A kind of FPGA-based internet of things edge computing power scheduling physical card for power generation enterprise

Abstract

This invention relates to a physical card for FPGA-based IoT edge computing scheduling in power generation enterprises, comprising a PCIe transmission module, a digital signal processor, a programmable logic processor, a memory module, a cache memory, a communication interface module, and a software module. The PCIe transmission module is connected to the digital signal processor, the programmable logic processor, and the cache memory, and the cache memory is connected to the memory module. The PCIe transmission module is used to connect to PC-like devices within the power generation enterprise, enabling unified management of these PC-like devices by an IoT management platform. The software module is used to implement host status monitoring, IoT MQTT communication, secure communication encryption, real-time edge data acquisition, and supporting unified edge computing services. This invention enables ordinary PC-like hosts to be uniformly managed by an IoT management platform while also possessing the capabilities of a dedicated edge computing terminal.

Description

Technical Field

[0001] This invention belongs to the field of power technology, and in particular relates to a physical card for FPGA-based IoT edge computing scheduling in power generation enterprises. Background Technology

[0002] Currently, the Internet of Things (IoT) technology has become a key component of the energy industrial internet, providing strong technical support for its model-driven and data-driven aspects. This requires deploying sensing terminals as comprehensively as possible in all production and management stages of energy production units to acquire rich and diverse data information. Relying on real-time and highly reliable data transmission technology, this information is transmitted to a data platform for data mining, fusion analysis, and feedback of the analysis results. This meets the inherent business needs of planning and construction, production decision-making, operation and maintenance, monitoring and control, and asset management during the expansion of industrialization and informatization and the process of intelligentization.

[0003] However, the application of IoT technology in power generation companies is not widespread, and the connectivity rate of industrial equipment is low. The main reasons for this situation are as follows:

[0004] 1) Cost reasons:

[0005] The price of these terminals ranges from several thousand yuan to tens of thousands of yuan, with edge terminals equipped with AI GPUs reaching as high as 100,000 yuan per unit. If there are a large number of these terminals, this undoubtedly increases the cost of information technology investment for power production units that are not already "wealthy".

[0006] 2) Edge computing has low utilization of existing machines.

[0007] Compared to other industries, power generation companies have relatively well-developed infrastructure in terms of information technology, networking, and automation. Their production and management areas possess a large number of PCs, industrial control computers, host computers, workstations, servers, and other machines. According to incomplete statistics, the average resource utilization rate of these machines is less than 50%. However, with the rapid development of the Internet of Things (IoT), various IoT terminal devices, represented by edge computing, are gradually converging with PC-like devices in terms of functionality and application. In other words, the boundary between edge computing terminals and PCs is becoming increasingly blurred. Therefore, utilizing existing PC-like devices as an important component of IoT terminal devices, alongside edge computing terminals, to jointly complete edge governance and edge computing for cloud-edge collaborative applications, not only fully utilizes existing "idle" computing power but also saves power generation companies investment and procurement costs.

[0008] However, the diversity of these numerous PCs, industrial PCs, host computers, workstations, servers, and other PC-like devices—including their varying architectures, operating systems, stability, and security requirements—poses a significant challenge for unified management by IoT management platforms. Because IoT management platforms need to deploy small applications such as "status monitoring agents," "MQTT communication agents," "secure communication encryption agents," and "edge data acquisition apps" within the managed terminal devices, it's difficult to effectively utilize this vast amount of "idle computing power" without a simple (easy to install and use) and reliable technology (product) that can rapidly transform these PC-like devices into "edge terminals."

[0009] 3) A key reason why many IoT application vendors recommend customized edge computing terminals to enterprise customers is to ensure information security. For example, to ensure secure communication between edge terminal devices and IoT management systems, most manufacturers use asymmetric encryption algorithms such as RSA and SH2. This requires placing the edge computing terminal's private key and the platform's public key within the terminal for authentication and decryption. Customized edge computing terminals provide a relatively closed yet secure operating environment.

[0010] However, the security levels of the numerous PCs used by power generation companies vary greatly. Some still use operating systems like Windows XP, some allow the free insertion and removal of USB flash drives and other removable storage media, and some lack any security hardening software. Therefore, without a simple (easy to install and use) yet highly secure technology (product) that can guarantee the rapid "edge terminal" transformation of PCs, it will be difficult to effectively utilize these numerous "idle computing powers." Summary of the Invention

[0011] The purpose of this invention is to provide a physical card for FPGA-based IoT edge computing scheduling in power generation enterprises. This physical card is interconnected with PC-type devices through a PCIe interface, and through four built-in integrated software modules—host status monitoring, IoT MQTT communication, secure communication encryption, and real-time edge acquisition—as well as a supporting unified edge computing service driver interface software module, it enables PC-type devices to possess the basic capabilities of conventional edge computing terminals.

[0012] This invention provides a physical card for IoT edge computing power scheduling based on FPGA for power generation enterprises, including a PCIe transmission module, a digital signal processor, a programmable logic processor, a memory module, a high-speed buffer memory, and a communication interface module, wherein the programmable logic processor has a built-in software module;

[0013] The PCIe transmission module is connected to the digital signal processor, the programmable logic processor, and the cache memory, and the cache memory is connected to the memory module.

[0014] The PCIE transmission module is used to connect with PC-type machines and devices within the power generation enterprise, so as to enable the PC-type machines and devices to be uniformly managed by the Internet of Things management platform.

[0015] The software module is used to implement host status monitoring, IoT MQTT communication, secure communication encryption, real-time edge data acquisition, and supporting unified edge computing services.

[0016] Furthermore, the PC-type machinery and equipment within the power generation enterprise includes various PCs, industrial control computers, workstations, and servers within the power generation enterprise.

[0017] Furthermore, the PCIE transmission module is integrated with PC-type machines in the production and management areas of the power generation enterprise through PCIE slots.

[0018] Furthermore, the software modules include a host status monitoring software module, an IoT MQTT communication software module, a secure communication encryption software module, an edge real-time acquisition software module, and an edge computing unified service driver interface software module;

[0019] The host status monitoring software module is used for:

[0020] Periodically acquire host PC-type device information, including: hardware information, device communication interface information, MAC address, IP address, memory usage rate, storage usage rate, CPU usage rate, last boot time, and current time;

[0021] Remotely obtain CPU utilization, memory utilization, and storage utilization thresholds issued by the IoT management platform, generate alarms in real time for indicators exceeding the thresholds, and upload them to the IoT management platform;

[0022] The hardware status information of the host PC-type host is periodically monitored and sent to the IoT management platform for computing power analysis. The analysis results are converted into instructions and sent to the physical card. After receiving the instructions, the physical card remotely obtains the specified edge application and executes it in the host PC-type host.

[0023] The IoT MQTT communication software module is used for:

[0024] The built-in MQTT client connects to the IoT management platform for authentication, heartbeat detection, topic message publishing and subscription, and time synchronization.

[0025] After unifying the edge computing results, edge-collected data, and PC-type host status information obtained by the host status monitoring software module, the data is sent to the IoT management platform.

[0026] The secure communication encryption software module is used for:

[0027] An encrypted communication channel is established using a built-in key negotiated with the main station to prevent information from being tampered with or destroyed during transmission.

[0028] The built-in source MAC, source port, target MAC, target port, and protocol key values ​​are used to calculate and verify the security of the collected data.

[0029] The edge real-time acquisition software module is used to acquire real-time production data in the host machine or real-time production data in the field analog-to-digital converter connected to the host machine.

[0030] The edge computing unified service driver interface software module is used for bidirectional communication between edge applications and physical cards, as well as network communication with host PC-like hosts.

[0031] Through the above solution, ordinary PC-type hosts, connected to FPGA-based IoT edge computing power scheduling physical cards used in power generation companies, are now uniformly managed by the IoT management platform and possess the capabilities of edge computing terminals on the market. A large number of PC-type hosts contribute their redundant computing power under the overall scheduling and management of the IoT management platform's computing power scheduling and AI analysis platform. This greatly promotes the construction of IoT, the development of the energy industrial internet, and the popularization of cloud-edge collaborative applications in power generation companies, while also saving customers' IT construction costs.

[0032] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0033] Figure 1 This is a schematic diagram of the installation of the physical card for FPGA-based IoT edge computing power scheduling in power generation enterprises according to the present invention;

[0034] Figure 2 This is a schematic diagram of the physical card for FPGA-based IoT edge computing power scheduling in power generation enterprises according to the present invention;

[0035] Figure 3 This is a schematic diagram of the implementation use case of the physical card for FPGA-based IoT edge computing power scheduling in power generation enterprises. Detailed Implementation

[0036] The following will further describe in detail the specific implementation manners of the present invention in conjunction with the accompanying drawings and embodiments. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.

[0037] Refer Figures 1 to 3 As shown, this embodiment provides a physical card for power generation enterprises' Internet of Things edge computing power scheduling based on FPGA, including a PCIE transmission module, a digital signal processor, a programmable logic processor, a memory module, a cache memory, and a communication interface module. The digital signal processor has a built-in software module;

[0038] The PCIE transmission module is connected to the digital signal processor, the programmable logic processor, and the cache memory, and the cache memory is connected to the memory module;

[0039] The PCIE transmission module is used to connect to PC-class machine devices within power generation enterprises to enable the PC-class machine devices to be uniformly managed by the Internet of Things management platform;

[0040] The software module is used to implement host status monitoring, Internet of Things MQTT communication, secure communication encryption, edge real-time acquisition, and supporting edge computing unified services.

[0041] FPGA, that is, the programmable logic processor, which is the field programmable gate array. It is a product developed further on the basis of programmable devices such as PAL, GAL, and CPLD. FPGA appears as a semi-custom circuit in the field of application-specific integrated circuits. It not only compensates for the deficiencies of custom circuits but also overcomes the shortcomings of the limited number of gate circuits in the original programmable devices. It has the advantages of being digital, serialized, miniaturized, low-power, multi-functional, standardized, good confidentiality, high integration, can be simulated and debugged on-site, and can be programmed repeatedly infinitely. FPGA has independent and powerful computing performance, extremely low power consumption, a compact form factor, and can build-in algorithms according to the actual application scenarios. This physical card gives full play to the powerful parallel processing ability of FPGA, and at the same time does not occupy the computing resources of the host computer, improving the overall performance of the system.

[0042] By installing a physical card based on the PCIE interface for various PC-class machine devices such as PCs, industrial control computers, workstations, and servers within power generation enterprises, it can enable ordinary PC-class hosts to be uniformly managed by the Internet of Things management platform and also possess the capabilities of dedicated edge computing terminals on the market. A large number of PC-class hosts, under the overall scheduling management of the computing power scheduling artificial intelligence analysis platform of the Internet of Things management platform, contribute their redundant computing power, which greatly promotes the construction of the Internet of Things in power generation enterprises, the construction of the energy industrial Internet, and the popularization of cloud-edge collaborative applications. In addition, it also saves the cost of information construction for customer enterprises.

[0043] Power equipment is diverse and geographically dispersed. In this embodiment, the PCIe transmission modules are integrated with PC-type machines in the production and management areas of power generation companies via PCIe slots (similar to graphics cards in computers). This rapidly transforms them into "edge terminals," enabling unified management by an IoT management platform and providing support for edge computing. This significantly reduces the IT infrastructure costs for power generation companies. Interconnecting with various PC-type machines in power generation companies via PCIe slots eliminates the need for traditional terminal devices. This approach allows users to fully utilize various PCs, industrial PCs, workstations, servers, and other PC-type machines, offering plug-and-play convenience and cost savings.

[0044] The FPGA board (programmable logic processor) integrates a host status monitoring software module, an IoT MQTT communication software module, a secure communication encryption software module, and an edge real-time acquisition software module, along with a matching edge computing unified service driver interface software module. This physical card interconnects with PC-like devices through a PCIe interface, enabling PC-like devices to possess the basic capabilities of a conventional edge computing terminal.

[0045] It should be noted that the purpose of the physical card is not to perform all edge computing functions, but to provide the core foundational capabilities of edge autonomy and edge computing for PC-like devices.

[0046] The functions of the physical card software module are described below:

[0047] 1) Host status monitoring software module

[0048] The physical card can periodically obtain information about the host PC-like devices through the host status monitoring software module. The information includes: hardware information, device communication interface information, MAC address, IP address, memory usage, storage usage, CPU usage, last boot time, and current time.

[0049] The physical card can remotely obtain threshold values ​​for indicators such as CPU utilization, memory utilization, and storage utilization issued by the IoT management platform through the host status monitoring software module, and can generate alarms in real time for indicators exceeding the threshold and upload them to the IoT management platform.

[0050] This software module periodically monitors the hardware status information of the host PC-like host and sends it to the IoT management platform. The computing power scheduling and artificial intelligence analysis platform of the IoT management platform will use trained professional model algorithms to perform computing power analysis using core information such as CPU model, CPU utilization, graphics card chip model, graphics card chip utilization, memory model, memory utilization, and network card throughput as parameters. The analysis results will be converted into instructions and sent to the physical card. After receiving the instructions, the physical card will remotely obtain the specified edge application (the complexity of the edge application depends on the algorithm analysis results) and execute it in the host PC-like host.

[0051] 2) IoT MQTT communication software module

[0052] The physical card has a built-in MQTT client, which is based on the standard MQTT protocol system. The basic MQTT functions it provides mainly include connection to the IoT management platform (identity) authentication, heartbeat detection, topic message publishing and message subscription, and time synchronization.

[0053] The physical card uses this software module to uniformly upload data such as edge computing results, edge-collected data, and PC-type host status information obtained by the host status monitoring software module to the IoT management platform.

[0054] 3) Secure communication encryption software module

[0055] To ensure that data exchange between the PC host and the IoT management platform is transmitted in encrypted form, the physical card contains a key negotiated with the main station. The IoT MQTT communication software module uses this key to establish an encrypted communication channel, preventing information from being tampered with or destroyed during transmission.

[0056] The physical card has a built-in CRC20 algorithm to calculate and verify the key value of the "five-tuple" (source MAC, source port, destination MAC, destination port, protocol) of IP packets, thereby ensuring the security of data collected by the edge real-time acquisition software module.

[0057] The FPGA chip in the physical card employs data mirroring technology, maintaining more than two backups of critical data within the chip. The chip logically divides the SRAM into primary and backup partitions. All write operations on critical data are performed simultaneously on both partitions. When both partitions are functioning correctly, data can be read from either partition; if one partition fails, data can still be read from the other.

[0058] 4) Edge Real-time Acquisition Software Module

[0059] The physical card incorporates three of the most commonly used communication acquisition modules for power generation enterprises, namely the Modbus acquisition module, the 104 acquisition module, and the 102 acquisition module. The edge real-time acquisition software module can acquire production real-time data in the host machine (memory database) in real time, or acquire production real-time data in the field digital-to-analog conversion device interconnected with the host machine in real time.

[0060] 5) Edge computing unified service driver interface software module

[0061] The edge computing unified service driver interface belongs to the supporting driver of the physical card. The physical card will remotely obtain the edge applications recommended by the computing power scheduling artificial intelligence analysis platform of the Internet of Things management platform and download them into the host PC-class host for execution. Each downloaded and executed edge application will communicate bidirectionally with the physical card through the unified edge computing unified service driver interface.

[0062] The edge computing unified service driver interface also completes functions such as network communication of the host PC-class host.

[0063] The execution logic of the edge application only cares about the calculation of edge services, and other basic functions related to the Internet of Things are completed by the physical card on behalf. The physical card not only realizes the rapid "edge terminal" of PC-class machines but also provides all the core basic functions of Internet of Things terminal management. The most crucial thing is that the physical card ensures that PC-class machines actually run various edge applications.

[0064] The use and working principle of this physical card are as follows:

[0065] 1) The physical card is interconnected with the PC-class host through the PCIE interface, and the operating system driver of this physical card is installed: In the normal state, the physical card indicator light shows green, otherwise it shows red.

[0066] 2) After the physical card is started, the FPGA chip calls the Internet of Things MQTT communication software module and the secure communication encryption software module to establish a secure connection with the Internet of Things management platform, providing encryption protection for two-way communication during operation.

[0067] 3) The FPGA chip periodically calls the instruction set of the host status monitoring software module, reads the collected hardware status information of the host PC-class host into the physical card cache memory, and then calls the instruction set of the Internet of Things MQTT communication software module to transmit it to the host machine through the PCIE for forwarding. When an alarm of medium level or above occurs, the physical card alarm light shows red.

[0068] 4) For common real-time data acquisition scenarios in power generation enterprises, the physical card, through a digital signal processing (DSP) chip and an FPGA chip, calls the edge real-time acquisition software module to acquire data from the power generation enterprise's control system. It can also efficiently perform exception adjustments (processing only changed data) and data quality pre-analysis on the acquired data. The physical card then calls the IoT MQTT communication software module instruction set to transmit the processed real-time data to the host machine via PCIe for forwarding to the IoT management platform.

[0069] 5) The FPGA chip listens for commands issued by the IoT management platform in real time. Upon receiving the instruction to schedule the edge algorithm, it immediately drives the physical card's driver software to remotely retrieve the edge application from the specified IP address (including port). After the driver software starts the edge application, it also begins the lifecycle management of that edge application. The edge application does not interact directly with the IoT management platform; all interactions are delegated to the physical card by the driver.

[0070] This IoT edge computing scheduling physical card integrates FPGA parallel processing technology, IoT communication technology, IoT edge terminal management technology, edge computing technology, and the interaction between the physical card and the IoT management platform's computing power scheduling AI analysis platform to achieve overall scheduling technology for edge computing applications. By installing a PCIE interface-based physical card on various PC-type machines such as PCs, industrial control computers, workstations, and servers within power generation enterprises, ordinary PC-type hosts can be uniformly managed by the IoT management platform, and also possess the capabilities of dedicated edge computing terminals on the market. This greatly promotes the construction of IoT, the construction of the energy industrial internet, and the popularization of cloud-edge collaborative applications in power generation enterprises, and also saves customers' information technology construction costs.

[0071] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A physical card for FPGA-based IoT edge computing power scheduling in power generation enterprises, characterized in that, It includes a PCIe transmission module, a digital signal processor, a programmable logic processor, a memory module, a cache memory, and a communication interface module, wherein the programmable logic processor has a built-in software module; The PCIe transmission module is connected to the digital signal processor, the programmable logic processor, and the cache memory, and the cache memory is connected to the memory module. The PCIE transmission module is used to connect with PC-type machines and devices within the power generation enterprise, so as to enable the PC-type machines and devices to be uniformly managed by the Internet of Things management platform. The software module is used to implement host status monitoring, IoT MQTT communication, secure communication encryption, real-time edge data acquisition, and supporting unified edge computing services.

2. The physical card for FPGA-based IoT edge computing scheduling in power generation enterprises according to claim 1, characterized in that, The PC-type equipment within the power generation enterprise includes various PCs, industrial control computers, workstations, and servers within the power generation enterprise.

3. The physical card for FPGA-based IoT edge computing scheduling in power generation enterprises according to claim 2, characterized in that, The PCIE transmission module is integrated with PC-type machines in the production and management areas of the power generation company through PCIE slots.

4. The physical card for FPGA-based IoT edge computing scheduling in power generation enterprises according to claim 1, characterized in that, The software modules include a host status monitoring software module, an IoT MQTT communication software module, a secure communication encryption software module, an edge real-time acquisition software module, and an edge computing unified service driver interface software module. The host status monitoring software module is used for: Periodically acquire host PC-type device information, including: hardware information, device communication interface information, MAC address, IP address, memory usage rate, storage usage rate, CPU usage rate, last boot time, and current time; Remotely obtain CPU utilization, memory utilization, and storage utilization thresholds issued by the IoT management platform, generate alarms in real time for indicators exceeding the thresholds, and upload them to the IoT management platform; The system periodically monitors the hardware status information of the host PC-type host and sends it to the IoT management platform for computing power analysis. The analysis results are converted into instructions and sent to the physical card. After receiving the instructions, the physical card remotely obtains the specified edge application and executes it in the host PC-type host. The IoT MQTT communication software module is used for: The built-in MQTT client connects to the IoT management platform for authentication, heartbeat detection, topic message publishing and subscription, and time synchronization. After unifying the edge computing results, edge-collected data, and PC-type host status information obtained by the host status monitoring software module, the data is sent to the IoT management platform. The secure communication encryption software module is used for: An encrypted communication channel is established using a built-in key negotiated with the main station to prevent information from being tampered with or destroyed during transmission. The built-in source MAC, source port, target MAC, target port, and protocol key values ​​are used to calculate and verify the security of the collected data. The edge real-time acquisition software module is used to acquire real-time production data in the host machine or real-time production data in the field analog-to-digital converter connected to the host machine. The edge computing unified service driver interface software module is used for bidirectional communication between edge applications and physical cards, as well as network communication with host PC-like hosts.

Images