AI computer control circuit based on domestic processor

Abstract

The utility model provides an AI computer control circuit based on a domestic processor, which comprises a power supply in power supply connection with a domestic processor main control circuit, and the domestic processor main control circuit is in communication connection with an AI computing circuit and a credible encryption circuit. The output end of the domestic processor main control circuit can also be connected with an electronic device with a WiFi communication function, an electronic device with a Bluetooth communication function, a USB interface electronic device, a Type-C interface electronic device, a mobile storage device, a DP signal display device and a DP signal-to-HDMI signal display device. The domestic processor main control circuit is internally provided with a main control chip U1 with the model of Feiteng D3000M, and the AI calculation circuit is internally provided with an AI calculation chip M2. The computer has the beneficial effects that the computer is extremely adaptive to localization application, the power consumption, the size and the production cost of a low-integration-level computer are greatly reduced, and the safety is higher.

Description

Technical Field

[0001] This utility model relates to the field of computer circuit technology, specifically to an AI computer control circuit based on a domestically produced processor. Background Technology

[0002] A computer, commonly known as a PC, is a modern electronic computing machine used for high-speed calculations. It can perform numerical calculations, logical calculations, and has storage and memory functions. It is a modern intelligent electronic device capable of running programs and automatically and rapidly processing massive amounts of data.

[0003] With the development of artificial intelligence technology, especially the widespread application of large language models such as DeepSeek and Tongyiqianwen, higher demands are placed on the computing power of computer equipment. Traditional general-purpose, low-integration computers have performance bottlenecks when handling these computationally demanding tasks, and often cannot meet the requirements for real-time performance and efficiency.

[0004] Currently, high-performance AI computers on the market mainly rely on imported CPU and GPU architectures. However, these solutions are insufficient in terms of domestic adaptation and support for localized applications. For example, they are not well adapted to DeepSeek. Moreover, imported CPUs and GPUs are usually very expensive, which does not provide a good user experience for enterprises and users. Utility Model Content

[0005] To address the problems in existing technologies, this utility model provides an AI computer control circuit based on a domestically produced processor. By setting up a power supply, a domestically produced processor main control circuit, an AI computing circuit, and a trusted encryption circuit in a mutually cooperating manner, the domestically produced processor main control circuit can utilize the AI ​​computing chip M2 and the trusted encryption circuit to improve the AI ​​computing performance and security performance of the AI ​​computer. This enables a low-power, highly integrated computer combination solution using the Phytium D3000M main control chip and a small M.2 AI card, which is highly adaptable to localized applications. It significantly reduces the power consumption, size, and production cost of low-integration computers, while providing higher security. This solves the problem of high cost and low compatibility of imported CPU and GPU architectures in the domestic AI computer market.

[0006] This utility model provides an AI computer control circuit based on a domestically produced processor, including a power supply, a main control circuit of the domestically produced processor, an AI computing circuit, and a trusted encryption circuit. The output terminal of the power supply is connected to the main control circuit of the domestically produced processor. The main control circuit of the domestically produced processor is communicatively connected to the AI ​​computing circuit and the trusted encryption circuit. The output terminal of the main control circuit of the domestically produced processor can also connect to electronic devices with WiFi communication function, electronic devices with Bluetooth communication function, USB interface electronic devices, Type-C interface electronic devices, mobile storage devices, DP signal display devices, and DP signal to HDMI signal display devices. The main control circuit of the domestically produced processor is equipped with a main control chip U1 of model Phytium D3000M, and the AI ​​computing circuit is equipped with an AI computing chip M2. The main control circuit of the domestically produced processor can call the AI ​​computing chip M2 and the trusted encryption circuit to improve the AI ​​computing performance and security performance of the AI ​​computer.

[0007] This utility model is further improved in that the main control chip U1 has 1148 pins. Pin A1 of the main control chip U1 is connected to the output terminal of the power supply. Pins A13, B12, E13, F12, C15, B14, G15, F14, A17, B16, E17, F16, C19, B18, G19, F18, J11, and H10 of the main control chip U1 are communicatively connected to the AI ​​computing circuit. Pins G27 and F26 of the main control chip U1 are communicatively connected to the trusted encryption circuit. Pins K28 and J29 of the main control chip U1 can connect to electronic devices with Bluetooth communication capabilities. Pins A21, B20, E21, and F20 of the main control chip U1 can connect to... For electronic devices with WiFi communication capabilities, pins BR1, BP2, BN7, and BM62 of the main control chip U1 can be connected to USB interface electronic devices; pins BH2 and BG1 of the main control chip U1 can be connected to Type-C interface electronic devices; pins BR25, BP26, BW25, and BV26 of the main control chip U1 can be connected to mobile storage devices; pins BW5, BV6, BR9, BP10, BW3, BY4, BR7, and BT8 of the main control chip U1 can be connected to DP signal display devices; and pins BR17, BP18, BR15, BT16, BR13, BP14, BR11, and BT12 of the main control chip U1 can be connected to DP signal to HDMI signal display devices.

[0008] This utility model is further improved in that the AI ​​computing chip M2 has 75 pins. The 5th, 7th, 11th, 13th, 17th, 19th, 23rd, 25th, 29th, 31st, 35th, 37th, 41st, 43rd, 47th, 49th, 53rd, and 55th pins of the AI ​​computing chip M2 are respectively connected to the A13th, B12th, E13th, F12th, C15th, B14th, G15th, F14th, A17th, B16th, E17th, F16th, C19th, B18th, G19th, F18th, J11th, and H10th pins of the main control chip U1. The 2nd and 4th pins of the AI ​​computing chip M2 are connected to a 3.3V constant voltage power supply.

[0009] The present invention is further improved by providing a trusted encryption chip U11 in the trusted encryption circuit. The trusted encryption chip U11 has 33 pins. The 20th and 21st pins of the trusted encryption chip U11 are connected to the G27th and F26th pins of the main control chip U1, respectively.

[0010] This utility model is further improved, and the model number of the AI ​​computing chip M2 is M2_M_KEY NASM0-B6701-TP15.

[0011] This utility model is further improved, and the trusted encryption chip U11 is model number Z32H330TC-SQN-2701.

[0012] Compared with existing technologies, the beneficial effects of this utility model are: it provides an AI computer control circuit based on a domestically produced processor. By setting up a power supply, a domestically produced processor main control circuit, an AI computing circuit, and a trusted encryption circuit in the AI ​​computer control circuit based on a domestically produced processor, the domestically produced processor main control circuit can call upon the AI ​​computing chip M2 and the trusted encryption circuit to improve the AI ​​computing performance and security performance of the AI ​​computer. It can realize a low-power, highly integrated computer combination solution with the Phytium D3000M main control chip and an M.2 small-size AI card, which is extremely suitable for localized applications and significantly reduces the power consumption, size, and production cost of low-integration computers. The powerful computing power of the Phytium D3000M and the acceleration capabilities of the small M.2 AI card enable it to complete complex AI tasks in a short time, significantly improving the performance of AI computers. The overall system power consumption is lower than that of traditional CPU and GPU architectures, with only about 12W when playing 1080P video. It is more suitable for edge computing scenarios that run for a long time. It supports a variety of large models and AI frameworks, making it highly adaptable and widely applicable in scientific research, education, industry and other fields. Moreover, the Phytium D3000M's built-in security mechanism, combined with the trusted module, provides additional protection for sensitive data processing, making it more secure. It solves the problem that existing AI computers in the Chinese domestic market mainly rely on imported CPU and GPU architectures, which are costly and have low compatibility. Attached Figure Description

[0013] To more clearly illustrate the solutions in this utility model or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this utility model. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0014] Figure 1 This is a schematic diagram of an AI computer control circuit based on a domestically produced processor according to this utility model.

[0015] Figure 2 This is a circuit diagram of the main control circuit of the domestically produced processor of this utility model;

[0016] Figure 3 This is a circuit diagram of the AI ​​computing circuit of this utility model;

[0017] Figure 4 This is a circuit diagram of the trusted encryption circuit of this utility model. Detailed Implementation

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein in the specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this invention are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or accompanying drawings of this invention are used to distinguish different objects, not to describe a particular order.

[0019] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the present invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0021] like Figures 1-4As shown, this utility model provides an AI computer control circuit based on a domestically produced processor, including a power supply, a main control circuit of the domestically produced processor, an AI computing circuit, and a trusted encryption circuit. The output of the power supply is connected to the main control circuit of the domestically produced processor for power supply. The main control circuit of the domestically produced processor is communicatively connected to the AI ​​computing circuit and the trusted encryption circuit. The output of the main control circuit of the domestically produced processor can also connect to electronic devices with WiFi communication function, electronic devices with Bluetooth communication function, USB interface electronic devices, Type-C interface electronic devices, mobile storage devices, DP signal display devices, and DP signal to HDMI signal display devices. The main control circuit of the domestically produced processor is equipped with a main control chip U1 of model number Phytium D3000M, and the AI ​​computing circuit is equipped with an AI computing chip M2. In this embodiment, the domestic processor's main control circuit can utilize the AI ​​computing chip M2 and trusted encryption circuit to enhance the AI ​​computing and security performance of the AI ​​computer. This enables a low-power, highly integrated computer combination solution using the Phytium D3000M main control chip and a small-size M.2 AI card. This solution is highly suitable for localized applications, significantly reducing the power consumption, size, and production cost of low-integration computers. The powerful computing power of the Phytium D3000M and the acceleration capabilities of the M.2 small-size AI card can complete complex AI tasks in a short time, greatly improving the performance of the AI ​​computer. The overall system power consumption is lower than that of traditional CPU and GPU architecture solutions, consuming only about 12W when playing 1080P video. This makes it more suitable for edge computing scenarios that operate for extended periods. It supports various large models and AI frameworks, is highly adaptable, and can be widely used in scientific research, education, industry, and other fields. Furthermore, the Phytium D3000M's built-in security mechanism, combined with the trusted module, provides additional protection for sensitive data processing, resulting in higher security.

[0022] like Figure 2As shown, the main control chip U1 has 1148 pins. Pin A1 of the main control chip U1 is connected to the output of the power supply. Pins A13, B12, E13, F12, C15, B14, G15, F14, A17, B16, E17, F16, C19, B18, G19, F18, J11, and H10 of the main control chip U1 are connected to the AI ​​computing circuit. Pins G27 and F26 of the main control chip U1 are connected to the trusted encryption circuit. Pins K28 and J29 of the main control chip U1 can be connected to electronic devices with Bluetooth communication capabilities. Pins A21, B20, E21, and F20 of the main control chip U1 can be connected to devices with WiFi communication capabilities. For electronic devices, pins BR1, BP2, BN7, and BM62 of the main control chip U1 can be connected to USB interface electronic devices; pins BH2 and BG1 of the main control chip U1 can be connected to Type-C interface electronic devices; pins BR25, BP26, BW25, and BV26 of the main control chip U1 can be connected to mobile storage devices; pins BW5, BV6, BR9, BP10, BW3, BY4, BR7, and BT8 of the main control chip U1 can be connected to DP signal display devices; and pins BR17, BP18, BR15, BT16, BR13, BP14, BR11, and BT12 of the main control chip U1 can be connected to DP signal to HDMI signal display devices. In this embodiment, Phytium D3000M is a high-performance processor based on ARM architecture, featuring eight 64-bit FTC862 cores, supporting multi-threaded parallel processing. Its built-in security features provide reliable protection for AI tasks. The main control chip U1 and the AI ​​computing chip M2 communicate via a 4-channel PCIe 3.0 interface, with a maximum speed of 8GB / s.

[0023] like Figure 3As shown, the AI ​​computing chip M2 has the model number M2_M_KEY NASM0-B6701-TP15. The AI ​​computing chip M2 has 75 pins. Pins 5, 7, 11, 13, 17, 19, 23, 25, 29, 31, 35, 37, 41, 43, 47, 49, 53, and 55 of the AI ​​computing chip M2 are connected to pins A13, B12, E13, F12, C15, B14, G15, F14, A17, B16, E17, F16, C19, B18, G19, F18, J11, and H10 of the main control chip U1, respectively. Pins 2 and 4 of the AI ​​computing chip M2 are connected to a 3.3V constant voltage power supply. In this embodiment, the AI ​​computing chip M2 is designed based on the M.2 form factor, with a small size (2280 standard), making it suitable for embedded environments. The AI ​​computing chip integrates dedicated AI acceleration chips (NPU and TPU) specifically for matrix operations and neural network inference, supporting multiple precision modes such as FP16 / INT8 to meet the performance requirements of different scenarios. The system supports dynamic link adjustment, optimizing bandwidth allocation according to the actual load. The AI ​​computing chip M2 is equipped with LPDDR4, and the D300M is equipped with LPDDR5 high-speed memory with a maximum capacity of 32GB, supporting NVMe SSD solid-state drives for loading large model parameters and caching intermediate results.

[0024] like Figure 4 As shown, the trusted encryption circuit includes a trusted encryption chip U11, model Z32H330TC-SQN-2701. The trusted encryption chip U11 has 33 pins, with pins 20 and 21 connected to pins G27 and F26 of the main control chip U1, respectively. In this embodiment, the trusted encryption chip U11 uses the Z32H330TC trusted module chip from Guomin Technology. The Z32H330TC is the first dual-algorithm trusted computing core product in the international trusted computing industry, incorporating both Chinese and international cryptographic algorithms. It fully supports multiple domestic computing platforms such as Zhaoxin, Loongson, Phytium, Hygon, and Huaxintong, supports a high-speed SPI interface, and provides hardware-isolated secure cryptographic algorithm services, platform integrity protection, and remote platform authentication. This chip is widely used in various servers, computers, tablets, smartphones, embedded systems, IoT devices, and other terminal devices. In the AI ​​computer of this invention, the main control chip U1 needs to verify the key with the trusted encryption chip U11 before accessing the BIOS. Only after the key verification is successful can the BIOS be loaded and the subsequent boot process proceed.

[0025] As can be seen from the above, this utility model provides an AI computer control circuit based on a domestically produced processor. By setting up a power supply, a main control circuit of the domestically produced processor, an AI computing circuit, and a trusted encryption circuit in the AI ​​computer control circuit based on the domestically produced processor, the main control circuit of the domestically produced processor can call the AI ​​computing chip M2 and the trusted encryption circuit to improve the AI ​​computing performance and security performance of the AI ​​computer. It can realize a low-power, highly integrated computer combination solution with the Phytium D3000M main control chip and the M.2 small-size AI card, which is extremely suitable for localized applications and significantly reduces the power consumption, size, and production cost of low-integration computers. The powerful Phytium D3000M... The computing power and acceleration capabilities of the M.2 small-size AI card enable complex AI tasks to be completed in a short time, significantly improving the performance of AI computers. The overall system power consumption is lower than that of traditional CPU and GPU architectures, with only about 12W when playing 1080P video. It is more suitable for edge computing scenarios that run for a long time. It supports a variety of large models and AI frameworks, and is highly adaptable. It can be widely used in scientific research, education, industry and other fields. Moreover, the built-in security mechanism of Phytium D3000M, combined with the trusted module, provides additional protection for sensitive data processing, making it more secure. It solves the problem that the current AI computers on the Chinese market mainly rely on imported CPU and GPU architectures, which are costly and have low compatibility.

[0026] The specific embodiments described above are preferred embodiments of this utility model, and are not intended to limit the specific scope of this utility model. The scope of this utility model includes but is not limited to the specific embodiments described above. All equivalent changes made in accordance with this utility model are within the protection scope of this utility model.

Claims

1. An AI computer control circuit based on a domestically produced processor, characterized in that: The system includes a power supply, a domestically produced processor main control circuit, an AI computing circuit, and a trusted encryption circuit. The output of the power supply is connected to the domestically produced processor main control circuit. The domestically produced processor main control circuit is communicatively connected to the AI ​​computing circuit and the trusted encryption circuit. The output of the domestically produced processor main control circuit can also connect to electronic devices with WiFi communication function, electronic devices with Bluetooth communication function, USB interface electronic devices, Type-C interface electronic devices, mobile storage devices, DP signal display devices, and DP signal to HDMI signal display devices. The domestically produced processor main control circuit contains a Phytium D3000M main control chip U1, and the AI ​​computing circuit contains an AI computing chip M2. The domestically produced processor main control circuit can utilize the AI ​​computing chip M2 and the trusted encryption circuit to improve the AI ​​computing performance and security performance of the AI ​​computer.

2. The AI ​​computer control circuit based on a domestically produced processor according to claim 1, characterized in that: The main control chip U1 has 1148 pins. Pin A1 of the main control chip U1 is connected to the output terminal of the power supply. Pins A13, B12, E13, F12, C15, B14, G15, F14, A17, B16, E17, F16, C19, B18, G19, F18, J11, and H10 of the main control chip U1 are communicatively connected to the AI ​​computing circuit. Pins G27 and F26 of the main control chip U1 are communicatively connected to the trusted encryption circuit. Pins K28 and J29 of the main control chip U1 can connect to electronic devices with Bluetooth communication capabilities. Pins A21, B20, E21, and F20 of the main control chip U1 can connect to devices with WiFi communication capabilities. The main control chip U1 has the following functions: pins BR1, BP2, BN7, and BM62 can be connected to USB interface electronic devices; pins BH2 and BG1 can be connected to Type-C interface electronic devices; pins BR25, BP26, BW25, and BV26 can be connected to mobile storage devices; pins BW5, BV6, BR9, BP10, BW3, BY4, BR7, and BT8 can be connected to DP signal display devices; and pins BR17, BP18, BR15, BT16, BR13, BP14, BR11, and BT12 can be connected to DP signal to HDMI signal display devices.

3. The AI ​​computer control circuit based on a domestically produced processor according to claim 2, characterized in that: The AI ​​computing chip M2 has 75 pins. Pins 5, 7, 11, 13, 17, 19, 23, 25, 29, 31, 35, 37, 41, 43, 47, 49, 53, and 55 of the AI ​​computing chip M2 are connected to pins A13, B12, E13, F12, C15, B14, G15, F14, A17, B16, E17, F16, C19, B18, G19, F18, J11, and H10 of the main control chip U1, respectively. Pins 2 and 4 of the AI ​​computing chip M2 are connected to a 3.3V constant voltage power supply.

4. The AI ​​computer control circuit based on a domestically produced processor according to claim 3, characterized in that: The trusted encryption circuit includes a trusted encryption chip U11, which has 33 pins. Pins 20 and 21 of the trusted encryption chip U11 are connected to pins G27 and F26 of the main control chip U1, respectively.

5. The AI ​​computer control circuit based on a domestically produced processor according to claim 4, characterized in that: The model number of the AI ​​computing chip M2 is M2_M_KEY NASM0-B6701-TP15.

6. The AI ​​computer control circuit based on a domestically produced processor according to claim 5, characterized in that: The trusted encryption chip U11 is model number Z32H330TC-SQN-2701.

Images