10results about How to "Guaranteed normal execution" patented technology

The invention relates to a gas detection and ventilation equipment linkage control method, system, equipment and medium, and the method comprises the steps: monitoring the gas concentration of a target area in real time through a gas detector, and generating a linkage instruction signal containing a self address identifier when the gas concentration reaches a preset alarm threshold value; the gas detectors directly send the linkage instruction signals to at least one set of specified ventilation equipment through a shared communication bus, and the shared communication bus is the same physical communication link for connecting the plurality of gas detectors, the at least one set of ventilation equipment and a gas alarm controller; and the specified ventilation equipment executes corresponding start-stop or speed regulation operation based on the received linkage instruction signal so as to solve the problems of response delay, single-point failure, single control strategy and the like of the existing centralized linkage system, thereby achieving the purpose of quicker, more reliable and more refined safety linkage control.

The application discloses a process path processing method and device and a semiconductor process equipment. The method comprises the following steps: obtaining the identification of a process path of a process task to be executed; obtaining the information of the corresponding process path according to the identification of the process path, wherein the information of the process path comprises the information of a plurality of process steps arranged in sequence and the information of at least one process chamber corresponding to each process step; adopting a graphical display control to perform graphical display on the process path according to the information of the process path, thereby obtaining a process path control, wherein the process path control comprises a plurality of process step node controls connected in sequence and at least one process chamber node control corresponding to each process step node control, and the attribute information of the process chamber node control comprises the identification of the corresponding process parameter. The application can intuitively show the process path of the material to the user, and can ensure the normal execution of the process flow and realize the confidentiality of the process parameters.

The invention relates to the technical field of vehicle control, and discloses a self-adaptive voice vehicle control method and device, a vehicle and a storage medium, and the method comprises the steps: obtaining first load information which is used for reflecting the load of multi-dimensional hardware in the vehicle; the corresponding current performance level is matched according to the first load information, the reasoning ability of the large language model is compressed according to the current performance level, and the reasoning ability is used for representing the input information amount and the output control message amount of the large language model; and processing a voice instruction of the user based on the compressed large language model to obtain a control message for controlling the vehicle. According to the method, the problem of dynamic balance of large model reasoning efficiency and hardware resources in the vehicle field is solved.

The present application relates to the technical field of vehicle control, and discloses a self-adaptive voice control method and device, a vehicle and a storage medium, the method comprising: obtaining first load information, the first load information being used to reflect the load level of multi-dimensional hardware in the vehicle; matching a corresponding current performance level according to the first load information, and compressing the inference ability of a large language model according to the current performance level, the inference ability being used to represent the input information amount and the output control message amount of the large language model; processing the voice instruction of a user based on the compressed large language model to obtain a control message used for controlling the vehicle. The present application solves the problem of dynamic balance between the inference efficiency of a large model and hardware resources in the field of vehicles.

This application provides a method, apparatus, terminal device, storage medium, and program product for chip operation detection. The method includes: acquiring real-time data streams of the data path according to a preset cycle during chip operation; invoking a preset comparison method to compare the real-time data streams with preset abnormal data streams to determine whether the real-time data streams and abnormal data streams are consistent; the abnormal data streams are data streams written into the monitoring unit after chip production and confirmed to cause errors; if consistent, determining a target avoidance method corresponding to the real-time data streams and invoking the target avoidance method to perform avoidance operations on the real-time data streams. This method can prevent errors caused by the chip continuing to run the real-time data streams and ensure the normal execution of the real-time data stream functions, thereby effectively detecting and avoiding errors that may occur during chip operation even after chip production, ensuring the safety and reliability of chip operation.

The application relates to an error injection method and device, computer equipment and a readable storage medium. The method comprises the following steps: determining an error injection requirement based on a configuration operation, the configuration operation being used for configuring at least one or a combination of a target logical address, an error type and a trigger condition; converting the error injection requirement into an error injection command, storing the error injection command into a content addressable memory (CAM) for scheduling, and obtaining a scheduling result; and according to the scheduling result, issuing the error injection command and a dynamic random access memory (DRAM) physical address to a corresponding DRAM control unit to execute the error injection through the DRAM control unit. Thus, the system can remain in a normal operation state during the entire error injection process, without entering a special test mode, ensuring the normal execution of a refresh operation, and the interaction between AXI internal and address mapping modules is reused, without affecting business continuity, so that the universality is enhanced.

This invention relates to the field of model parameter encryption and discloses a method, terminal device, and storage medium for protecting model weight parameters. The method for protecting model weight parameters includes: dividing the model weight parameters to be protected into multiple weight levels; performing layered encryption on the multiple weight levels to obtain layered encrypted model weights; if an inference task requirement is detected, dynamically selecting necessary parts of the model weights for partial decryption to obtain plaintext local model weight parameters, and responding to the inference task requirement based on the plaintext local model weight parameters; if the inference task requirement is met, restoring the plaintext local model weight parameters to their encrypted state. This invention improves the security of model weights throughout the entire storage and inference process.

This application provides a task execution plugin, an inference task execution method, an apparatus, a medium, and a product. The method includes: when a terminal device generates an inference task, the task execution plugin can determine a target task execution unit from among its provided candidate task execution units to implement the inference task. In the target runtime environment provided by the terminal device that generated the inference task, the target task execution unit can be used to execute the inference task. In the above method, the task execution plugin has compatibility, meaning that different task execution units among the candidate task execution units are suitable for different runtime environments provided by different terminal devices. Therefore, regardless of the runtime environment in which the inference task is generated by the terminal device, the task execution plugin can determine the target task execution unit from the candidate task execution units, and the target task execution unit will perform the inference task execution, thus ensuring the normal execution of the inference task.

Embodiments of the present application provide a satellite remote sensing method and device, electronic equipment and storage medium, comprising: receiving a user input perception detection request on the ground side, wherein the perception detection request comprises a perception service requirement and a target area; filtering a second communication satellite set from a first communication satellite set capable of covering the target area, wherein the communication satellites in the second communication satellite set have perception functions and exist idle beams not occupied by communication services; determining a satellite remote sensing strategy according to the perception service requirement and the perception parameters of the communication satellites in the second communication satellite set; and controlling the communication satellites in the second communication satellite set to utilize the idle beams to perform remote sensing detection on the target area by using the satellite remote sensing strategy to obtain a perception result. Thus, the satellite resources are fully utilized and the perception capability of the satellite system is improved.