Sinsegye: With the design concept of integrating computing and control, we aim to solve industrial control problems

From cloud computing to big data, from artificial intelligence to industrial Internet, these advanced technologies have brought unprecedented opportunities and challenges to industrial manufacturing. With the development of technology, many industrial equipment require data exchange and real-time control, which also puts forward higher requirements for the computing power of control equipment.

As a company with strong technological advantages in PC control, dual domain operating system development, motion control algorithms, and artificial intelligence algorithms, Sinsegye is committed to providing efficient, stable, and reliable industrial automation and intelligent solutions for industrial scenarios.

During an interview with Gong Gong, Song Jianwei, an expert in embedded software and operating system products at Sinsegye, stated that since its establishment, Sinsegye has been committed to developing AI products that integrate computing and control based on PC architecture. This not only provides users with diverse functional options and expansion possibilities, but also elevates PC control performance to new heights, achieving efficient utilization of system resources.

Sinsegye Embedded Software and Operating System Product Expert Song Jianwei  

Subverting traditional control architecture, integrating computing and control to unleash industrial potential

In the traditional split architecture, the control system is divided into two main parts: the upper computer and the lower computer, mainly composed of an industrial control computer and various PLC controllers. Not only is it bulky, but the real-time performance of the system is difficult to improve with the computing power of the industrial control computer, which cannot meet the processing of large amounts of real-time data. Deploying multiple controllers in the system leads to a geometric increase in system complexity.

Compared with the traditional combination of industrial personal computer and PLC, the intelligent machine products of Sinsegye combine edge computing with control. Through a set of controller and a set of software and programs, they can simultaneously have the functions of motion control, logic control, machine vision, configuration display, etc., which can make the industrial control equipment better adapt to the complex and changing production environment, and have higher real-time performance and computational scalability.

Traditional architectures are difficult to achieve effective breakthroughs in handling the real-time and stability requirements involved in a large number of servo motors. When the number of axes exceeds 32, the complexity of traditional architectures doubles. By simplifying the architecture, artificial intelligence machines can easily cope with large-scale scenarios with 256 axes, "Song Jianwei added.

In addition, traditional communication between upper and lower computers has been replaced by cross domain shared memory in artificial intelligence machines, resulting in data throughput 2-3 orders of magnitude faster than traditional Ethernet communication. Compared to traditional architectures, due to the lack of excessive communication configuration and line maintenance, the integration of AI programming and IEC standard process programming can also reduce deployment costs, personnel requirements, and corresponding costs for the application of artificial intelligence machines.

In terms of software, Sinsegye has launched a dual domain operating system MetaOS and MetaHypervisor virtualization management software based on Xenomai, which can achieve efficient resource management and precise task scheduling. Through software definition, Sinegye has achieved scalability in hardware computing power, and through the isolation of real-time and non real time domains of the operating system, as well as the configuration of MetaHypervisor, the computing power is reasonably allocated and flexibly scaled according to the customer's on-site needs. It truly achieves the application of computing power to key industrial needs on the customer's site through software definition.

In addition, the Codesys PLC platform is seamlessly integrated, and Sinsegye has also launched MetaFacture/MetaRTE software, which not only supports IEC61131-3 programming, compilation, download, and operation functions, but also supports common industrial communications such as EtherCAT, Profinet, EthernetIP, DeviceNet, CANOpen, Modbus, OPC, etc. It also integrates hundreds of Sinsegye's self-developed visual algorithm function blocks, such as shape recognition, color recognition, measurement, text recognition, etc.

With the help of the Xenomai framework, under the full load of non real time domain CPU and GPU pressure, the industrial intelligence machine achieves stable output capability of EtherCAT 128 axis 1-millisecond DC synchronization, and high real-time performance with jitter less than 30 microseconds. Sinegye's industrial intelligence machine products have shown significant advantages in control and can meet strict control accuracy requirements. In addition, AI products also have powerful computing power. This is mainly due to close cooperation with hardware such as Intel, Hikvision, Nvidia, Cambrian, etc. The powerful computing power of these partners endows AI machines with outstanding performance in artificial intelligence applications, thereby bringing users a new computerized automation experience.

With stable performance, industrial intelligence machines using the Xenomai framework have played a key role in many industries. For example, the control and production decision-making system provided by Sinsegye for leading polysilicon enterprises has achieved stable operation for 15 consecutive months. By deploying the intelligent control system MesPV for polycrystalline silicon reduction furnaces, functions such as furnace management, batch management, curve viewing, parameter control, and issuance can be achieved. Real time judgment of the reaction situation inside the furnace, identification of the atomization status inside the furnace, and reduction of electricity consumption can be achieved. Each batch can save 26000-52000 kWh of electricity.

The outstanding performance of AI machines is due to the joint efforts of Sinsegye's excellent R&D team and professional service team, as well as the adoption of the widely validated and recognized open-source Xenomai framework and open-source Linux foundation by IT and OT, which ensure the stability and reliability of AI products in various devices, on-site environments, and other environments.

Sinsegye provides control and production decision-making systems for leading polysilicon enterprises

Dual domain operating system, creating an efficient and stable system environment

The launch of artificial intelligence machines not only effectively solves the integration problem between OT and IT, but also has the ability to solve key issues such as high computing power demand, mutual interference, high cost, and insufficient reliability. However, in complex industrial environments, different control requirements need to be processed simultaneously by an AI machine. How can efficient allocation and management of resources be achieved, while ensuring mutual isolation between real-time and non real time tasks?

During the development process, the main challenge is how to achieve isolation between real-time and non real time domains on a single device, "said Song Jianwei. The real-time domain is mainly responsible for tasks that require real-time response, such as motion control algorithms and communication with motor servos. The execution of algorithms such as motion control interpolation and electronic cam also requires a large amount of computing resources. The non real time domain is mainly used for tasks such as image recognition, video stream processing, and artificial intelligence inference. These tasks do not require high real-time performance, but have a significant demand for data throughput and computing resources.

Sinsegye's MetaOS is based on Xenomai+MetaHypervisor, which achieves complete isolation between real-time and non real time domains, and desktop resources can be allocated through MetaHypervisor. Even when there are updates, restarts, or blue screens in the desktop system, real-time process control tasks within the domain will not be affected.

Just like the intelligent clone function of mobile phones, using MetaHypervisor to virtualize resource management can flexibly partition physical resources to handle real-time and non real time work resources separately. Depending on the nature of the job, tasks with high time sensitivity can be processed in the real-time domain; Tasks with low time sensitivity can be processed in non real time domain. In this way, real-time and non real time tasks can avoid mutual interference.