288 results about "Function Code" patented technology

The invention provides a multi-agent control system based on a large language model. The system comprises an analogue simulation subsystem and a coordination control subsystem. The analogue simulation subsystem comprises a data acquisition module, an initial reward function generation module, an error correction module, a dense reward function generation module and a strategy network updating module, wherein the data acquisition module is used for acquiring a multi-agent reinforcement learning training code; the initial reward function generation module is used for generating an initial reward function code; the error correction module is used for generating executable reward function codes; the dense reward function generation module is used for generating dense reward function codes; and the strategy network updating module is used for obtaining the strategy network with the maximized reward. The coordination control subsystem comprises a data collection module and a strategy distribution module; wherein the data collection module is used for receiving observation data; and the strategy distribution module is used for guiding multiple agents to execute a control task by adopting a strategy network generation action for maximizing rewards based on the observation data.

The invention relates to a code file rewriting method and device based on a large language model, and the method comprises the steps: analyzing a source code file, obtaining the structure of a source code, and enabling the structure of the source code to comprise the definition of each function and the position of each function code in the source code file; respectively inputting each function code into a large language model, and respectively processing according to each cue word to complete rewriting of each function code; and combining each rewritten function code corresponding to each position in the source code file to form a rewritten code file. According to the invention, a scheme for optimizing and rewriting the codes based on the large language model and the file level can be provided.

The invention discloses a code vulnerability detection method, a code vulnerability rule configuration method, a static single assignment SSA code acquisition method, devices corresponding to the methods, and electronic equipment. The code vulnerability detection method comprises the following steps: acquiring a plurality of source code files and code vulnerability rules of an application program; generating an abstract syntax tree of the source code file, and obtaining a function call relationship; and traversing the abstract syntax tree, and obtaining a cross-function risk propagation path according to the function call relationship and the code vulnerability rule. By the adoption of the processing mode, the function calling relation and the abstract syntax tree are fused to conduct cross-function code vulnerability detection, cross-function vulnerability detection of context sensitivity, flow sensitivity and domain sensitivity is achieved, and vulnerability false report and missing report are avoided; therefore, the accuracy and recall rate of vulnerability detection can be effectively improved.

The invention discloses an agent-based test case generation method and device, equipment and a medium, and relates to the technical field of software testing, the method comprises the following steps: in response to a test case generation instruction for a target function, displaying a test case framework of the target function; in response to an interaction instruction for a preset position in the test case framework, obtaining a target test code of the target function; wherein the interaction instruction is used for triggering interaction between a first agent and a second agent, and the first agent is configured to be used for analyzing a logic structure of a function code of a target function to obtain a structure code of a test point of the target function; the second intelligent agent is configured to be used for generating a target test code based on the structure code of the test point; and displaying the target test code at a preset position to obtain a target test case of the target function. The problem that the generation accuracy of the test case is low can be solved.

The present disclosure is directed toward systems, methods, and non-transitory computer-readable media for generating function code and test code to determine that the function code satisfies an intent query. For example, the disclosed systems utilize a large language model to process an intent query and generate function code that defines one or more processes whose execution satisfies the intent query. In addition, the disclosed systems also utilize the large language model to generate test code that defines a function test for determining that the function code satisfies the intent query. Moreover, the disclosed systems can execute the test code to determine that the function code satisfies the intent query.

The invention relates to the field of electric power monitoring, in particular to an electric power monitoring system intrusion detection method and system based on flow analysis. The method comprises the following steps: collecting network traffic, analyzing and recombining to obtain structured session data; time sequence behavior features and function code distribution features are extracted to construct a multi-dimensional feature set; inputting the feature set into a compliance rule base and a behavior baseline model in parallel, and respectively outputting a rule matching result and an abnormal deviation degree score; generating a comprehensive threat index by adopting a weighted decision fusion strategy; and when the index exceeds a dynamic threshold value, intrusion is determined and an alarm is given. According to the invention, the problem of insufficient precision and adaptability caused by single feature dimension and isolated detection mechanism is solved.

The present disclosure is directed toward systems, methods, and non-transitory computer-readable media for generating function code and test code to determine that the function code satisfies an intent query. For example, the disclosed systems utilize a large language model to process an intent query and generate function code that defines one or more processes whose execution satisfies the intent query. In addition, the disclosed systems also utilize the large language model to generate test code that defines a function test for determining that the function code satisfies the intent query. Moreover, the disclosed systems can execute the test code to determine that the function code satisfies the intent query.

The invention provides a Modbus protocol security enhancement method and system in a unicast communication scene, and relates to the technical field of industrial control networks. Comprising the following steps: sharing a first ciphertext and a second ciphertext between a master node and a slave node; the master node generates a first message authentication code according to the second ciphertext, a preset key and the first random number, and sends the first message to the slave node by using the target function code; the slave node authenticates the identity of the master node according to the first message authentication code, the first ciphertext, the preset key and the second random number; when the slave node successfully authenticates the identity of the master node, the slave node generates a second message authentication code according to the first ciphertext, the preset key and the second random number, and sends a second message to the master node by using the target function code; and the master node authenticates the identity of the slave node according to the second message authentication code, the second ciphertext, the preset key and the first random number. Therefore, the validity of the identities of the two communication parties can be ensured, and the integrity and confidentiality of the data in the transmission process are guaranteed.

The invention provides a micro-isolation and differential encryption method and system based on industrial protocol perception and a storage medium, and the method comprises the steps: firstly carrying out the static filtering of an original flow through a preset port, and obtaining a to-be-analyzed first communication message; analyzing a message behavior mode by using a time sequence feature analysis model, and extracting a semantic tag containing a function code, a data point address and a data value; if the data value exceeds the safety threshold value, real-time blocking and alarming are carried out; otherwise, generating a dynamic communication strategy based on the function code and the data point address; when the micro-isolation rule is triggered, dynamically limiting the access authority of the equipment; when the difference encryption rule is triggered, an encryption algorithm is selected according to the data sensitivity, and an algorithm acceleration module is called for execution. According to the method, micro-isolation is realized through protocol semantic analysis, and illegal function code injection is prevented; a differential encryption mechanism is adopted to reduce encryption overhead, and meanwhile, the real-time requirement of an industrial scene is met; in addition, encryption strength is dynamically adjusted in combination with channel quality, and operation reliability under extreme working conditions is guaranteed.

The invention provides a multi-discontinuous register reading method for improving Modbus protocol communication rate and a function code. The multi-discontinuous register reading method comprises the following steps: defining a function code 43H in a Modbus protocol; the method comprises the following steps: determining the type, address list and number of discontinuous registers to be read through a master device, assembling a request frame, and sending the request frame to a slave device; after the slave device receives the request frame, address matching and verification are carried out on the request frame, and if the address is not matched or verification fails, the frame is discarded; if the address is matched and the verification is passed, analyzing, judging whether abnormity exists or not, and executing related operation based on a judgment result; the master device receives the response frame returned by the slave device, verifies the response frame, and analyzes a data list in the response frame to obtain data of each register if the response frame passes the verification and is a normal response frame; if verification fails or the frame is responded abnormally, error processing is carried out, and the multi-discontinuous register reading method which can improve the communication rate, reduce the communication overhead and enhance the real-time performance and is high in compatibility is provided.

The invention discloses an automatic configuration method for dispatching access of plant station equipment, and the method comprises the steps: A, obtaining communication data from the plant station equipment, and enabling the communication data to comprise a frame structure field; b, based on the frame structure field, identifying a communication protocol type by comparing an initial identifier, a function code field and a frame length field; c, in combination with the communication protocol type and the equipment attribute field, an access configuration strategy template is matched, and the template comprises a communication parameter set, an interface adaptation rule and a signal mapping relation; d, generating a structured configuration data file according to the configuration strategy template; and E, deploying the configuration data file to a scheduling system to complete equipment communication registration and data interaction initialization.

The invention discloses a GUI (Graphical User Interface) program fuzz test method and system, computer equipment and a storage medium, and the method comprises the steps: firstly designing a hyper-call interface of a KVM (Keyboard Virtual Machine), and then designing a QEMU (Query Empirical Mode Unit) monitor for responding to a call number and a corresponding function code of the hyper-call interface of the KVM; the method comprises the following steps: writing a code library in which a call number and a corresponding function code are packaged for external call, injecting the code library into a target GUI program, creating a Windows agent layer under a Windows system, finishing a fuzzy test on the target GUI program through the Windows agent layer, creating a Linux agent layer under a Linux system, and finishing a fuzzy test on the target GUI program through the Linux agent layer. According to the method, the fuzzy test range of the GUI program can be expanded, and the fuzzy test speed of the GUI program is increased.

The invention belongs to the technical field of intrusion detection, and relates to an intrusion detection method, system and device for a roadside heterogeneous device network, and the method comprises the steps: calculating an action feature value for describing the change condition of a function code in front and back frames of data according to the function code in each frame of data, and obtaining an action feature vector; calculating each function code in each frame of data to read a time interval of a previous register with the same serial number to obtain a time interval vector; constructing a network behavior vector through the action feature vector and the time interval vector; clustering network behavior vectors constructed by normal network data through an anomaly detection model based on a clustering algorithm to obtain a central point and a threshold distance of each cluster; and judging whether the data points are abnormal data or not according to whether the network behavior vectors of the data points in the network data meet the characteristics of the outliers or not. The zero-day attack detection method is small in computing power requirement and can effectively detect zero-day attack at a low false alarm rate.

The present application discloses a method for encryption of obfuscation code, an apparatus, a device and a readable medium. The method for encryption of obfuscation code includes: obtaining position information of all of function codes in a to-be-encrypted program; based on the position information, determining, one by one, whether a skipping instruction exists in the corresponding function code; when a skipping instruction exists in one of the function codes, saving the skipping instruction and corresponding address information into a decryption datum body, and replacing the skipping instruction with a breakpoint instruction; and in response to execution of the preceding one step being completed for all of the function codes, based on the decryption datum body and an encrypted program obtained by replacing the instructions, generating an encrypted file.

The present disclosure is directed toward systems, methods, and non-transitory computer-readable media for generating function code and test code to determine that the function code satisfies an intent query. For example, the disclosed systems utilize a large language model to process an intent query and generate function code that defines one or more processes whose execution satisfies the intent query. In addition, the disclosed systems also utilize the large language model to generate test code that defines a function test for determining that the function code satisfies the intent query. Moreover, the disclosed systems can execute the test code to determine that the function code satisfies the intent query.

The invention provides a method for developing a tool function for an intelligent agent through computing power of an intelligent computing center, and relates to the technical field of intelligent computing centers, intelligent computing centers and computing power infrastructures, and the method comprises the following steps: S1, a manufacturing intelligent agent receives task information provided by a user through an interactive interface and used for developing the tool function for a target intelligent agent; s2, the manufacturing agent determines development requirements of the tool function according to the task information, and the development requirements at least comprise function requirements of the tool function; and S3, the manufacturing agent automatically generates a code of the tool function of the target agent according to the development requirement of the tool function. In the invention, through the manufacturing agent running in the intelligent computing center, the process of communicating with the user, confirming the development demand, automatically completing the development of the tool function of the target agent and replacing the manual development of the tool function code can be realized, so that the development cost is effectively saved, the intelligent computing center can provide sufficient computing power resources, and the development efficiency is improved. And the development efficiency is greatly improved.

The invention relates to the field of network card firmware repairing and upgrading, in particular to a network card firmware hot patching method, a storage medium and electronic equipment. According to the method, a patch file is received and loaded to a reserved area of a network card SRAM (Static Random Access Memory), an address acquisition logic is executed to generate a patch symbol table, a function to be updated is determined by matching a function jump table, address pointing information is updated to point to a target function code in the SRAM, and indirect calling based on the jump table is realized. According to the method, memory resource waste is solved through function granularity updating, address accuracy is ensured through dynamic symbol table matching, non-inductive switching of services is achieved through an indirect calling mechanism, equipment does not need to be restarted, and the continuity and reliability of network services are improved.

The invention discloses a path task planning method for a magnetic micro robot in a complex environment, and relates to the technical field of micro-nano control, and the method comprises the steps: carrying out the formalized coding of a multi-target task demand through employing linear sequential logic LTL; a decision module based on a DQN neural network is constructed, a real-time environment state and an action execution result are combined and input to a multi-level reward function coded by LTL to generate a reward and punishment signal, and the reward and punishment signal is fed back to the network to guide a path optimization decision; in the network optimization process, a logic correction state transition model is introduced to increase the probability of compliance action in a weighted manner, so that the state transition process of the magnetic micro robot is shifted towards the direction meeting LTL constraint. According to the method, task logic features are extracted through bidirectional GRU, and the task logic features are embedded into the neural network as priori knowledge of decision making, so that the problem of low efficiency of processing time sequence tasks through traditional DQN is solved, and the interpretability and safety of decision making are improved through explicit logic constraints.

The embodiment of the invention provides a data collection method and system, electronic equipment and a storage medium, and is applied to the technical field of industrial communication. According to the data acquisition method and system, the electronic equipment and the storage medium provided by the embodiment of the invention, the register objects with the same acquisition period, the same slave station ID, the same function code and continuous register addresses are aggregated in the same second register aggregation group, and the data request frame is generated for the same second register aggregation group; the register objects in the second register aggregation group can share one data request frame. Compared with the prior art in which each register object generates an independent data request frame, the data acquisition method and device have the advantages that the number of the data request frames can be obviously reduced, so that excessive system resources are prevented from being occupied, and the data acquisition efficiency is further improved.

The invention provides a device access adaptation code generation method and system, a device and a medium, and the code generation method comprises the steps: inputting the behavior of a target device based on a declarative description language, and obtaining a first input result; on the basis of a declarative description language, performing complementation description input on the design adapted to the target equipment to obtain a second input result; deducing a program flow chart of a target function of the to-be-implemented code according to the first input result and the second input result; and generating an implementation code of the target function according to the program flow chart of the target function and a preset rule. According to the method, equipment behaviors are modeled through the declarative description language, and when equipment of the same kind is accessed but equipment of different manufacturers is accessed, input information can be repeatedly utilized, so that repeated labor is avoided; and meanwhile, a derivation algorithm is introduced to solve a program flow chart realized by target function codes, the program flow chart can be used as input of code generation and can also be used as a carrier for realizing logic through manual inspection, and the code inspection efficiency is improved.

The invention relates to the technical field of hardware acceleration, and provides a convolutional neural network hardware acceleration method and system, and the method comprises the steps: obtaining a convolutional neural network instruction which comprises an img2col instruction, a systolic array multiplication instruction and a col2img instruction; analyzing an operation instruction code and a function code field of each instruction, determining a target register, a source register and calculation parameters, determining an instruction function and generating a control signal; data to be operated are taken out from the source register and transmitted to the corresponding operation module together with the control signal and the calculation parameter to be calculated, and then the data are written back to the target register; through hardware optimization strategies such as multi-level address superposition, systolic array data multiplexing, double-buffer window construction and a parallel comparison tree, extra power consumption caused by multiple times of data loading and storage and intermediate result calculation is reduced, and data handling overhead, calculation delay and control complexity are greatly reduced.

The invention relates to the technical field of computer software, and discloses a bridging method and system for a cross-language interface of KaihongOS, a terminal and a storage medium. The method comprises the steps that an interface definition file is analyzed, and function declaration information of a target interface is obtained; analyzing a native interface definition file to obtain implementation information of a native interface corresponding to the target interface; and based on the function declaration information and the implementation information, according to a preset bridging rule, automatically generating a bridging function code conforming to a target operation environment, and bridging a cross-language interface through the bridging function code. According to the method, through automatic analysis and code generation, the bridging function code meeting the target operation environment is automatically generated by improving the code quality, and repeated labor and potential errors in manual writing are avoided. A preset bridging rule ensures the consistency and maintainability of codes, the technical threshold can be effectively reduced, and the project cycle is shortened.

The invention provides a compiler rear-end generation enhancement method and device based on a large language model, and the method comprises the steps: converting a target description file at the rear end of a compiler into a structured attribute graph, training a graph neural network, enabling the graph neural network to learn the similarity between the rear ends of the existing compiler, and obtaining the target description file; generating a graph embedding vector representing the structural characteristics of each existing compiler for the rear end of each existing compiler; training the multi-modal model according to the graph embedding vector and the text representation at the rear end of the existing compiler, so as to predict the similarity between the functions at the rear end of the compiler; inputting the text representation of the target function at the rear end of the new compiler and the graph embedding vector at the rear end of the new compiler into the trained multi-modal model to obtain a similarity score between the target function at the rear end of the new compiler and the function at the rear end of the existing compiler; screening out the most similar existing function as a high-quality example; and according to the specification description of the high-quality example and the target function, an example prompt is constructed, and the large language model generates a target function code at the rear end of a new compiler.

The invention belongs to the technical field of code abstract generation, and particularly relates to a code abstract generation method and device.The code abstract generation method comprises the following specific steps that S1, function source codes and corresponding code abstracts are preprocessed, and a multi-modal function code abstract data set is obtained; s3, designing a two-stage fine tuning method oriented to a code abstract generation task, taking X0 and E as input, performing fine tuning on an original CodePTMs model to obtain a fine tuning prefix vector code abstract model CodePTMs' model, and outputting a code abstract; and S4, dividing the multimodal function code summary data set, and training the CodePTMs' model by using the multimodal function code summary data set. Through the two-stage fine tuning method oriented to the code abstract generation task, the deep semantic modeling capability of the model for the code can be improved, so that the accuracy of generating the code abstract by the model is improved.

The invention discloses a dynamic monitoring platform of a pollution source processing system, which relates to the technical field of pollution source processing, and comprises the following steps: the dynamic monitoring platform loads a preset protocol template through a device ID, dynamically adjusts legality communication parameters, locks a device type function code, and generates an initialization verification report; generating a dynamic polling task based on the initialization report, collecting equipment data in parallel through a thread pool, and executing validity mark judgment to form a standardized data packet; an analysis engine extracts plaintext parameters in the standardized data packet, converts a state code for service description and marks data quality based on a rule base, and generates a structured state analysis report; and performing multi-level anomaly detection on the structured state analysis report, triggering a hierarchical warning in combination with a logic association rule, accurately positioning a risk and triggering control strategy matching. According to the method, multi-dimensional judgment of data quality is realized through three-stage verification (frame structure verification, data type forced conversion and logic integrity verification) and a state code linkage mechanism.

An embodiment of the present invention provides a method, system, and storage medium for constructing a calculator for amount calculation. The method includes: before creating a calculator that implements amount calculation, obtaining a corresponding function code based on the amount calculation information of the functional dimension in a corresponding dimension system according to the amount calculation type, loading an amount calculation rule corresponding to the function code, and constructing a functional dimension branch of the calculator based on the amount calculation rule corresponding to the function code; obtaining an operation code corresponding to the operation dimension in the dimension system corresponding to the amount calculation type, and constructing an operation dimension branch of the calculator based on the operation code corresponding to the operation dimension; and constructing a calculator, the calculator including the functional dimension branch and the operation dimension branch, the functional dimension branch and the operation dimension branch being in a parallel relationship. This improves the scalability of the code, reduces the maintenance cost of the code, and improves the debuggability and reusability of the code.

The embodiment of the invention provides an intelligent agent system for graphics processor development and a graphics processor development method, and is applied to the technical field of graphics processor development. The intelligent agent system comprises a first intelligent agent, a second intelligent agent, a plurality of third intelligent agents and a fourth intelligent agent. The first intelligent agent is configured to receive a development task and drive the other intelligent agents to execute corresponding operations in sequence or in parallel so as to complete the development task; the second agent is configured to maintain related documents; the plurality of third agents are configured to generate function codes based on related documents; and the fourth agent is configured to perform joint test on the function codes to generate test information. The plurality of third agents includes a simulator agent, an assembler agent, and a test program agent. According to the intelligent agent system, related documents serve as a single fact source, a simulator, an assembler and a test program are automatically processed through multi-intelligent-agent cooperation, and the consistency of all components consuming the related documents is guaranteed.

The invention relates to the technical field of dust explosion prevention in the textile industry, in particular to a dust explosion prevention method and system in the textile industry and a storage medium. The explosion-proof method comprises the steps that a bus communication architecture based on a two-wire system power line carrier is constructed, a multi-host communication mode adopting a private protocol is established between a controller and a plurality of detection probes, and a message structure of the private protocol comprises a function code field, a number field, a data field and a CRC verification field, wherein the function code field definition comprises 01H alarm coding, 02H parameter modification coding and 03H heartbeat request coding. According to the invention, a multi-host communication architecture based on a private protocol is constructed, dynamic distribution of bus control right and dust flow feature detection are deeply coupled, and a zero-latency master-slave switching mechanism when an alarm event is triggered is realized.

The present invention discloses a method for automatically compiling and running C / C++ code for Huawei Ascend acceleration cards. By virtue of the characteristics of the Ascend platform provided by the Ascend acceleration card processor to the upper layer and its custom operators, combined with C / C++ language compilation, the overall process of operator function development and invocation is integrated with the data management and operation scheduling capabilities of the host and the Ascend acceleration card processor, so as to implement the execution of custom function code on the Ascend acceleration card; among them, it includes a unified description model for custom operator functions, a data model for memory management of the host and the Ascend acceleration card processor, an automated configuration program for custom operators of the Ascend platform, and a call execution system for custom operators.

The application discloses a reliability evaluation and optimization method and system for large model function code, a medium and a program product, and belongs to the technical field of artificial intelligence. A demand prompt word, a test case and expected output are acquired to form a standardized test data set; an execution consistency score, a semantic consistency index and a static structure consistency index of the code to be evaluated are calculated and fused into an intention consistency comprehensive score; a safety score is obtained by performing static safety analysis on the code to be evaluated; the intention consistency comprehensive score and the safety score are fused into a final comprehensive score, and the reliability is judged; if the reliability is not good, a defect label is generated according to the scores of the indexes and weighted fusion is performed to obtain optimization guidance information, and the code is corrected or regenerated according to the information. The application comprehensively evaluates the code quality from four dimensions of execution, semantics, structure and safety, highlights the short board effect by adopting a fusion strategy, realizes closed-loop automatic optimization, and significantly improves the reliability and safety of the function code.