A unified cloud computing system for ubiquitous power Internet of Things
By collecting operating status data in real time and dynamically dividing computing tasks, the problem of insufficient computing power of smart meters is solved, the efficient operation of a unified cloud computing system is achieved, the data transmission cost and cloud platform load are reduced, and the computing efficiency of smart meters is enhanced. Functional adaptability.
Patent Information
- Application Number
- CN202010155507.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-03-09
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2040-03-09
AI Technical Summary
Due to insufficient computing power, existing smart meters cannot effectively utilize the advantages of cloud computing platforms and smart meters, resulting in high data transmission costs, excessive load on cloud computing platforms, and the inability to dynamically adjust functions to meet the needs of ubiquitous power grids.
A unified cloud computing system was designed to collect operating status data in real time, label and divide computing tasks, and divide them into multiple computing chips that are allocated to smart meters. Dynamic task reconstruction strategy is adopted to improve computing efficiency and reduce data migration. and energy consumption.
It has achieved the expansion of the computing power of the cloud computing platform, improved the computing efficiency of smart meters, reduced data transmission costs and cloud platform load, and enhanced the functional adaptability of smart meters.
Smart Images

Figure CN111245958B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of ubiquitous power Internet of Things, and in particular to a unified cloud computing system for the ubiquitous power Internet of Things. Background Art
[0002] The introduction of ubiquitous power grid has greatly increased the demand for smart meter functions. Traditional smart meters can perform relatively fixed and common functions, and their programs are mostly solidified after the meters leave the factory, and new functions cannot be added over time. At the same time, smart meters cannot meet the needs of subsequent functional changes due to the use of microcontrollers with poor computing power. Therefore, when the function changes, the PCB board needs to be redesigned, which increases manpower and material costs.
[0003] Therefore, the new generation of smart meters proposes a structure composed of multiple computing chips. Tasks with different computing requirements are distributed to different computing chips. At the same time, all sites are equipped with corresponding cloud computing platforms.
[0004] However, in the face of the powerful data base of the ubiquitous power grid, there is no efficient and practical way to combine the two so as to fully utilize the advantages of cloud platforms and smart meters and fully complete the methods and systems for site allocation tasks. This means that all computing tasks need to be completed on the cloud computing platform, resulting in high data transmission costs and high load on the cloud computing platform. Summary of the invention
[0005] The present application provides a unified cloud computing system for ubiquitous power Internet of Things, which fully utilizes the advantages of cloud platforms and smart meters, fully completes the task of site allocation, and solves the problem of huge data received by the cloud computing platform, increased costs, and increased load on the cloud computing platform.
[0006] The technical solutions adopted by this application to solve the above technical problems are as follows:
[0007] A unified cloud computing system for ubiquitous power Internet of Things, including:
[0008] The user platform includes at least one smart meter, the smart meter includes a communication chip, a computing chip and a storage chip, the smart meter is connected to the cloud computing platform through the communication chip, the computing chip is used to execute the computing task sent by the cloud computing platform, and return the computing result to the cloud computing platform through the communication chip, and the storage chip is used to store the operation data and computing data generated by the smart meter;
[0009] A site-based cloud computing platform, on which a monitoring unit and a server are running, wherein the monitoring unit is used to collect the operating status data of the cloud computing platform and the smart meter in real time, and to form a computing task to be issued according to the operating status data;
[0010] The cloud computing platform runs a server and a monitoring unit for monitoring the server, wherein the monitoring unit includes a first task scheduling unit, which is used to collect the operation status data of the cloud computing platform and the smart meter in real time, and to mark and divide the computing tasks to obtain the marked computing tasks after the marking and division, and to send the marked computing tasks to the second task scheduling unit;
[0011] A second task scheduling unit is run in the smart meter, and is used for cutting and dividing the marked calculation task after receiving the marked calculation task, and sending the cut and divided marked calculation task to the computing chip.
[0012] Optionally, the computing chip includes at least: a CPU chip, an FPGA chip and an ASIC chip; the CPU chip, the FPGA chip and the ASIC chip are all integrated inside the smart meter and connected to the cloud computing platform via the communication chip.
[0013] Optionally, the operating status data includes resource utilization of the cloud computing platform, busyness of the smart meter, physical topology of the smart meter and load status of the computing chip.
[0014] Optionally, the communication connection includes a wireless network, a private network or a public Internet of Things.
[0015] Optionally, an operating system runs in the smart meter to support the running of multiple applications.
[0016] Optionally, the computing task includes a computing-intensive task, and the computing-intensive task includes encryption and decryption and numerical calculation.
[0017] The technical solution provided by this application includes the following beneficial technical effects:
[0018] The present application provides a unified cloud computing system for ubiquitous power Internet of Things. It collects the operating status data of the cloud computing platform and smart meters in real time, and labels and divides the computing tasks according to the operating status data information to obtain the labeled computing tasks after labeling and division, and sends the labeled computing tasks to the second task scheduling unit in the smart meter. After receiving the labeled computing tasks, the second task scheduling unit cuts and divides the labeled computing tasks, and sends the cut and divided labeled computing tasks to the computing chip. The present application proposes a unified computing system based on the interconnection of the site-based cloud computing platform and the user's smart meter computing chip. The task assigned to the computing chip in the smart meter is divided by a task division strategy combining dynamic and static. On the one hand, it can expand the computing power of the cloud computing platform of the current site and form a computing pool with the computing power of the user's smart meter. On the other hand, it can improve the efficiency of the computing tasks inside the smart meter by dynamically reconstructing the tasks. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the embodiments are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.
[0020] Figure 1 A schematic diagram of the structure of a unified cloud computing system for ubiquitous power Internet of Things provided in an embodiment of the present application;
[0021] Figure 2 A schematic diagram of task division of a smart meter provided in an embodiment of the present application;
[0022] Figure 3 A schematic diagram of task division of a fault-tolerant group provided in an embodiment of the present application. DETAILED DESCRIPTION
[0023] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the application embodiments will be clearly and completely described below in conjunction with the drawings in the embodiments of this application; it is obvious that the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.
[0024] In order to efficiently utilize the computing resources of smart meters, this application proposes a collaborative task reconfigurable loading technology for smart meters with multiple computing chips. According to the requirements of the task, the task is divided at the site, and then the task is transmitted to the smart meter to be completed by the hardware of the smart meter in collaboration, thereby reducing the complexity of task division and improving resource utilization. At the same time, cloud computing technology is used to interconnect the smart meter and the cloud computing platform of the remote site to form a unified and efficient cloud computing platform.
[0025] In the ubiquitous power Internet of Things, smart meters carry the core tasks of the underlying foundation. In addition to recording the user's electricity usage and providing a data source for the analysis and mining of the user's electricity usage, the smart meters provided in the embodiments of the present application can also complete some localized analysis and calculation tasks. The present application designs a method for dynamically reconstructing tasks, with the aim of enabling the user's smart meter to quickly complete the task and return the result after the remote site issues the task, thereby improving the utilization rate of the computing chip and shortening the task completion time.
[0026] The embodiment of the present application provides a unified cloud computing system for ubiquitous power Internet of Things, including: a user platform, including at least one smart meter, the smart meter including a communication chip, a computing chip and a storage chip, and the smart meter is also provided with other chips required for normal use, which are not described in detail in the embodiment of the present application. The smart meter is connected to the cloud computing platform through the communication chip, the computing chip is used to execute the computing tasks sent by the cloud computing platform, and return the computing results to the cloud computing platform through the communication chip, and the storage chip is used to store the operation data and computing data generated by the smart meter;
[0027] Based on the site's cloud computing platform, the site has its own cloud computing platform. Normally, the site can independently complete big data analysis tasks. When the site's resources are tight, the site's tasks can be assigned to smart meters to achieve task linkage between the cloud computing platform and the smart meter, and use the computing resources of the smart meter to complete the task. Through the interconnection of smart meters and cloud computing platforms, the amount of data migration is reduced and energy consumption is reduced. At the same time, the task reconfiguration in the smart meter improves the operating efficiency of the task and shortens the task completion time. In the traditional mode, the smart meter will not receive the computing tasks sent by the site, but only transmits the collected data back to the site, and the site processes the data. The embodiment of the present application proposes to assign the computing tasks to the smart meter and send the calculation results back to the site instead of sending the data back to the site, thereby reducing the amount of data migration and reducing energy consumption.
[0028] The cloud computing platform runs a server and a monitoring unit for monitoring the server, wherein the monitoring unit includes a first task scheduling unit, which is used to collect the operation status data of the cloud computing platform and the smart meter in real time, and to mark and divide the computing tasks to obtain the marked computing tasks after the marking and division, and to send the marked computing tasks to the second task scheduling unit;
[0029] A second task scheduling unit is run in the smart meter, and is used for cutting and dividing the marked calculation task after receiving the marked calculation task, and sending the cut and divided marked calculation task to the computing chip.
[0030] The reason for dividing the computing tasks is mainly due to two aspects. First, when the site's cloud computing platform performs computing tasks, the data migration cost is relatively high. It is cheaper to complete the corresponding calculation at the user's smart meter than to transmit the smart meter data back to the site's cloud platform and then calculate it through the cloud platform. On the other hand, when the site's computing resources are insufficient, the computing resources of idle user smart meters can be used to form a computing resource pool to reduce the load on the cloud computing platform.
[0031] Optionally, the computing chip at least includes: a CPU chip, an FPGA chip and an ASIC chip, and the CPU chip, the FPGA chip and the ASIC chip are all integrated inside the smart meter and connected to the cloud computing platform through the communication chip. The computing tasks of the ASIC chip are solidified when leaving the factory, such as the common metering function of the meter, and the metering method generally adopts a unified and unchanged metering method; the CPU chip generally adopts ARM or domestic chips, and runs an operating system or application on it. The FPGA chip is a high-performance computing chip that can run various tasks. Unlike the CPU chip, the FPGA chip can run tasks in parallel on hardware, especially computing-intensive tasks.
[0032] After the site cloud computing platform issues the task, the smart meter dynamically divides the task by analyzing the task. Tasks generally have their own characteristics. Some tasks are mainly computationally intensive, and some tasks require frequent access to storage and reading of stored data. The dynamic task reconfiguration method proposed in the embodiment of the present application, when dividing the task, the cloud computing platform marks the task on the cloud computing platform side, relies on the powerful computing resources of the cloud computing platform, analyzes the characteristics of the task, and cuts the task at the same time, dividing the task into subtasks, that is, tasks suitable for CPU operation and tasks suitable for FPGA operation. The cloud platform marks and divides the task and packages it to the smart meter. After receiving the task, the second task scheduling unit connected to the smart meter first decomposes the task according to the mark, and divides the task into multiple small tasks according to the characteristics of the task. At the same time, the second task scheduling unit of the smart meter then starts to analyze the resource utilization of the computing chip. In the embodiment of the present application, a greedy algorithm is used to first assign subtasks to the FPGA chip as much as possible. When the task running on the FPGA chip is full, the subtask is then assigned to the CPU chip for completion.
[0033] Optionally, the operation status data includes resource utilization of the cloud computing platform, busyness of the smart meter, physical topology of the smart meter and load status of the computing chip. The resource utilization includes resource utilization of the CPU memory, network, hard disk, etc. of the physical computer of the site cloud computing platform; the cloud computing platform constructs a corresponding physical topology according to the physical location of the smart meter, thereby marking and dividing the computing tasks.
[0034] Optionally, the communication connection includes a wireless network, a private network or a public Internet of Things.
[0035] Optionally, an operating system runs in the smart meter to support the running of multiple applications.
[0036] Optionally, the computing tasks include computing-intensive tasks, which include encryption and decryption and numerical calculations. The computing-intensive tasks also include other tasks issued by the site, which are not described in detail in the embodiments of the present application and can be set according to actual usage requirements.
[0037] As an implementation mode of the present application, when it is applied, the cloud computing platform of the site issues tasks to the smart meter. First, it establishes communication with the smart meter through the communication module, obtains the current operating status of the smart meter, and understands the status including the usage of the computing chip of the smart meter, so as to prepare for the task issuance. After the site obtains the status of the smart meter, it decomposes and issues the task. The task decomposition method adopts a multi-copy method. Generally speaking, the computing power of the user's smart meter is relatively weak compared with the computing power of the cloud computing platform. At the same time, the distance between the user's smart meter and the site is relatively far. Therefore, during the task issuance process, the monitoring unit first constructs a fault-tolerant group according to the physical topological structure of the smart meter. The smart meters in the same area are divided into different fault-tolerant groups to prevent the failure of task issuance or the inability to return within the specified time after the task is completed due to regional power supply failure or network failure. The construction of the fault-tolerant group can also be constructed in other ways. After the task is issued, the smart meter calculates the task and then returns the result to the cloud computing platform. After receiving the returned result, the cloud computing platform summarizes the result and returns it to the user or terminal. In the process of collecting task results, the cloud computing platform can adopt performance-first and reliability-first solutions. The performance-first solution is to use the result returned first by the same fault-tolerant group as the calculation result, and directly discard the results returned later. Because the calculations of smart meters in the fault-tolerant group are mostly the same, using the first returned result can greatly save time. The reliability-first solution is that the cloud computing platform waits for all smart meters in the fault-tolerant group to return results and then uses voting to verify the results. The majority of results are used as the final calculation result. This strategy is mainly used for applications with strict reliability requirements.
[0038] In order to enable the technical personnel in the field of the present application to have a clearer understanding of the technical solution provided in the present application, the present application also provides the following specific implementation methods:
[0039] like Figure 1The figure shows the structure of a unified computing platform based on cloud computing platform and smart meters. In terms of physical topology, the Internet is used to interconnect smart meters and cloud computing platform. In terms of logical topology, user smart meters are deployed to run corresponding cloud computing applications to realize the control of smart meters by the site cloud computing platform. In this way, the unification between cloud computing platform and smart meters is realized. During the operation of the cloud computing platform, the smart meter communicates with the cloud computing platform through periodic heartbeat information. If no signal is received within the set time period, the smart meter is considered to be disconnected. The cloud computing platform then actively sends a connection request and sends inquiry information to its physically adjacent smart meters to locate the source of the smart meter failure. If it is due to network reasons, the cloud computing platform only marks the smart meter offline. If it is due to a smart meter failure, the cloud computing platform issues a corresponding warning message and staff will come to repair it. Among them, the periodic heartbeat information includes a specific data packet, and the data packet includes the self-test information of the smart meter. The errors in the self-test information are sent to the cloud platform through the data packet, and the cloud platform records the errors. The cloud platform cannot analyze some sudden failures. Network failures can cause the smart meter to lose connection, and the cloud computing platform can obtain the corresponding information. Smart meters randomly add their own physical topology information when sending periodic heartbeat signals. After receiving this information, the cloud computing platform uses the physical topology structure of the smart meters to construct a fault-tolerant group and divides the smart meters in different physical locations into the same fault-tolerant group. In the subsequent task distribution process, the same task is sent to different smart meters in the same fault-tolerant group to achieve reliable operation of the task.
[0040] like Figure 2 The figure shows the task division diagram of the smart meter. The figure shows the task format issued from the cloud computing platform. A total task consists of multiple small tasks. Different subtasks have different task attributes. After the tasks cut by the cloud computing platform are transmitted to the smart meter, the smart meter first analyzes the tasks to be performed, and then the CPU chip starts to load the tasks. The tasks assigned to the FPGA chip are also guided by the CPU chip. The CPU chip first analyzes the information of the task, and then loads the task into the FPGA chip for operation. The FPGA chip can generally run multiple tasks at the same time, but due to limited resources, the tasks running at the same time cannot conflict. Therefore, the CPU chip will reorganize and sort the tasks assigned to the FPGA chip, and load the tasks into the FPGA chip in turn. And interact during the operation of the FPGA chip to control the completion of the task. After the FPGA chip completes the task, the result is retrieved, packaged and merged with other tasks, and the result is transmitted back.
[0041] like Figure 3The figure shows that the site cloud computing platform cross-sends the same task to different smart meters for execution. The cloud computing platform first obtains the physical locations of different smart meters, and establishes a corresponding tree structure based on the physical locations. Each smart meter is represented by a node on the tree. When performing multiple copies of the task, the cloud computing platform selects several relatively dispersed nodes as the smart meters selected for the task execution based on the distance between the tree structures, and then crosses the tasks. Figure 3 In the figure, the task is sent to two smart meters for execution. Here, the number of replicas can be adjusted according to the importance of the task to ensure the reliability and correctness of the task.
[0042] It should be noted that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that an article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprises a..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0043] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest range consistent with the principles and novel features disclosed herein.
[0044] It should be understood that the present application is not limited to what has been described above and shown in the drawings, and various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.
Claims
1. A unified cloud computing system for ubiquitous power Internet of Things, characterized in that: include: The user platform includes at least one smart meter, the smart meter includes a communication chip, a computing chip and a storage chip, the smart meter is connected to the cloud computing platform through the communication chip, and the smart meter communicates with the cloud computing platform through periodic heartbeat information; The computing chip is used to execute the computing tasks sent by the cloud computing platform and return the computing results to the cloud computing platform through the communication chip. The computing chip at least includes: a CPU chip, an FPGA chip and an ASIC chip. The CPU chip, the FPGA chip and the ASIC chip are all integrated inside the smart meter. The storage chip is used to store the operation data and calculation data generated by the smart meter. A site-based cloud computing platform, on which a server and a monitoring unit for monitoring the server are running, wherein the monitoring unit includes a first task scheduling unit, which is used to collect the operation status data of the cloud computing platform and the smart meter in real time, and to mark and divide the computing tasks, obtain the marked computing tasks after the marking and division, and send the marked computing tasks to a second task scheduling unit, wherein the first task scheduling unit constructs a fault-tolerant group according to the physical topological structure of the smart meter during the task issuance process, and adopts a performance priority or reliability priority scheme during the task result collection process; A second task scheduling unit is run in the smart meter, and is used for cutting and dividing the marked calculation task after receiving the marked calculation task, and sending the cut and divided marked calculation task to the computing chip.
2. The unified cloud computing system for ubiquitous power Internet of Things according to claim 1 is characterized in that: The operation status data includes the resource utilization rate of the cloud computing platform, the busyness of the smart meter, the physical topology of the smart meter and the load status of the computing chip.
3. The unified cloud computing system for ubiquitous power Internet of Things according to claim 1 is characterized in that: The communication connection includes a wireless network, a private network or a public Internet of Things.
4. The unified cloud computing system for ubiquitous power Internet of Things according to claim 1 is characterized in that: The smart meter runs an operating system to support the running of multiple applications.
5. The unified cloud computing system for ubiquitous power Internet of Things according to claim 1 is characterized in that: The computing tasks include computing-intensive tasks, and the computing-intensive tasks include encryption and decryption and numerical calculations.
