management. The platform comprises a database, an equipment analysis module, a bottleneck analysis module and an intelligent management module, the equipment analysis module and the bottleneck analysis module respectively deeply analyze the real-time demand of the equipment and the communication state of the processing node to obtain a real-time index and a bottleneck index, so that the monitoring and evaluation of the equipment state and the evaluation of the performance condition of the processing node are realized, and a guarantee is provided for the intelligent management of the Internet of Things equipment management platform; the intelligent management module comprehensively analyzes the real-time index and the bottleneck index, can timely identify whether the equipment management of the processing nodes has a delay risk, controls the Internet of Things equipment management platform to enter a level-to-level management mode, and reasonably distributes the equipment to the corresponding processing nodes, and the plurality of processing nodes carry out parallel processing. The efficiency of the Internet of Things equipment management platform can be improved, and better equipment management and service experience can be provided for users.

Description

Technical Field

[0001] The present invention relates to the technical field of device management, and particularly to an Internet of Things device management platform based on big data. Background Art

[0002] With the development of Internet of Things technology, the Internet of Things device management platform has become an important tool for enterprises to manage devices; by connecting various devices within the enterprise to a unified Internet of Things device management platform, enterprises can achieve centralized management and monitoring of diverse devices to improve work efficiency and reduce management costs;

[0003] Different devices have different requirements for quick response and real-time performance; for example, industrial automation devices have very high requirements for real-time performance and need to respond and process data in a timely manner; while other devices, such as smart home devices, have relatively lower requirements for response; currently, when there are bottlenecks in data processing (communication bottlenecks and hardware bottlenecks, etc.) in the Internet of Things device management platform, it is easy to occur that devices with high response requirements are not processed in a timely manner, resulting in problems such as delay in device status analysis and execution delay of control instructions, thus affecting the normal operation of the devices. Summary of the Invention

[0004] Based on this, it is necessary to provide an Internet of Things device management platform based on big data for the problems mentioned in the above background art.

[0005] The object of the present invention can be achieved by the following technical solutions: An Internet of Things device management platform based on big data, including: a database, a device analysis module, a bottleneck analysis module, and an intelligent management module;

[0006] The device analysis module evaluates the real-time performance of the device based on device information to obtain a real-time index;

[0007] The bottleneck analysis module comprehensively evaluates the performance of each processing node based on hardware information and communication information to obtain a bottleneck index;

[0008] The intelligent management module performs load balancing through comprehensive analysis based on the received real-time index and bottleneck index, and generates corresponding management measures for intelligent management; specifically as follows:

[0009] Allocate according to the real-time performance of the device and the bottleneck index of the processing node:

[0010] Step 1: Select the device with the largest real-time index as the target device;

[0011] Step 2: Obtain the number of devices in the processing node allocation list, denoted as S2, and the real-time index Ψ1 corresponding to each device; sum up the real-time indexes corresponding to all devices in the allocation list to obtain a demand sum value, denoted as Ψ3; use the set formula Calculations are performed to obtain the processing value Xz of the processing node, where h6 and h7 are respectively set proportionality coefficients; the target device is assigned to the processing node with the largest processing value, and the number of devices in the allocation list of this processing node increases by one.

[0012] Step 3: According to Steps 1 to 2 until all devices are allocated.

[0013] Manage the corresponding devices according to the bottleneck index of the processing node:

[0014] Compare and analyze the bottleneck index with the set bottleneck threshold. When the bottleneck index is greater than the set bottleneck threshold, the processing node enters the hierarchical management mode.

[0015] Hierarchical management mode: Retrieve each device allocated by the processing node and the corresponding real-time index, compare and analyze the real-time index with the set real-time interval. When the real-time index is greater than the maximum value in the set real-time interval, the device corresponding to the real-time index is recorded as a high-priority device; when the real-time index is within the set real-time interval, the device corresponding to the real-time index is recorded as a medium-priority device; when the real-time index is less than the minimum value in the set real-time interval, the device corresponding to the real-time index is recorded as a low-priority device; data transmission, data analysis, and control execution are carried out in sequence according to the priority level of the devices.

[0016] In some embodiments, the device analysis module evaluates the real-time performance of the device based on device information to obtain a real-time index, and the real-time performance evaluation is as follows:

[0017] 201: Deeply analyze the connection relationship of the device in the enterprise network topology structure to obtain a connection coefficient;

[0018] 202: Mark the temperatures of each part of the device on the 3D model of the device according to the corresponding positions to obtain a 3D temperature model of the device, and perform quantitative analysis based on this to obtain a model temperature distribution value; use the acquisition time as the abscissa, and the power, vibration value, and model temperature distribution value as the ordinates to obtain a graph of the relationship between the power, vibration value, and model temperature distribution value changing with time, and perform analysis of the relationship graph based on this to obtain a state anomaly value;

[0019] 203: Retrieve the historical state anomaly values of the device and the corresponding generation times, and obtain an abnormal time trajectory based on this. Perform cumulative analysis on the abnormal time trajectory to obtain an aging coefficient;

[0020] 204: Substitute the aging coefficient Lδ, importance coefficient SH, and state anomaly value P into the set formula ψ1 = ln(h1×Lδ + h2×SH + h3×P + 1) for calculation to obtain the real-time index Ψ1 of the device, where h1, h2, and h3 are respectively set proportionality coefficients.

[0021] In some embodiments, a deep analysis is performed based on the connection relationship of the device in the enterprise network topology to obtain a connection coefficient, which is specifically as follows:

[0022] Arbitrarily select one of the devices and denote it as the target device. Denote the devices in direct connection with the target device as directly related devices, and count the number of directly related devices as S1. Denote the devices in indirect connection with the target device as indirectly related devices, and count the number of indirectly related devices as Hi, where i = 1, 2, 3... m1, and m1 takes positive integer values. m1 represents the total number of devices between the target device and other devices. When i = 1, H1 represents the indirectly related devices with one directly related device in between the target device. When i = 2, H2 represents the indirectly related devices with one directly related device and one indirectly related device in between the target device. And so on. When i = m1, Hm1 represents the indirectly related devices with one directly related device and m1 - 1 indirectly related devices in between the target device. Substitute the number S1 of directly related devices and the number Hi of indirectly related devices into the set formula for calculation to obtain the connection coefficient SH, where a0 and a i are respectively set proportionality coefficients, and a0 > a1 > a2 >... > a m1 > 0; thus, the connection coefficient of each device can be obtained.

[0023] In some embodiments, the temperatures of each part of the device are marked on the three-dimensional model of the device according to the corresponding positions to obtain the three-dimensional temperature model of the device, and a quantitative analysis is performed based on this to obtain the model temperature distribution value, which is specifically as follows:

[0024] Retrieve the temperatures of each part in the three-dimensional temperature model of the device at different acquisition times and denote them as Tjγ; where j = 1, 2, 3... m2, γ = 1, 2, 3... m3; m2 and m3 take positive integer values; m2 represents the total number of acquisition times, and m3 represents the total number of parts in the three-dimensional temperature model of the device; j is any one of the acquisition time serial numbers; γ is any one of the part serial numbers. Compare and analyze the temperatures of each part with the set temperature range to divide each part into high-temperature parts, medium-temperature parts, and low-temperature parts. Count the numbers of high-temperature parts, medium-temperature parts, and low-temperature parts in the three-dimensional temperature model of the device, and denote them as D1, D2, and D3 respectively. Substitute Tjγ, D1, D2, and D3 into the set formula for calculation to obtain the model temperature distribution value corresponding to this acquisition time, where d1 and d2 are respectively set proportionality coefficients, and is the average temperature of each part of the three-dimensional temperature model of the device at the i-th acquisition time.

[0025] In some embodiments, taking the acquisition time as the abscissa and the power, vibration value, and model temperature distribution value as the ordinates to obtain a graph showing the relationship between the power, vibration value, and model temperature distribution value changing over time, and performing graph analysis based on this to obtain the state outlier value. The graph analysis process is as follows:

[0026] Taking the acquisition time as the abscissa, and respectively taking the power, vibration value, and model temperature distribution value as the ordinates to construct a two-dimensional rectangular coordinate system. Input the power, vibration value, and model temperature distribution value into the coordinate system in the order of the corresponding acquisition times. Denote the positions of the power, vibration value, and model temperature distribution value in the coordinate system as the power point, vibration point, and distribution point respectively; use smooth curves to successively connect the power point, vibration point, and distribution point to obtain a graph showing the relationship between the power, vibration value, and model temperature distribution value changing over time; make tangents to the curves at the positions of the power point, vibration point, and distribution point to obtain the power tangent, vibration tangent, and distribution tangent respectively; calculate the slopes of the power tangent, vibration tangent, and distribution tangent respectively, and denote them as the power slope K Gj 、vibration slope K Zj and distribution slope K TCj ; use the set formula group to perform calculations to obtain the state outlier value P, where b1, b2, b3, b4, b5, b6, b7, b8, b9 are respectively set proportionality coefficients, Gj is the power corresponding to the j-th acquisition time, Zj is the vibration value corresponding to the j-th acquisition time, is the average value of the powers corresponding to different acquisition times, is the average value of the vibration values corresponding to different acquisition times, is the average value of the model temperature distribution values corresponding to different acquisition times, and BG, BZ, and BF are respectively the standard power, standard vibration value, and standard model temperature distribution value of this device.

[0027] In some embodiments, perform cumulative analysis on the abnormal time trajectory to obtain the aging coefficient, specifically as follows:

[0028] Retrieve the abnormal values of the device's historical status and their corresponding generation times, and obtain the abnormal time trajectory of the device based on this; compare and analyze the status abnormal values with the set abnormal intervals to classify the status abnormal values into highly abnormal values, moderately abnormal values, and lowly abnormal values, and record their corresponding generation times as highly abnormal times, moderately abnormal times, and lowly abnormal times; mark the highly abnormal times, moderately abnormal times, and lowly abnormal times in the abnormal time trajectory, calculate the interval durations between adjacent highly abnormal times, moderately abnormal times, and lowly abnormal times respectively, which are recorded as the first-level interval duration, second-level interval duration, and third-level interval duration, and calculate their respective means to obtain the first-level interval mean L1, second-level interval mean L2, and third-level interval mean L3; sum up the highly abnormal values, moderately abnormal values, and lowly abnormal values respectively to obtain the first-level abnormal value, second-level abnormal value, and third-level abnormal value, and record them as L4, L5, and L6 respectively; retrieve the usage duration of the device, which is recorded as L7, and substitute L1, L2, L3, L4, L5, L6, and L7 into the set formula for calculation to obtain the aging coefficient Lδ, where c1, c2, c3, c4, c5, and c6 are respectively the set proportionality coefficients.

[0029] In some embodiments, the bottleneck analysis module comprehensively evaluates the performance of each processing node based on hardware information and communication information to obtain a bottleneck index, specifically as follows:

[0030] 701: Conduct quantitative analysis based on the hardware information of each processing node to obtain a hardware status value;

[0031] 702: Refine the analysis of the packet loss rate and response duration of each device to obtain a packet loss coefficient and a response coefficient;

[0032] 703: Substitute the hardware status value Yz, response duration Q1jf, packet loss rate Q2jf, connection success rate Q3j, delay coefficient μ1, and packet loss coefficient μ2 into the set formula for calculation to obtain the communication status value Qzj corresponding to the acquisition moment, where q5, q6, and q7 are respectively the set proportionality coefficients; conduct trend and fluctuation analysis based on the communication status values at different acquisition moments to obtain the bottleneck index.

[0033] In some embodiments, conduct trend and fluctuation analysis based on the communication status values at different acquisition moments to obtain the bottleneck index, specifically as follows:

[0034] Taking the acquisition time as the abscissa and the communication status value as the ordinate, a two-dimensional rectangular coordinate system is constructed, and the communication status values are input into the coordinate system in the order of the corresponding acquisition times; the positions of the communication status values in the coordinate system are recorded as communication points, and a smooth curve is used to connect the communication points in sequence to obtain a curve graph of the communication status value changing with time; at the communication points, a tangent line of the curve is made and the tangent line expression is obtained by data fitting, and the derivative of the expression is calculated to obtain the derivative of the communication point denoted as Uj; the derivatives greater than zero are summed to obtain the monotonic increase denoted as A1, and the derivatives less than zero are summed and the absolute value is taken to obtain the monotonic decrease denoted as A2; using the set formula for calculation to obtain the bottleneck index Ψ2, where h4 and h5 are respectively set proportionality coefficients, is the average value of the derivatives at different acquisition times; When, then m5 takes an even value; when When, then m5 takes an odd value, and when When, then m5 takes a value of zero.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] 1. The device analysis module and the bottleneck analysis module respectively deeply analyze the real-time requirements of the device and the communication status of the processing node to obtain the real-time index and the bottleneck index, realize the monitoring and evaluation of the device status and the performance status evaluation of the processing node, timely identify the bottleneck, and provide guarantee for the intelligent management of the Internet of Things device management platform;

[0037] 2. The intelligent management module can timely identify whether there is a delay risk in the device management of the processing node by comprehensively analyzing the real-time index and the bottleneck index, and control the Internet of Things device management platform to enter the hierarchical management mode, so as to avoid the problem that the devices with high real-time requirements for delay are not processed in time, affecting the normal operation of the enterprise business; at the same time, the devices are reasonably allocated to the corresponding processing nodes, and multiple processing nodes process in parallel, effectively improving the device management efficiency, and can improve the efficiency, reliability and stability of the Internet of Things device management platform, and provide a better device management and service experience for users;

[0038] In summary, the device analysis module, the bottleneck analysis module and the intelligent management module form a collaborative working system, which can better monitor the device status in real time, identify potential bottlenecks, and take corresponding intelligent management measures, which helps to improve the efficiency and reliability of the Internet of Things device management platform, ensure the timely processing of devices with high response requirements, and reduce data processing delay and the impact on the normal operation of the devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] To more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0040] Figure 1 It is a block diagram of the principle of the present invention. Specific embodiments

[0041] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention with reference to the drawings. Many specific details are set forth in the following description to fully understand the present invention. However, the present invention can be implemented in many other ways different from those described herein. Those skilled in the art can make similar improvements without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0042] As Figure 1 shown, for the Internet of Things device management platform based on big data, the Internet of Things device management platform is respectively communicatively connected to devices 1, 2, 3... n to collect device information; the database collects communication information and hardware information and stores them; the platform includes: a database, a device analysis module, a bottleneck analysis module, and an intelligent management module;

[0043] The device analysis module comprehensively analyzes the real-time performance and operating status of the device based on the device information to obtain a real-time index and sends it to the intelligent management module; specifically:

[0044] Retrieve device information, where the device information includes the device's three-dimensional model, power, vibration value, temperature of each part of the device, and connection relationship with other devices; mark the temperature of each part of the device on the device's three-dimensional model according to the corresponding position to obtain the device's three-dimensional temperature model;

[0045] Arbitrarily select one of the devices and denote it as the target device. Denote the devices in a direct connection relationship with the target device as directly related devices, and count the number of directly related devices as S1. Denote the devices in an indirect connection relationship with the target device as indirectly related devices, and count the number of indirectly related devices as Hi, where i = 1, 2, 3... m1, and m1 takes positive integer values. m1 represents the total number of devices between the target device and other devices. When i = 1, H1 represents the indirectly related devices with one directly related device in between the target device. When i = 2, H2 represents the indirectly related devices with one directly related device and one indirectly related device in between the target device. And so on. When i = m1, Hm1 represents the indirectly related devices with one directly related device and m1 - 1 indirectly related devices in between the target device. Use the set formula for calculation to obtain the connection coefficient SH, where a0 and a i are respectively set proportionality coefficients, and a0 > a1 > a2 >... > a m1 > 0. Thus, the connection coefficient of each device can be obtained. It should be noted that the larger the connection coefficient, the more important the device is in the enterprise network topology structure;

[0046] Denote the temperature in the three-dimensional temperature model of the device at different acquisition times as Tjγ; where j = 1, 2, 3... m2, γ = 1, 2, 3... m3; m2 and m3 take positive integer values. m2 represents the total number of acquisition times, and m3 represents the total number of parts in the three-dimensional temperature model of the device. j is any one of the acquisition time serial numbers; γ is any one of the part serial numbers. Compare and analyze the temperature of each part with the set temperature range. When the temperature is greater than the maximum value in the set temperature range, then mark this part as a high-temperature part. When the temperature is within the set temperature range, then mark this part as a medium-temperature part. When the temperature is less than the minimum value in the set temperature range, then mark this part as a low-temperature part. Count the number of high-temperature parts, medium-temperature parts, and low-temperature parts in the three-dimensional temperature model of the device, and denote them as D1, D2, and D3 respectively. Use the set formula for calculation to obtain the model temperature distribution value corresponding to this acquisition time, where d1 and d2 are respectively set proportionality coefficients, is the average temperature of each part of the three-dimensional temperature model of the device at the i-th acquisition moment; taking the acquisition moment as the abscissa and the power, vibration value, and model temperature distribution value as the ordinates respectively, a two-dimensional rectangular coordinate system is constructed. The power, vibration value, and model temperature distribution value are input into the coordinate system in sequence according to the corresponding acquisition moments. The positions of the power, vibration value, and model temperature distribution value in the coordinate system are denoted as the power point, vibration point, and distribution point; smooth curves are used to connect the power point, vibration point, and distribution point in sequence to obtain the relationship diagrams of the power, vibration value, and model temperature distribution value changing with time; tangents to the curves are made at the positions of the power point, vibration point, and distribution point to obtain the power tangent, vibration tangent, and distribution tangent; the slopes of the power tangent, vibration tangent, and distribution tangent are calculated respectively and denoted as the power slope K Gj , vibration slope K Zj and distribution slope K TCj ; using the set formula group for calculation to obtain the power anomaly value Gp, vibration anomaly value Zp, temperature distribution anomaly value Fp, and state anomaly value P, where b1, b2, b3, b4, b5, b6, b7, b8, b9 are respectively set proportionality coefficients, Gj is the power corresponding to the j-th acquisition moment, Zj is the vibration value corresponding to the j-th acquisition moment, is the average value of the powers corresponding to different acquisition moments, is the average value of the vibration values corresponding to different acquisition moments, is the average value of the model temperature distribution values corresponding to different acquisition moments, BG, BZ, and BF are respectively the standard power, standard vibration value, and standard model temperature distribution value of the device; it should be noted that the larger the state anomaly value of the device, the greater the risk that the device is in an abnormal operating state;

[0047] Retrieve the abnormal values of the device's historical status and their corresponding generation times, and obtain the abnormal time trajectory of the device based on this; compare and analyze the status abnormal values with the set abnormal range. When the status abnormal value is greater than the maximum value in the set abnormal range, record this status abnormal value as a highly abnormal value and its corresponding generation time as a highly abnormal time; when the status abnormal value is within the set abnormal range, record this status abnormal value as a moderately abnormal value and its corresponding generation time as a moderately abnormal time; when the status abnormal value is less than the minimum value in the set abnormal range, record this status abnormal value as a lowly abnormal value and its corresponding generation time as a lowly abnormal time; mark the highly abnormal time, moderately abnormal time, and lowly abnormal time in the abnormal time trajectory, calculate the interval durations between adjacent highly abnormal times, moderately abnormal times, and lowly abnormal times respectively, which are recorded as the first-level interval duration, second-level interval duration, and third-level interval duration, and calculate their respective means to obtain the first-level interval mean L1, second-level interval mean L2, and third-level interval mean L3; sum up the highly abnormal values, moderately abnormal values, and lowly abnormal values respectively to obtain the first-level abnormal value, second-level abnormal value, and third-level abnormal value, and record them as L4, L5, and L6 respectively; retrieve the usage duration of the device, which is recorded as L7, and use the set formula for calculation to obtain the aging coefficient Lδ, where c1, c2, c3, c4, c5, c6 are respectively set proportionality coefficients;

[0048] Calculate the real-time index Ψ1 of the device by using the set formula ψ1 = ln(h1×Lδ + h2×SH + h3×P + 1) with the aging coefficient Lδ, importance coefficient SH, and status abnormal value P, where h1, h2, h3 are respectively set proportionality coefficients; it should be noted that in the network topology structure of the device enterprise, the more important it is and the greater the risk of abnormal operation status, it means that a higher real-time response requirement is needed for the device to ensure its normal operation;

[0049] Obtain the real-time index of the device through comprehensive analysis of the importance of the device in the enterprise network topology structure and the operating status of the device, so as to measure the device's demand for real-time response and provide technical support for the intelligent management of the device.

[0050] The bottleneck analysis module analyzes the performance status of each processing node of the platform based on communication information and hardware information to obtain a bottleneck index and send it to the intelligent management module; specifically:

[0051] The processing node refers to the server in the Internet of Things device management platform that is used to process data and execute computing tasks. These processing nodes can be physical servers or entities in virtualized environments such as virtual machines or containers;

[0052] Hardware status analysis:

[0053] Arbitrarily select a processing node and retrieve the hardware information, where the hardware information includes server type, storage memory, and running memory; set that each server type corresponds to a type value, compare and match the server type with all the set server types to obtain the corresponding type value, and record it as Y1; calculate the hardware status value Yz through the set formula Yz = y1×Y1 + y2×Y2 + y3×Y3 using the storage memory Y2, running memory Y3, and type value Y1, where y1, y2, and y3 are respectively the set proportionality coefficients;

[0054] Communication status analysis:

[0055] Arbitrarily select a processing node and retrieve the communication information at different collection times, where the communication information includes the response duration of each device, the packet loss rate of each device, and the connection success rate, and record them as Q1jf, Q2jf, and Q3j respectively, where f = 1, 2, 3... m4, m4 takes positive integer values, and m4 represents the total number of devices; f is the serial number of any one of the devices; the time from when the server receives a request from the device to when it returns to the device is recorded as one response, and the duration of receiving and returning is the response duration of this device; the connection success rate is the ratio between the number of devices in the connected state at a certain moment and the total number of devices requesting to connect to the server;

[0056] Compare and analyze the response duration with the set response interval. When the response duration is greater than the maximum value in the set response interval, it indicates that there is a relatively serious delay in the response between the device and the server, and then mark this device as a first-level delay device; when the response duration is within the set response interval, then mark this device as a second-level delay device; when the response duration is less than the minimum value in the set response interval, then mark this device as a third-level delay device; respectively count the numbers of first-level delay devices, second-level delay devices, and third-level delay devices, and record them as W1, W2, and W3 respectively; compare and analyze the packet loss rate with the set packet loss interval. When the packet loss rate is greater than the maximum value in the set packet loss interval, then mark this device as a serious packet loss device; when the packet loss rate is within the set packet loss interval, then mark this device as a moderate packet loss device, and when the packet loss rate is less than the minimum value in the set packet loss interval, then mark this device as a minor packet loss device; respectively count the serious packet loss devices, moderate packet loss devices, and minor packet loss devices, and record them as W4, W5, and W6 respectively; use the set formula group to calculate to obtain the delay coefficient μ1 and the packet loss coefficient μ2, where q1, q2, q3, q4 are respectively the set proportionality coefficients;

[0057] Pass the hardware status value Yz, response duration Q1jf, packet loss rate Q2jf, connection success rate Q3j, delay coefficient μ1, and packet loss coefficient μ2 through the set formula Calculations are performed to obtain the communication status value Qzj corresponding to the acquisition moment, where q5, q6, and q7 are respectively set proportionality coefficients; taking the acquisition moment as the abscissa and the communication status value as the ordinate to construct a two-dimensional rectangular coordinate system, and inputting the communication status values into the coordinate system successively according to the corresponding acquisition moments; recording the positions of the communication status values in the coordinate system as communication points, and connecting the communication points successively with a smooth curve to obtain a curve graph of the communication status value changing with time; making a tangent to the curve at the communication point and using data fitting to obtain the tangent expression, and performing a derivative operation on the expression to obtain the derivative of this communication point denoted as Uj; summing the derivatives greater than zero to obtain the monotonic increase degree denoted as A1, and summing the derivatives less than zero and taking the absolute value to obtain the monotonic decrease degree denoted as A2; using the set formula Calculations are performed to obtain the bottleneck index Ψ2, where h4 and h5 are respectively set proportionality coefficients, is the average value of the derivatives at different acquisition moments; when holds, then m5 takes an even value; when holds, then m5 takes an odd value, and when holds, then m5 takes a value of zero; and so on to obtain the bottleneck index of each processing node, and sending the generated bottleneck index to the intelligent management module;

[0058] By comprehensively analyzing the hardware state and communication state, it is possible to comprehensively evaluate the performance status of each processing node of the Internet of Things device management platform, accurately identify the bottlenecks of the processing nodes, and help system administrators discover and solve performance bottleneck problems in a timely manner, thereby improving the stability and reliability of the system and ensuring the normal operation of the Internet of Things device management platform.

[0059] The intelligent management module performs load balancing through comprehensive analysis based on the received real-time index and bottleneck index, and generates corresponding management measures for intelligent management; specifically:

[0060] Allocation is carried out according to the real-time nature of the device and the bottleneck index of the processing node:

[0061] Step 1: Select the device with the largest real-time index as the target device;

[0062] Step 2: Obtain the number of devices in the processing node allocation list denoted as S2 and the real-time index Ψ1 corresponding to each device; sum up the real-time indexes corresponding to all devices in the allocation list to obtain the demand sum value denoted as Ψ3; use the set formula Calculations are performed to obtain the processing value Xz of the processing node, where h6 and h7 are respectively set proportionality coefficients; allocate the target device to the processing node with the largest processing value, and the number of devices in the allocation list of this processing node increases by one;

[0063] Step 3: Based on Steps 1 to 2 until all devices are allocated, perform device management allocation by analyzing the processing capabilities of processing nodes to achieve load balancing, avoid the situation where some nodes are overloaded while other nodes have idle resources, and achieve reasonable utilization of resources;

[0064] Perform corresponding device management according to the bottleneck index of the processing node:

[0065] Compare and analyze the bottleneck index with the set bottleneck threshold. When the bottleneck index is greater than the set bottleneck threshold, it indicates that there is a risk of delay in data transmission, data processing, and control execution of the processing node at this time, and the processing node enters the hierarchical management mode;

[0066] Hierarchical management mode: Retrieve each device allocated to the processing node and the corresponding real-time index, compare and analyze the real-time index with the set real-time range. When the real-time index is greater than the maximum value in the set real-time range, the device corresponding to the real-time index is recorded as a high-priority device; when the real-time index is within the set real-time range, the device corresponding to the real-time index is recorded as a medium-priority device; when the real-time index is less than the minimum value in the set real-time range, the device corresponding to the real-time index is recorded as a low-priority device; perform data transmission, data analysis, and control execution in sequence according to the priority level of the device to ensure that high-priority devices are processed and resource-allocated in a timely manner, improving the efficiency and response speed of device management;

[0067] By comprehensively analyzing the real-time index and the bottleneck index, it is possible to timely identify whether there is a risk of delay in the device management of the processing node, and control the Internet of Things device management platform to enter the hierarchical management mode, thereby avoiding the problem that devices with high real-time requirements for delay are not processed in a timely manner, affecting the normal operation of the enterprise's business; at the same time, reasonably allocate devices to the corresponding processing nodes, and multiple processing nodes can effectively improve the device management efficiency in parallel, which can improve the efficiency, reliability, and stability of the Internet of Things device management platform, and provide users with a better device management and service experience.

[0068] The technical features of the above-described embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0069] The above-described embodiments merely represent several implementation manners of the present invention, and their descriptions are relatively specific and detailed, but should not be construed as limiting the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention patent should be subject to the appended claims.

Claims

1. IoT device management platform based on big data, characterized by: include: Equipment analysis module, bottleneck analysis module and intelligent management module; The device analysis module evaluates the real-time performance of the device based on the device information to obtain a real-time index; The bottleneck analysis module comprehensively evaluates the performance of each processing node based on hardware information and communication information to obtain a bottleneck index; The intelligent management module performs a comprehensive analysis based on the received real-time index and bottleneck index to balance the load and generate corresponding management measures for intelligent management; the details are as follows: Allocation is based on the real-time performance of the device and the bottleneck index of the processing node: Step 1: Select the device with the largest real-time index as the target device; Step 2: Obtain the number of devices in the processing node allocation list and the real-time index corresponding to each device, sum the real-time indexes corresponding to all devices in the allocation list to obtain the demand and value, and perform a formulaic calculation and analysis on it and the number of devices to obtain the processing value of the processing node; assign the target device to the processing node with the largest processing value, and the number of devices in the allocation list of the processing node increases by one; Step 3: Follow steps 1 to 2 until all devices are assigned; Perform corresponding equipment management based on the bottleneck index of the processing node: Compare and analyze the bottleneck index with the set bottleneck threshold. When the bottleneck index is greater than the set bottleneck threshold, the processing node is controlled to enter the hierarchical management mode. Hierarchical management mode: retrieve the devices assigned to the processing node and the corresponding real-time index, compare and analyze the real-time index with the set real-time interval, when the real-time index is greater than the maximum value in the set real-time interval, the device corresponding to the real-time index is recorded as a high-priority device; when the real-time index is within the set real-time interval, the device corresponding to the real-time index is recorded as a medium-priority device; when the real-time index is less than the minimum value in the set real-time interval, the device corresponding to the real-time index is recorded as a low-priority device; data transmission, data analysis and control execution are carried out in sequence according to the priority level of the equipment.

2. The IoT device management platform based on big data according to claim 1 is characterized in that: The device analysis module evaluates the real-time performance of the device based on the device information to obtain a real-time index, where the real-time performance evaluation is as follows: 201: Conduct in-depth analysis based on the connection relationship of the device in the enterprise network topology to obtain the connection coefficient; 202: Mark the temperature of each part of the equipment on the three-dimensional model of the equipment according to the corresponding position to obtain the three-dimensional temperature model of the equipment, and perform quantitative analysis based on it to obtain the model temperature distribution value; use the acquisition time as the horizontal coordinate, and use the power, vibration value and model temperature distribution value as the vertical coordinate to obtain a relationship diagram of the power, vibration value and model temperature distribution value over time, and perform relationship diagram analysis based on this to obtain the state abnormal value; 203: Retrieve the abnormal value of the device historical state and the corresponding generation time, and obtain the abnormal time trajectory based on it, and perform cumulative analysis on the abnormal time trajectory to obtain the aging coefficient; 204: The aging coefficient, the important coefficient and the abnormal state value are calculated and analyzed in a formula to obtain a real-time index.

3. The IoT device management platform based on big data according to claim 2 is characterized in that: Based on the connection relationship of the equipment in the enterprise network topology, a deeper analysis is performed to obtain the connection coefficient, as follows: Take any one of the devices and record it as the target device, record the devices that are directly connected to the target device as directly related devices, and count the number of directly related devices; record the devices that are indirectly connected to the target device as indirectly related devices, and count the number of indirectly related devices; The number of directly related equipment and the number of indirectly related equipment are calculated and analyzed in a formula to obtain the connection coefficient.

4. The IoT device management platform based on big data according to claim 3 is characterized in that: The temperatures of various parts of the equipment are marked on the three-dimensional model of the equipment according to the corresponding positions to obtain the three-dimensional temperature model of the equipment, and quantitative analysis is performed based on this to obtain the model temperature distribution value, as follows: The temperature of each part in the three-dimensional temperature model of the equipment at different acquisition times is retrieved, and the temperature of each part is compared and analyzed with the set temperature range to divide each part into a high-temperature part, a medium-temperature part and a low-temperature part; the number of high-temperature parts, medium-temperature parts and low-temperature parts in the three-dimensional temperature model of the equipment is counted, and it is analyzed by formula calculation with the temperature of each part to obtain the model temperature distribution value.

5. The IoT device management platform based on big data according to claim 4 is characterized in that: The acquisition time is used as the horizontal axis, and the power, vibration value and model temperature distribution value are used as the vertical axis to obtain the relationship diagram of power, vibration value and model temperature distribution value over time, and the relationship diagram is parsed based on this to obtain the state abnormal value, wherein the relationship diagram parsing process is: A two-dimensional rectangular coordinate system is constructed with the acquisition time as the horizontal coordinate and the power, vibration value and model temperature distribution value as the vertical coordinate. The power, vibration value and model temperature distribution value are input into the coordinate system in sequence according to the corresponding acquisition time. The positions of the power, vibration value and model temperature distribution value in the coordinate system are recorded as power points, vibration points and distribution points. A smooth curve is used to connect the power points, vibration points and distribution points in sequence to obtain a time-varying relationship diagram of the power, vibration value and model temperature distribution value. Tangents of the curves are drawn at the positions of the power points, vibration points and distribution points to obtain the power tangent, vibration tangent and distribution tangent. The slopes of the power tangent, vibration tangent and distribution tangent are calculated respectively, and recorded as the power slope K. Gj , vibration slope K Zj and the distribution slope K TCj ; Using the set formula group Calculation is performed to obtain the state abnormality value P, where b1, b2, b3, b4, b5, b6, b7, b8, and b9 are the set proportional coefficients respectively, Gj is the power corresponding to the j acquisition moment, Zj is the vibration value corresponding to the j acquisition moment, G is the average value of the power corresponding to different acquisition moments, Z is the average value of the vibration values ​​corresponding to different acquisition moments, F is the average value of the model temperature distribution values ​​corresponding to different acquisition moments, and BG, BZ, and BF are the standard power, standard vibration value, and standard model temperature distribution value of the equipment respectively.

6. The IoT device management platform based on big data according to claim 5 is characterized in that: The abnormal time trajectory is cumulatively analyzed to obtain the aging coefficient, as follows: Retrieve the historical status anomaly values ​​of the equipment and the corresponding generation time, and obtain the abnormal time trajectory of the equipment accordingly; compare and analyze the status anomaly values ​​with the set abnormal interval to divide the status anomaly values ​​into high anomaly values, moderate anomaly values ​​and low anomaly values, and record their corresponding generation times as high anomaly moments, moderate anomaly moments and low anomaly moments; mark the high anomaly moments, moderate anomaly moments and low anomaly moments in the abnormal time trajectory, calculate the interval durations between two adjacent high anomaly moments, moderate anomaly moments and low anomaly moments respectively, and record them as the first-level interval duration, the second-level interval duration and the third-level interval duration, and average them respectively to obtain the first-level interval mean, the second-level interval mean and the third-level interval mean; sum the high anomaly values, moderate anomaly values ​​and low anomaly values ​​respectively to obtain the first-level anomaly values, the second-level anomaly values ​​and the third-level anomaly values, and perform formulaic calculation and analysis with the first-level interval mean, the second-level interval mean and the third-level interval mean to obtain the aging coefficient.

7. The IoT device management platform based on big data according to claim 1, characterized in that: The bottleneck analysis module comprehensively evaluates the performance of each processing node based on hardware information and communication information to obtain the bottleneck index, as follows: 701: randomly select a processing node, and perform quantitative analysis based on its hardware information to obtain a hardware status value of the processing node; 702: performing detailed analysis on the packet loss rate and response time of each device to obtain a packet loss coefficient and a response coefficient; 703: Perform formula calculation and analysis on the hardware status value, response time, packet loss rate, connection success rate, delay coefficient and packet loss coefficient to obtain the communication status value; perform trend and fluctuation analysis on the communication status values ​​at different collection times to obtain the bottleneck index.

8. The IoT device management platform based on big data according to claim 7 is characterized in that: The bottleneck index is obtained by performing trend and fluctuation analysis based on the communication status values ​​at different collection times, as follows: A two-dimensional rectangular coordinate system is constructed with the acquisition time as the horizontal coordinate and the communication state value as the vertical coordinate, and the communication state value is input into the coordinate system in sequence according to the corresponding acquisition time; the position of the communication state value in the coordinate system is recorded as the communication point, and the communication points are connected in sequence with a smooth curve to obtain a curve graph of the communication state value changing with time; a tangent line is drawn at the communication point and the tangent expression is obtained by data fitting, and the derivative of the communication point is obtained by differentiation operation, and it is recorded as Uj; the derivatives greater than zero are summed up to obtain a monotonic increase, which is recorded as A1, and the derivatives less than zero are summed up and the absolute value is taken to obtain a monotonic decrease, which is recorded as A2; using the set formula Calculate to get the bottleneck index Ψ2, where h4 and h5 are the set proportional coefficients, is the mean value of the derivative at different acquisition times; when When , m5 takes an even number; when When m5 is an odd number, When , m5 takes the value of zero.

Images