Internet of Things edge cloud all-in-one machine and control method thereof

Abstract

The invention relates to the technical field of Internet of Things information processing, in particular to an Internet of Things edge cloud all-in-one machine and a control method thereof. The Internet of Things edge cloud all-in-one machine comprises a signal label acquisition module used for carrying out feature identification on signal information, acquiring signal labels, and carrying out statistics on the signal labels in a preset time period to generate a signal label data set; the calculation and analysis module is used for obtaining a to-be-started instruction set through analysis according to the signal label data set. And the sorting analysis module is used for analyzing the to-be-started instruction set, obtaining a starting instruction and running the corresponding Internet of Things equipment according to the starting instruction. And the historical data storage module is used for storing the historical data. According to the invention, each piece of Internet of Things equipment in the Internet of Things area can be organically coordinated to work, so that the self-intelligent coordination and cooperation capability of the Internet of Things equipment can be realized, and the working efficiency and the user experience of the Internet of Things equipment are improved.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) information processing technology, specifically to an IoT edge cloud integrated machine and its control method. Background Technology

[0002] The traditional internet primarily connects people with devices like computers and mobile phones for information exchange. The Internet of Things (IoT), on the other hand, connects various physical objects (such as home appliances, cars, machines, and crops) to the network, enabling them to communicate with each other, be remotely controlled, or send collected data to the cloud. A simple analogy: if the internet is likened to a "network of the human brain and nervous system," responsible for thinking and transmitting information, then the IoT is like a "network of the entire planet's sensory and motor nervous system," responsible for sensing the physical world (temperature, humidity, location, state, etc.) and executing specific actions. The key components of the IoT are: the perception layer, the network layer, the platform layer, and the application layer. The perception layer includes various sensors (such as temperature sensors, GPS, and cameras) and actuators. They are responsible for collecting data and executing commands. The network layer is responsible for data transmission, including communication technologies such as Wi-Fi, Bluetooth, 5G, and NB-IoT. The platform layer is typically a cloud platform, responsible for storing, processing, and analyzing massive amounts of device data. The application layer transforms the processed data into valuable services, such as mobile app notifications, smart home automation, and industrial predictive maintenance.

[0003] In traditional IoT architectures, all data is unconditionally uploaded to the cloud for processing. However, this leads to several problems: high latency, high bandwidth pressure, data security and privacy concerns, and network dependency. Edge computing was developed to address these issues. Its core idea is to "decentralize computing power to the vicinity of the data source (the 'edge') for processing, rather than transmitting everything to the distant cloud." The IoT edge cloud appliance is a key product for implementing the IoT concept in complex industrial and enterprise scenarios. By bringing cloud computing capabilities to the edge, it solves the core pain points of real-time performance, bandwidth, security, and reliability.

[0004] Existing edge cloud integrated machines start IoT devices through user commands or sensors, which cannot perform intelligent linkage management of various IoT devices within the IoT area, resulting in low operating efficiency of IoT devices. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides an integrated IoT edge cloud appliance for intelligently selecting and managing the interconnectedness of various IoT devices within an IoT area, thereby improving the utilization efficiency of IoT devices. The integrated IoT edge cloud appliance includes: The signal tag acquisition module is used to identify the features of signal information and acquire signal tags, and to generate a signal tag dataset by statistically analyzing the signal tags within a preset time period.

[0006] The calculation and analysis module is used to analyze the signal tag dataset to obtain the set of instructions to be started.

[0007] The sorting and analysis module is used to analyze the set of startup instructions to obtain startup instructions, and run the corresponding IoT device according to the startup instructions.

[0008] The historical data storage module is used to store historical data.

[0009] Preferred: The preferred signal information includes: equipment status data, process parameter data, performance data, business data and / or feature identification data.

[0010] Preferred: Device status data includes: IoT device power on / off, running, standby, fault status, etc.

[0011] Preferred process parameters include: temperature, pressure, rotational speed, flow rate, volume, height, etc.

[0012] Preferred performance data includes: vibration, noise, current, voltage, etc.

[0013] Preferred data includes: production counts, energy consumption, etc.

[0014] Preferred method for obtaining signal tag dataset includes: obtaining signal information and the signal type to which the signal information belongs; then determining whether the signal information belongs to the signal tag database of the signal type based on the signal type; if yes, then labeling the signal information as a signal tag; if no, then the signal information does not belong to the signal tag; and collecting the labeled signal tags to form a signal tag dataset.

[0015] Preferred: The pre-set signal tag-run instruction information table is generated by constructing historical data.

[0016] Preferably, the method for obtaining the set of instructions to be started includes: numbering the signal tags in the signal tag dataset as i; obtaining the signal tags at the current time point from the signal tag dataset; then, based on the signal tags running at the current time point, pairing other signal tags to obtain an n-order tag pairing subset, 1≤n≤I-1, where I is the total number of signal tags in the signal tag dataset, i=1,2,…,I, and I>1; after pairing, performing timing optimization on the n-order tag pairing subset; and then searching the signal tag-running instruction information table to obtain the signal tag combination paired with the n-order tag pairing subset as the instructions to be started. The instructions to be started are then gathered together to form the set of instructions to be started.

[0017] Preferred method for obtaining n-order label pairing subset recipes includes: performing n-order pairing from the sorted i-th signal label based on the current time point, and after the pairing is completed; pairing the sorted i+1-th signal label until i=I-1, and then performing n+1-order pairing until n=I-1 and then ending.

[0018] Preferred: The signal tag-run instruction information table is constructed by combining a pre-set information matching table with historical data.

[0019] Preferred method for selecting the start instruction includes: calculating the pairing index p between each tag pairing subset and the corresponding start instruction in the set of start instructions, and then selecting the start instruction with the largest pairing index p as the start instruction.

[0020] Preferably: the pairing index Where T is the time length of historical data from the current time point, T' is the standard time length, n is the order of the label pairing subset, and ε is the probability of obtaining the label pairing subset in the historical data.

[0021] An IoT edge cloud integrated machine control method is provided for intelligent selection and coordinated management of various IoT devices within an IoT area, thereby improving the utilization efficiency of IoT devices. The IoT edge cloud integrated machine control method may include the following steps: S1. Perform feature recognition on the signal information and obtain signal tags, and statistically generate a signal tag dataset by collecting the signal tags within a preset time period.

[0022] S2. Analyze the signal tag dataset to obtain the set of instructions to be started.

[0023] S3. Determine if the number of startup instructions in the startup instruction set is 0; if yes, do not send a startup instruction; if no, execute S4.

[0024] S4. Determine if the number of instructions to be started in the instruction set is 1; if yes, mark the instruction to be started as the start instruction; if no, execute S5.

[0025] S5. Analyze one of the startup instructions in the startup instruction set to obtain the startup instruction, and run the corresponding IoT device according to the startup instruction.

[0026] The technical effects and advantages of this invention are as follows: The IoT edge cloud all-in-one machine can analyze the signal information to obtain the start command to be executed, and then drive the IoT devices in a modular way. This is not limited to the operation of individual IoT devices. The IoT devices are managed organically in a modular way, which can be biased towards user habits, so that the various IoT devices in the IoT area can work in an organic and coordinated manner. This enables the IoT devices to intelligently select their own direction and operate, which greatly improves work efficiency and user experience. Attached Figure Description

[0027] Figure 1 This is a structural block diagram of an IoT edge cloud integrated machine proposed in this invention.

[0028] Figure 2 This is a flowchart of a method for obtaining signal tag datasets in an IoT edge cloud all-in-one machine proposed in this invention.

[0029] Figure 3 This is a flowchart of a method for obtaining a signal start-up instruction set in an IoT edge cloud all-in-one machine proposed in this invention.

[0030] Figure 4 This is a flowchart of an IoT edge cloud integrated machine control method proposed in this invention. Detailed Implementation

[0031] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the invention, and should not be construed as limiting the invention. Rather, embodiments of the invention include all variations, modifications, and equivalents falling within the spirit and scope of the appended claims.

[0032] Example 1 refer to Figure 1 This embodiment proposes an IoT edge cloud integrated machine for intelligently selecting and managing the interconnectedness of various IoT devices within an IoT area, thereby improving the utilization efficiency of IoT devices. The IoT edge cloud integrated machine may include: The signal tag acquisition module is used to identify signal information by its features and acquire signal tags, and to statistically generate a signal tag dataset by collecting signal tags within a preset time period. The IoT edge cloud all-in-one machine needs to connect to various IoT devices within the IoT area, requiring it to have electronic device access and protocol adaptation capabilities: it connects to various types of IoT devices through a multi-protocol adaptive gateway module. These IoT devices can be production and service equipment and sensors, including PLCs, sensors, robots, etc. The IoT edge cloud all-in-one machine automatically identifies the IoT device type and communication protocol, and performs protocol conversion and data standardization. Each IoT device within the IoT area can be automatically identified by the IoT edge cloud all-in-one machine, and then manually reviewed and bound. Alternatively, the IoT edge cloud all-in-one machine can automatically identify and bind each IoT device entering the IoT area, depending on the security level of the IoT area. The IoT area can be a fixed area, such as a factory area, office floors, etc., or it can be a fixed brand of equipment, communication protocol, or type of robot, depending on actual usage needs. The signal information refers to data containing information content obtained or sent by various IoT devices. IoT devices include LCs, sensors, CNC machine tools, cameras, speakers, or recorders, etc., and can be any IoT device within the IoT area. For example, it could be an IoT refrigerator or washing machine. The signal information here can include device status data, process parameter data, performance data, business data, and / or feature identification data. Device status data records the status of IoT devices, allowing us to obtain their operating status. This can include: device power on / off, running, standby, fault status, etc.; for example, the closing signal of an electronic door lock, or the signal of a washing machine in spin-dry mode. Process parameter data reflects the environmental state, including: temperature, pressure, rotational speed, flow rate, volume, height, etc. For example, a temperature sensor detects the ambient temperature. Performance data reflects the operational performance of IoT devices, including: vibration, noise, current, voltage, etc. For example, a noise sensor detects the decibel level of friction noise from a high-speed motor. Business data records the characteristic data of the IoT device's operating phase, including: production counts, energy consumption, etc. For example, the quantity of products detected by a counter during production. Feature recognition data refers to data on the specific characteristics of objects identified by IoT devices. Examples include characteristic photographic images captured by cameras and timbre identified by sound collectors. Further details will not be elaborated upon here. Not all signal information is signal tagging; signal tags here refer to pre-labeled signal information.For example, facial features captured by a camera, voice recognition of a specific person, or temperature sensors detecting a specific temperature—these are the signal information we define as signal tags. (Reference) Figure 2 The specific real-time process involves acquiring signal information, determining its signal type, and then checking if it belongs to the corresponding signal tag database. If it does, the signal is tagged; otherwise, it is not. Collecting these tagged signals creates a signal tag dataset. For example, if a temperature sensor detects a temperature of 30°C in a production workshop, and the signal type is temperature, we can directly search the temperature-related signal tag database. This eliminates the need to compare all signal types, significantly reducing comparison time. We can then directly compare the temperature type with a pre-built temperature signal information set, such as one containing "{temperature 30 degrees - air conditioning started...}", thus tagged 30 degrees as the signal tag. This is just a simple example; actual operation requires code recognition, which will not be elaborated here. The preset time period can be set according to specific needs. For example, for the production department, the timeframe could be 1 day, and for a single signal tag, it could be 1 second. We collect the signal tags within this timeframe, classify and sort them to generate a signal tag dataset. The classification can be based on IoT devices, and the sorting can be based on time, for example: {Camera, 12:00, image recognition information; Counter, 12:02, count of 1000; ...}. Of course, this is just a simple example, and other statistical settings are not excluded; other cases will not be elaborated upon here.

[0033] The computational analysis module is used to analyze the signal tag dataset to obtain the set of instructions to be started. For an electronic device, its control system can analyze the received signals and then send instruction signals to the corresponding working units of the electronic device. These working units belong to the electronic device, and their operating programs are pre-set. However, for an IoT edge cloud integrated machine, the challenge is greater because its management scope is no longer limited to the electronic device itself, but rather it needs to receive signals from all IoT devices within the entire IoT area. The object of management is all IoT devices, thus breaking free from the individual constraints of individual devices and requiring global control of the entire IoT area, which is not simply expanding the management scope. Within an electronic device, the relationships between various operating instructions are fixed and pre-set. However, the operation of IoT devices varies greatly. It may differ from user preferences and operating environments, and the operating relationships between various operating instructions may also differ, resulting in many uncertainties. This brings great difficulties to the control of the IoT edge cloud integrated machine. There are various methods to obtain the set of instructions to be started. One approach is to search a pre-defined signal tag-run instruction information table based on the signal tags at the current time point in the signal tag dataset. This table will then identify the run instructions most frequently associated with the current tag. These run instructions can then be used as the set of instructions to be started. The associated run instructions are the run instructions that need to be started after the current signal tag has run. These run instructions are not limited to IoT devices, nor are they the sum of all the run programs of a single IoT device. They can be divided according to function. For example, for a washing machine, automatic spin-drying is one run instruction, and rinsing-spin-drying is another. This is just a simple example. This method allows for the modularization of various IoT functions, facilitating intelligent selection. When there are many IoT devices and run instructions are frequently executed, more than one run instruction will be obtained. These will be aggregated to form the set of instructions to be started. The pre-defined signal tag-run instruction information table can be generated from historical data. It can be constructed based on the frequency of the sequence of run instructions and the signal tag within a preset time period, typically one day or one week. Alternatively, it can be directly based on the last associated run instruction of the current signal tag. However, this method analyzes each signal tag individually, and the analysis is fragmented by the signals, failing to provide a comprehensive overview. This leads to inaccurate and error-prone operation of IoT devices in the IoT edge cloud integrated machine, making it unable to connect with the current operating environment. Furthermore, it collects a large amount of redundant operating command signals, significantly increasing computational complexity for later calculations. (Reference) Figure 3The method for obtaining the set of instructions to be started may further include: numbering the signal tags in the signal tag dataset as i, obtaining the signal tags at the current time point from the signal tag dataset, and then pairing other signal tags based on the signal tags running at the current time point to obtain an n-order tag pairing subset, 1≤n≤I-1, where I is the total number of signal tags in the signal tag set, i=1,2,…,I, and I>1. I>1 means that there must be at least two signal tags in the signal tag set so that they can be paired with other signal tags based on the signal tags at the current time point. If there is only one signal tag, then pairing is not necessary, and a table can be directly looked up. The tag combination in the pairing subset here can be two or I-1. Under normal circumstances, there are not too many signal tags in the signal tag set, so the computational load of pairing is not too large. Of course, this does not exclude the use of complex IoT edge cloud integrated machines. The specific pairing process can be based on the current time point, starting with the i-th ranked signal tag and performing n-order pairings. For example, obtaining {Camera, 12:00, Image Recognition Information --- Temperature Sensor, 12:10, 30℃}, after pairing, pairing is performed on the (i+1)-th ranked signal tag until i=i-1, then n+1-order pairings are performed until n=i-1. After pairing, timing optimization is performed on the n-order tag pairing subset, and then the signal tag-running instruction information table is consulted to obtain the signal tag combination paired with the n-order tag pairing subset as the start-up instruction. The start-up instructions are then gathered together to form the start-up instruction set. The timing optimization here converts absolute time to relative time. For example, the timing of {Camera, 12:00, Image Recognition Information --- Temperature Sensor, 12:10, 30℃} is optimized to {Camera, Image Recognition Information --- 10min --- Temperature Sensor, 30℃}. Of course, this is just a simple example and may not be universally applicable. Matching two signal tag combinations involves comparing whether their contents are identical, a technique already in development and not detailed here. Alternatively, the signal tag-operation command information table can be a pre-set matching table, requiring manual construction. For example, {temperature sensor 30℃; infrared sensor area has people -- start air conditioner}. This is just a simple example. Such a manually set matching table facilitates the initial operation of the IoT edge cloud all-in-one machine. Data self-optimization and self-learning can be achieved through comparison with historical data. This can be built using the edge intelligence (AI inference) of the IoT edge cloud all-in-one machine. Edge intelligence (AI inference) requires a built-in AI chip in the IoT edge cloud all-in-one machine to run AI algorithms locally; this function is existing technology and will not be detailed here. A better approach is to construct the signal tag-operation command information table by combining a pre-set matching table with historical data. Using the pre-set matching table as a foundation, historical data is used for optimization and self-learning; details will not be elaborated here.The set of instructions to be started obtained by this method can take into account the combination of signal tags in both the longitudinal and lateral directions of time. The instructions to be started are obtained based on this combination of signal tags, which breaks away from the limitation of matching a single signal tag and truly realizes the organic pairing of IoT device operation instructions, which facilitates multi-group joint operation.

[0034] The sorting and analysis module analyzes the set of commands to be started to obtain the commands and then runs the corresponding IoT devices based on these commands. There are three scenarios for the commands to be started in the set. The first scenario is that there are no commands to be started. In this case, it is not necessary to start the IoT devices within the IoT area. If a user needs to start a particular IoT device, they can do so manually. The IoT edge cloud integrated machine can store this usage record in historical data for self-learning. If the user does not use any IoT devices subsequently, it means that the user does not need any further IoT device commands. The second scenario is that there is only one command to be started. This is the simplest; the command can be started directly within a preset time period. The preset time period can be manually set or the relative time difference between the signal tag executed at the current time point and the subsequent start command execution in historical data. Details will not be elaborated here. The third scenario involves multiple commands to be executed within the set of commands to be started. In this case, selection is necessary, and various methods exist. One approach is to directly choose the command corresponding to the most recent tag pair subset. This time can be the average time of the signal tags within the tag pair subset (excluding the current time point). However, this method relies on historical data being easily overwritten and distorted by new data. Another approach is to select the command corresponding to the highest-order tag pair subset within the signal tag set. This method selects the most suitable signal tag match with a high degree of matching, but it is susceptible to inaccuracies due to changes in the environment or preferences. The selection method for the command to be started can also include calculating the pairing index p between each tag pair subset and the corresponding command in the set of commands to be started. The method for calculating the pairing index can be... Where T represents the time elapsed between the historical data and the current time point, and can be taken as the time elapsed between the last signal tag in the historical data and the current time point. T' is the standard time length, the value of which can be determined according to the specific calculation situation, and needs to consider the frequency of occurrence of the current tag pairing subset. Its value is generally between 1 minute and 1 day, but other time lengths are not excluded, which will not be elaborated here. n is the order of the tag pairing subset. ε is the probability of obtaining the tag pairing subset in the historical data, which can be based on the ratio of the number of times the tag pairing subset is associated with the pending command to the total number of occurrences in the historical data, which will not be elaborated here. Then, the pending command with the largest pairing index p is selected as the starting command. The calculation using this method not only considers the most suitable signal tag, but also analyzes factors that change over time, without being distorted by the exclusion of time factors. By analyzing historical data, tag pairing subsets with low probabilities are quickly eliminated, making the selection of the starting command more accurate. The IoT edge cloud all-in-one machine sends the starting command to the corresponding IoT device, and makes it run the starting command to perform the corresponding operation. This method enables various IoT devices within an IoT area to work in a coordinated manner, thereby achieving autonomous intelligent operation and improving work efficiency and user experience. For IoT areas where operating habits and preferences change frequently, the startup command can be played or displayed before execution for user approval. If a startup command does not meet user expectations, the user can directly close or adjust it. Voice recognition or gestures can be used, and the IoT edge cloud integrated machine records and stores these actions for analysis and learning. Specific details will not be elaborated upon here.

[0035] The historical data storage module is used to store data from IoT devices. This facilitates the analysis, extraction, and optimization of historical data by the IoT edge cloud appliance. The specifics are based on existing technology and will not be elaborated upon here.

[0036] Example 2 refer to Figure 4 An IoT edge cloud integrated machine control method is disclosed, which is used to intelligently select and manage the linkage of various IoT devices within an IoT area, thereby improving the utilization efficiency of IoT devices. The IoT edge cloud integrated machine control method may include the following steps: S1. Perform feature recognition on the signal information and obtain signal tags, and statistically generate a signal tag dataset by collecting the signal tags within a preset time period.

[0037] S2. Analyze the signal tag dataset to obtain the set of instructions to be started.

[0038] S3. Determine if the number of startup instructions in the startup instruction set is 0; if yes, do not send a startup instruction; if no, execute S4.

[0039] S4. Determine if the number of instructions to be started in the instruction set is 1; if yes, mark the instruction to be started as the start instruction; if no, execute S5.

[0040] S5. Analyze one of the startup instructions in the startup instruction set to obtain the startup instruction, and run the corresponding IoT device according to the startup instruction.

[0041] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0042] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An integrated IoT edge cloud appliance, characterized in that, The IoT edge cloud all-in-one machine includes: The signal tag acquisition module is used to identify the features of signal information and acquire signal tags, and to generate a signal tag dataset by statistically analyzing the signal tags within a preset time period. The computational analysis module is used to analyze the signal tag dataset to obtain the set of instructions to be started. The sorting and analysis module is used to analyze the set of instructions to be started and obtain the start instructions, and then run the corresponding IoT device according to the start instructions. The historical data storage module is used to store historical data.

2. The IoT edge cloud integrated machine according to claim 1, characterized in that, Signal information includes: equipment status data, process parameter data, performance data, business data, and / or feature identification data.

3. The IoT edge cloud integrated machine according to claim 2, characterized in that, Device status data includes: IoT device power on / off, running, standby, and / or fault.

4. The IoT edge cloud integrated machine according to claim 2, characterized in that, Process parameters include: temperature, pressure, rotational speed, flow rate, volume and / or height.

5. The IoT edge cloud integrated machine according to claim 2, characterized in that, Performance data include: vibration, noise, current and / or voltage.

6. The IoT edge cloud integrated machine according to claim 2, characterized in that, Business data includes: production counts and / or energy consumption.

7. The IoT edge cloud integrated machine according to claim 1, characterized in that, The method for obtaining a signal tag dataset includes: obtaining signal information and the signal type to which the signal information belongs; then determining whether the signal information belongs to the signal tag database of that signal type based on the signal type; if it does, then the signal information is labeled as a signal tag; if not, then the signal information does not belong to a signal tag; and collecting all the labeled signal tags to form a signal tag dataset.

8. The IoT edge cloud integrated machine according to claim 1, characterized in that, The method for obtaining the set of instructions to be started includes: numbering the signal tags in the signal tag dataset as i, obtaining the signal tags at the current time point from the signal tag dataset, and then pairing other signal tags based on the signal tags running at the current time point to obtain an n-order tag pairing subset, 1≤n≤I-1, where I is the total number of signal tags in the signal tag set, i=1,2,…,I, and I>1. After pairing, timing optimization is performed on the n-order tag pairing subset. Then, the signal tag-running instruction information table is searched to obtain the signal tag combination paired with the n-order tag pairing subset as the instructions to be started. The instructions to be started are then gathered together to form the set of instructions to be started.

9. An IoT edge cloud integrated machine according to claim 8, characterized in that, The method for obtaining the n-order label pairing subset recipe includes: performing n-order pairing from the sorted i-th signal label based on the current time point, and after the pairing is completed; pairing the sorted i+1-th signal label until i=I-1, and then performing n+1-order pairing until n=I-1 and then ending.

10. A control method for an integrated IoT edge cloud machine, characterized in that, The IoT edge cloud all-in-one machine control method includes the following steps: S1. Perform feature recognition on the signal information and obtain signal tags, and statistically generate a signal tag dataset by collecting the signal tags within a preset time period; S2. Analyze the signal tag dataset to obtain the set of instructions to be started; S3. Determine if the number of startup instructions in the startup instruction set is 0; if yes, do not send a startup instruction; if no, execute S4. S4. Determine if the number of instructions to be started in the instruction set is 1; if yes, mark the instruction to be started as the start instruction; if no, execute S5. S5. Analyze one of the startup instructions in the startup instruction set to obtain the startup instruction, and run the corresponding IoT device according to the startup instruction.

Images