T/CES 228-2023 Active Group standards

T/CES 228-2023 Electricity artificial intelligence integrated sensing and computing system testing method

Publish Date: 2023-10-19 Implement Date: 2023-10-20 For services related to genuine standard inquiry, procurement, translation, and other related services in China, please Contact Us

Price: $ 20.00

Out of stock

Basic Information

Standard Code: T/CES 228-2023

Standard Type: Group standards

Standard Status: Active

is_force_gb: no

CCS Name: -

ICS Name: Comprehensive Application of Information Technology

Publish Date: 2023-10-19

Implement Date: 2023-10-20

Scope

Scope: This document specifies the test methods for power artificial intelligence integrated systems of sensing, storage, and computing, which can evaluate the functions of AI-based systems and the performance of typical application scenarios, including image classification, object detection, object segmentation, object recognition, object tracking, behavior detection, speech recognition, text recognition, local voice wake-up, and load identification. This document is applicable to manufacturers, research institutions, application developers, and third-party organizations for testing and evaluating power artificial intelligence integrated systems, as well as to manufacturers for the design of such systems. Main technical content: With the development of new applications such as the ubiquitous power Internet of Things and edge computing, the scale and number of power artificial intelligence models have increased dramatically, and the requirements for data collection, storage, and processing in various application scenarios continue to rise. The traditional architecture of separating storage and processors in data processing leads to the von Neumann bottleneck problem, manifested as storage walls, bandwidth walls, and power consumption walls. How to effectively break through the von Neumann bottleneck and provide computing power support for the development of related technologies has become an urgent issue in the current industry. The storage-computing integration technology directly uses storage devices for data processing, featuring ultra-high computing power and energy efficiency, making it a more suitable architecture for AI computing than the von Neumann architecture. AspenCore predicts that it will be one of the top ten technology trends in the global semiconductor industry in 2022. The "sensing-storage-computing integration" adds sensing to the storage-computing integration architecture, which solves the power consumption bottleneck of data transfer in the