379 results about "Rapid convergence" patented technology

The invention provides a power grid evolution behavior modeling method and system based on dynamic digital twinning, and relates to the technical field of intelligent power grid digital twinning. According to the method, interaction characteristics of power station output, user load response and an energy storage scheduling strategy are captured in real time through a digital twin interface, behavior coupling factors containing complementarity and conflict weight are formed, and a power grid response relation and a multi-scale demand elastic curved surface are constructed by using an implicit tensor fusion technology and nonlinear projection. And through the steps of virtual behavior anchor point disturbance and the like, offset is separated to identify an unbalance region, an evolution trajectory is generated, an optimal collaborative path is extracted, and the weight of a behavior coupling factor is dynamically adjusted, so that rapid convergence of a global collaborative stability domain is realized. The method covers a behavior coupling factor generation module and the like, supports various operations, effectively solves many problems in power grid resource interaction, and improves dynamic adaptability and cooperation efficiency of power grid evolution modeling.

The invention discloses an urban power grid real-time load collaborative peak regulation method based on multi-energy complementation and AI scheduling. The method comprises the following steps of multi-source data access and high-dimensional feature space construction, embedded entropy calculation and interactive network construction, domain knowledge and data-driven model fusion, hierarchical scheduling and dual-stage optimization, and real-time decision and closed-loop feedback. According to the method, the real-time performance and hierarchical scheduling thought are emphasized, and an organic closed loop is formed on the three aspects of intra-day scheduling, hour-level rolling correction and minute-level or second-level emergency response. Meanwhile, by means of a multi-stage optimizer switching mechanism, the model can complete rapid convergence of high-dimensional parameters in a short time, finer strategy fine adjustment is carried out in the later period, and the accuracy and reliability of a peak regulation scheme are guaranteed; the method can be applied to advanced power grid systems such as intelligent power grid dispatching, a multi-energy collaborative optimization platform and demand side response management, and has the characteristics of high real-time performance, strong adaptability and good expandability.

The invention discloses a self-adaptive logarithmic terminal sliding mode control method and device based on observation, and relates to the technical field of sliding mode control. According to the method, the terminal attractors are nested into the natural logarithmic function, terminal sliding mode control and logarithmic sliding mode control are effectively combined, rapid convergence can be achieved near and far away from the balance point, the global dynamics of the sliding mode surface is improved, and the sliding mode control efficiency is improved. Meanwhile, a self-adaptive logarithmic terminal sliding mode controller is constructed on the basis of super-spiral sliding mode control and self-adaptive gain, the self-adaptive gain is updated according to a self-adaptive law, the self-adaptive law is constructed on the basis of a preset nonlinear obstacle function, intelligent self-adaption of the super-spiral sliding mode control gain is achieved, and therefore sliding mode variables are restrained within a preset boundary; and the interference estimation error is quickly converged based on the estimation of the observer on the interference, the global quick convergence of the state tracking error is realized under the constraint of the sliding mode variable, the steady-state precision is high, the chattering phenomenon is weakened, and the convergence efficiency and the control precision are improved.

The invention discloses a finite iteration error tracking learning control method for robot accurate trajectory tracking. The method comprises the steps of establishing a kinetic equation of a repeated operation robot system, determining a discrete dynamic model of the robot system, designing a finite iteration error tracking learning controller and carrying out simulation verification. According to the method, under the conditions that the initial position of the robot system is arbitrary and the expected trajectory changes, the number of iterations required for achieving arbitrary precision tracking of the given expected error trajectory by the output error of the robot system can be calculated by designing an error tracking learning controller and only utilizing error information obtained through first iteration. According to the method, the limitation that zero error convergence needs to depend on infinite iterations in a traditional iterative learning control method is effectively overcome; meanwhile, it is ensured that the robot system can still achieve the preset tracking precision with the rapid convergence capacity under the disturbance conditions that the initial position is arbitrary and the expected trajectory changes, and the practicability and the popularization value of iterative learning control in actual engineering are improved.

A motor control method based on an improved composite reaching law and a sliding mode observer belongs to the technical field of permanent magnet synchronous motor system control, and comprises the following steps: firstly, considering motor internal parameter change and external load disturbance, and establishing a disturbance-containing permanent magnet synchronous motor mathematical model and a motor state equation; secondly, providing a new switching function, constructing an improved composite reaching law, completing stability proving, and analyzing rapid convergence and buffeting suppression capability of the improved composite reaching law; then constructing an anti-saturation integral sliding mode surface, and designing a sliding mode speed controller according to the permanent magnet synchronous motor mathematical model containing disturbance, the anti-saturation integral sliding mode surface and the improved composite reaching law; and finally, aiming at disturbance in the controller, designing an improved sliding mode disturbance observer, and completing stability proving. The contradiction between fast convergence and buffeting suppression in an existing sliding mode reaching law scheme is solved, and the provided improved sliding mode observer improves the observation precision and the response speed.

The invention relates to the technical field of adaptive filtering, discloses a variable-step robust adaptive filter and a filter network, and aims to solve the technical problem that a traditional adaptive filter cannot effectively consider convergence speed, steady-state precision and impulse noise interference resistance. A non-linear error factor is introduced, large errors caused by impulse noise are effectively suppressed through a generalized function form of the non-linear error factor, and the robustness of the filtering process is remarkably improved; meanwhile, a variable step length mechanism is provided, the theoretically optimal candidate step length is calculated on line by minimizing the mean square deviation of the next moment, and a self-adaptive target variable step length is generated through truncation and time smoothing processing, so that dynamic balance is realized between rapid convergence and low-steady-state maladjustment. According to the invention, the filter is expanded to the distributed network, each node follows a diffusion strategy of first updating and then combining, and global information sharing and performance collaborative optimization are realized by using a combined coefficient.

The invention relates to the technical field of path planning, and discloses an AGV path planning method and system based on an improved ant colony algorithm, and the method comprises the steps: obtaining a position coordinate and a scheduling end point of an automatic guided vehicle; carrying out map modeling on the schedulable area; constructing the actual elements of the path planning into algorithm elements of an ant colony algorithm; and performing path optimization by using an improved ant colony algorithm to obtain an optimal planning path. The robustness of the algorithm is enhanced, rapid convergence of the algorithm is enhanced, local optimum is avoided, and the path planning quality is improved. The algorithm output randomness is avoided, global repeated operation is introduced to improve the ant colony algorithm mechanism, the global optimality of path planning is improved, and the efficiency and quality of AGV path planning and the adaptability of path planning in different environments are improved.

The invention relates to the technical field of optical communication, and discloses an optical module adaptive test and rapid convergence method based on reinforcement learning, which can dynamically generate and execute an optimal test sequence for optical modules of different types or states through an off-line trained RL intelligent agent module, avoids redundant steps in a fixed process, and improves the test efficiency. On the premise of ensuring the test coverage, the test time is obviously shortened, and meanwhile, the detection probability of potential defects is improved; through the closed-loop test of the integrated RL intelligent agent module, the test action can be adjusted in real time according to the state of a single optical module, so that the test process can quickly adapt to the individual difference and process fluctuation of the device, the online quick convergence of the test strategy is realized, and the accuracy and stability of the test are guaranteed; and through online iterative optimization of the RL agent module, new data and new changes in the production process can be continuously learned, so that a test strategy is continuously evolved, and the optimal performance is kept for a long time.

The invention relates to the technical field of wireless communication, and discloses an unloading method in an ultra-dense millimeter wave MEC network. Aiming at the problems of insufficient terminal capability, high energy consumption of an ultra-dense base station, millimeter wave coverage limitation, communication security risk and the like under the condition of sharp increase of a calculation-intensive task, the method comprises the following steps: firstly, obtaining network basic information, constructing a network architecture containing communication, calculation unloading and a security model, and establishing a multi-constraint optimization problem; based on the optimization problem, initializing a multi-strategy black-wing plinuary optimization algorithm population through logic mapping and elite selection; updating individual positions through global and local search in algorithm iteration, and screening historical optimal individuals; and finally, configuring and unloading resources according to the configuration. According to the method, the NOMA technology, a multi-step unloading framework and a hybrid communication mode are combined, the optimization efficiency is improved through a multi-strategy black-wing optimization algorithm, the local energy consumption is reduced, the communication rate is improved, the safety is guaranteed, the optimal scheme meeting time delay and safety constraints is rapidly converged, and the user experience is improved.

The invention relates to a rigid spacecraft attitude control method and system based on finite time constraint, and belongs to the technical field of spacecraft attitude control. Comprising the following steps: respectively constructing a spacecraft attitude control system model and a finite time convergence performance function containing a preset time constant; according to the finite time convergence performance function, dynamic attitude constraint is applied to the system state of the spacecraft attitude control system model; converting the attitude constraint into differential homeomorphic mapping of an unconstrained space; obtaining an error conversion variable according to differential homeomorphic mapping; and designing a virtual control law and constructing an error compensation module according to the error conversion variable, and designing an actual control input law according to the virtual control law and the error compensation module. According to the method, rapid convergence of the spacecraft attitude within the preset time is ensured, the dynamic response precision of the spacecraft in the rapid attitude maneuver process is remarkably improved, the control period delay is effectively avoided, and the power consumption is reduced.

The invention relates to the technical field of large model optimization, and discloses a large model parameter optimization and adaptive adjustment method and device based on reinforcement learning, and the method comprises the steps: obtaining a current operation state of a target large model, and carrying out the comprehensive evaluation of the operation state of the target large model through combining with data features, and obtaining an operation state vector; constructing an intelligent agent, and inputting the operation state vector into the intelligent agent to obtain a parameter optimization strategy; optimizing parameters of the target large model based on a parameter optimization strategy to obtain an optimized large model; calculating a plurality of preset indexes for optimizing the large model by utilizing a reward function to serve as reward results; and adaptively adjusting parameters of the intelligent agent based on the reward result. In training and practical application of the large model, large model parameters are dynamically adjusted by means of an intelligent agent, it is ensured that rapid convergence can be achieved in the initial stage of training, the global optimal solution can be accurately approached in the later stage of training, the training efficiency and quality are effectively improved, and by means of a reward mechanism, the intelligent agent is made to adjust the parameters in a self-adaptive mode, and the generalization ability of the large model is improved.

The invention belongs to the technical field of intelligent manufacturing and production scheduling, discloses a flexible job shop dynamic scheduling method and system considering machine aging, and designs a hierarchical environmental response strategy which can firstly evaluate the severity of environmental change. When the change is not violent, only a lightweight local optimization strategy is adopted for fine adjustment; and when the change is relatively violent, a global reconstruction strategy combining knowledge migration, reinitialization and directional repair is started. Therefore, the algorithm can intelligently allocate computing resources according to the intensity of environment change, blind global search is avoided, and the response speed and the operation efficiency of the algorithm are greatly improved. The design effectively balances the exploration and utilization capabilities of the algorithm, and maintains the diversity of the population while ensuring rapid convergence, thereby obtaining a group of Pareto optimal solution sets with good convergence and wider distribution.

The invention relates to the technical field of power distribution network control, and particularly discloses a power distribution network distributed cooperative control system, which comprises a sensing acquisition module, a state judgment module, a communication interaction module, a main node setting module, a consensus control module, a control execution module and an exception handling module. According to the scheme, a dynamic main node selection method is adopted, the main nodes are dynamically arranged in each control period, the multiple main nodes can guide different local sub-graphs respectively, partition cooperative adjustment is achieved, the efficient response capacity and robustness of the system are improved, and rapid convergence of the system is facilitated through target-driven reactive power adjustment; state convergence among PV nodes is achieved through a consensus control method, it is guaranteed that all PV nodes output reactive power in equal proportion according to the capacities of the PV nodes through node identification and command transmission, overload or idle of some PV nodes is avoided, system-level reactive synchronous control is achieved, the self-adaptive capacity of the system is effectively improved, and network topology changes are flexibly coped with.

The invention discloses a whole network reference unified Beidou B2b constant deviation estimation method, which comprises the following steps of: firstly, performing Beidou B2b clock error reference transformation by applying Beidou B2b service DCB information to realize constant deviation parameterization separation; secondly, adding a single-station virtual reference, and constructing a constant deviation full-rank estimation model based on a single station; then, the constant deviation information of the whole network is integrated, the reference difference of a plurality of single station solutions is eliminated, and B2b constant deviation estimation with the unified reference of the whole network is achieved; and finally, under the constant deviation constraint of the unified reference, the performance of terminal positioning convergence speed, precision and the like is improved. According to the method, constant deviation estimation based on the whole network is realized, the stability and availability of Beidou B2b constant deviation estimation are improved, the problems of to-be-estimated parameter hopping and even filtering divergence possibly caused when a user side adopts constant deviations of different references are avoided, and then rapid convergence and instantaneous horizontal decimeter-level positioning of a terminal user are guaranteed.

The invention discloses an exoskeleton robot fixed time control method based on output constraint and disturbance observation, relates to the field of robot control, constructs a fixed time neural controller (FTNC), and ensures that a system is stable in fixed time by combining a universal obstacle Lyapunov function (UBLF) and a preset performance function (PPF), so as to improve the system stability. The convergence time is irrelevant to the initial state; meanwhile, a radial basis function neural network (RBFNN) and a nonlinear disturbance observer (NDO) are adopted to jointly compensate the uncertainty of the model and the man-machine interaction disturbance; by dynamically adjusting UBLF boundary conditions, joint angle errors are limited, and training safety is guaranteed. According to the method, a fixed time control theory is adopted, fixed time is combined with a preset performance function (PPF) and a universal barrier Lyapunov function (UBLF), and the output of the exoskeleton robot is strictly limited, so that the rapid convergence and transient performance requirements of an exoskeleton system are met at the same time.

The invention discloses an aspheric surface processing intelligent feedback control system based on digital twinning, and the system comprises a digital twinning modeling and baseline calibration module which is used for building an aspheric surface processing digital twinning body and generating a baseline; the twinborn update and virtual-real residual calculation module is used for calculating virtual-real residual update digital twinborn bodies; the reinforcement learning candidate strategy generation module is used for generating reinforcement learning candidate strategies through the reinforcement learning strategy network; the model predictive control optimization module is used for establishing a model predictive control optimization problem and performing rolling prediction; the double-strategy fusion control module is used for executing virtual processing simulation and generating a control vector; and the virtual-real closed-loop adaptive convergence execution module is used for executing aspheric surface processing to realize closed-loop adaptive convergence control. According to the method, digital twinning is combined with reinforcement learning and model prediction control, intelligent optimization of aspheric surface machining is achieved, and the method has the advantages of being high in precision and stability and rapid in convergence.

The invention discloses a sub-microsecond simulation-based network construction type multi-converter modeling method, and relates to the technical field of converter modeling and simulation. Discretizing a switch element, and constructing a generalized constant admittance switch model based on an LC equivalent circuit; then, expansion modeling from a single converter to a multi-converter system is realized through a discrete system state matrix, and a complete network-forming type multi-converter model including a control link is established in combination with inertia characteristics of a virtual synchronous generator; and finally, building a simulation model of the network construction type multi-converter system in the PSCAD / EMTDC, and verifying the correctness and effectiveness of the method. According to the method, the generalized constant admittance switch model and the inertia characteristic of the virtual synchronous generator are fused, good applicability and stability are achieved under the sub-microsecond-level simulation step length, the system simulation precision and efficiency are remarkably improved, and meanwhile rapid convergence of the transient process of the network construction type multi-converter and active supporting of the power grid voltage frequency are achieved.

The invention discloses a micro-grid cooperative scheduling control method, system, equipment and medium based on edge calculation, and relates to the technical field of computer platform load balancing, and the method comprises the steps: collecting power grid operation data through a micro-grid unit, carrying out the short-time power prediction of the micro-grid unit according to the power grid operation data, and obtaining a load prediction value; deploying a genetic algorithm at an edge node, and optimizing a load prediction value to obtain a scheduling strategy parameter; the scheduling strategy parameters are input into the micro-grid units, a game model is constructed to carry out multi-micro-grid cooperative game, and optimal scheduling strategy parameters are output; and performing power output scheduling through a multi-stage control mechanism. According to the method, through multi-dimensional data acquisition and sequence feature learning of micro-grid units, high-precision short-time power prediction is realized, a rapid convergence local optimization strategy is obtained, it is ensured that multiple micro-grids reach Nash equilibrium under autonomous conditions, and rapid response to sudden disturbance and effective correction of steady-state errors are realized.

The invention relates to the technical field of image recognition, and particularly provides a multi-view point cloud registration method, which comprises the following steps of: performing coarse registration on a source point cloud and a target point cloud based on multi-dimensional features by using an RANSAC (Random Sample Consensus) algorithm in a coarse registration stage to obtain a coarse registration transformation matrix; in the fine registration stage, the coarse registration transformation matrix is used as an initial value, point cloud registration is carried out in at least two resolution spaces, segmented iterative optimization is carried out by using an error loss function set in each resolution space, rapid convergence is carried out in a low-resolution space through an iterative nearest point algorithm, and the point cloud registration is realized. And local geometric alignment optimization is carried out in other resolution spaces through a generalized iterative nearest point algorithm, and fine registration of the source point cloud and the target point cloud is completed after multi-resolution space progressive optimization registration. According to the method, the robustness of feature matching of the low-overlap region is remarkably improved, the registration speed and precision are balanced, and the limitation of a traditional point cloud registration method under the low-overlap and non-ideal point cloud condition is solved.

The invention discloses an unmanned aerial vehicle flight path planning method based on a hybrid WOA-CEM-DRL algorithm, and the method is based on a hybrid WOA (whale optimization algorithm), a cross entropy method (CEM) and deep reinforcement learning (DRL), and aims at solving the problems that an existing algorithm is slow in convergence speed and is liable to fall into a local optimal solution in a complex three-dimensional environment. The invention provides a three-stage dynamic coevolution framework. In the early stage of evolution, population diversity is enhanced through global disturbance of CEM; in the middle stage of evolution, exploration and development are balanced; and in the later stage of evolution, performing fine tuning on the elite solution by using gradient information of the DRL. According to the method, global search of WOA, efficient sampling of CEM and rapid convergence capability of DRL are fused, and the method is mainly used for accurately generating a safe and smooth optimal track for an unmanned aerial vehicle in a complex three-dimensional environment.

The invention discloses an electromagnetic metasurface unit structure optimization method based on constraint space mapping, and belongs to the technical field of electromagnetic metamaterial design. In order to solve the problems that in the prior art, constraint conditions are difficult to process in the metasurface unit structure optimization process, and optimization efficiency is low, the method mainly adopts the mode that multi-dimensional structure variables are mapped to a convex constraint space formed by inequality constraint conditions, and efficient optimization is carried out in combination with a multi-objective optimization algorithm. A structural variable vector is represented by using a direction vector and a proportionality coefficient, so that an initial solution generation process meeting complex constraints is simplified, and meanwhile, the global search capability of the algorithm and the convergence speed in a constraint space are improved. According to the method, the local optimal solution can be quickly converged under the complex constraint condition, and the Pareto optimal solution is effectively found, so that an efficient and accurate solution is provided for multi-objective optimization of the electromagnetic metasurface unit structure.

The invention discloses a robot gait test optimization method and device, equipment and a storage medium. According to the method, interval expansion is carried out on the historical gait parameters, and new virtual gait parameters are randomly generated, so that the limitation of a fixed parameter interval in a traditional gait optimization process can be broken through, and the diversity and environmental adaptability of gait parameter sampling are improved; the friction coefficient in the current test environment is determined, the friction coefficient and the virtual gait parameters are jointly input into the simulation model, and an obtained simulation test result evaluates a weighted fusion objective function composed of stability, energy efficiency and anthropomorphism, so that dynamic tradeoff and collaborative optimization are realized among multiple performance indexes subsequently. According to an error between a simulation test result and an actual test result, a value interval of a virtual gait parameter, a friction coefficient learning rate and a target function weight coefficient are dynamically adjusted, finally, rapid convergence of simulation and actual test results can be realized, and the iteration efficiency of a gait optimization process and the comprehensive performance of a final gait are improved.

The invention relates to the technical field of intelligent control, and discloses a multi-agent consistency detection system, which effectively improves the cooperative control capability of a multi-agent system in a dynamic environment. A real-time topology updating mechanism guarantees rapid convergence when a network structure changes, a prediction compensation module significantly reduces control delay caused by environmental interference, and a multi-dimensional health monitoring system improves the accuracy and efficiency of fault recovery. Functional decoupling is achieved through the modular design, system expansion and maintenance are facilitated, autonomous adjustment and optimization of a recovery strategy are achieved through a strategy optimization mechanism driven by reinforcement learning, the fault recovery time is shortened, and the success rate is increased. A secondary health verification mechanism ensures effectiveness of recovery operation, and secondary faults caused by continuous operation of the system in an incomplete recovery state are avoided.

The invention discloses a business processing method and device based on an ant colony algorithm, electronic equipment and a medium. The method comprises the following steps: acquiring a business service request, and analyzing the business service request to obtain a business task; determining each agent matched with the business task; wherein the intelligent agents are used for cooperative processing of business tasks; planning the cooperation path of each agent based on an ant colony algorithm to obtain a target path; and according to the cooperation sequence of the intelligent agents in the target path, executing the business task by using the intelligent agents in sequence. According to the technical scheme, efficient collaboration between agents is achieved through the improved ant colony algorithm. According to the system, a multi-dimensional pheromone matrix is adopted to record and transmit collaborative experience among departments, a knowledge verification mechanism based on swarm intelligence is established, and dynamic optimization and rapid convergence of a cross-department business process are realized.

Provided are techniques for the calibration of distillation learning from a teacher model to a student model. Specifically, the present disclosure proposes systems and methods that provide convergence with both high quality and speed. That is, example proposed systems both enable the distillation loss to be minimized at the probability mean value in the probability domain of the teacher's predictions distributions while also providing a loss that is nicely (e.g., symmetrically and / or strongly) convex around an optimum in the logit and / or probability domains (e.g., including far from the minimum) to encourage fast convergence of gradient based methods (e.g., irrespective of distance from the minimum).

The invention provides a self-adaptive fixed time tracking control method based on event triggering, which is used for solving the technical problem that a virtual control law is easy to generate singularity in the traditional backstepping method design. The method comprises the following steps: establishing a mathematical model of a non-strict feedback nonlinear system with input time delay, and introducing an auxiliary system to counteract delay influence; an error dynamic system is obtained through coordinate transformation; according to system fixed time convergence, designing a piecewise function, introducing a backstepping method to design a controller and an adaptive law, using an RBFNN to approach an unknown function therein, designing a virtual control signal corresponding to each error signal, and obtaining an actual controller and an adaptive law; according to the actual self-adaptive controller, event triggering conditions are designed, when the event triggering moment is reached, the event triggering conditions containing input are triggered, and information transmission and response are carried out. An event triggering mechanism is designed, the number of times of control instruction transmission is reduced, communication resources are saved, and rapid convergence and safe operation of the system state are achieved.

According to the communication method and the communication device based on the LDPC code, an area B of an LDPC basis matrix used for encoding or decoding comprises a first square matrix, the first square matrix comprises at least two columns with the column weight larger than or equal to 3 and / or at least one column with the column weight being an even number larger than or equal to 4, and the row weight of each row of the first square matrix is larger than or equal to 2. Thus, the column weight of the columns included in the first square matrix, namely the B area, is not limited to 2 or 3 and can be a numerical value except 2 and 3. In addition, the first square matrix comprises at least two columns with the column weight larger than or equal to 3 and / or at least one column with the column weight being an even number larger than or equal to 4, and the connecting edge density of the first square matrix is higher. Therefore, the degree distribution of the coding structure of the LDPC basis matrix is more flexible, the edge connection density is higher, the realization of rapid convergence is facilitated, and the coding or decoding performance is improved.

The invention discloses a drainage basin pollution discharge transaction knowledge reasoning method based on TransE and Compgcn, and relates to the field of knowledge graph embedding and the field of environmental resources. According to the method, the pre-training embedding of TransE is used as the initial representation of the combined graph convolutional network, so that the training efficiency and the final performance of the model are remarkably improved. The objective of the invention is to provide efficient and accurate derivation support for drainage right transactions in a drainage basin. The method comprises the following steps: firstly, carrying out relation modeling on a knowledge graph by adopting a TransE model to generate high-quality initial entity and relation vector representation; secondly, inputting the embedded images into CompGCN, performing graph convolution operation by utilizing the combination operation characteristics of the embedded images, and introducing an attention mechanism to optimize node representation update so as to capture complex graph structure information; and finally, through iteration of the training process and cost function optimization, rapid convergence and performance improvement of the model are realized. The method is suitable for tasks such as knowledge graph completion and system recommendation, so as to help to improve prediction precision and calculation efficiency.

The application discloses a data center micro-grid electric calculation cooperative double-layer deep reinforcement learning scheduling method and belongs to the technical field of power systems. Firstly, aiming at the high renewable energy penetration scene of the geographical distributed data center micro-grid, a double-layer multi-agent deep reinforcement learning framework is proposed. The double-layer multi-agent deep reinforcement learning framework containing a global layer and a local layer is constructed. The global layer is distributed by centralized deep reinforcement learning training to calculate the task, and the local layer is optimized by decentralized deep reinforcement learning to internally dispatch energy. The layered multi-agent double-delay deep deterministic policy gradient algorithm is combined to realize cross-layer interaction and rapid convergence. Through space-time task adjustment and energy storage cooperation, the application reduces the cost of non-renewable energy by 15%-20%, and the convergence speed is improved by 5 times compared with the prior art. Moreover, the application supports dynamic data backup, significantly improves the system reliability and economy while maintaining the service quality.

The invention relates to the technical field of computers and provides a functional coverage rate code conversion method, computer equipment and a medium. The conversion method comprises the steps of obtaining a first verification plan; analyzing the first verification plan to obtain a plurality of coverage groups corresponding to the plurality of verification scene combinations; the method comprises the steps of generating a plurality of coverage groups, generating a function coverage rate statistical result of the plurality of coverage groups, and then generating a simulation case based on the function coverage rate statistical result of the plurality of coverage groups, the simulation case being used for supplementing a function coverage rate code associated with a first verification plan. In this way, automatic conversion of the function coverage rate codes and rapid convergence of the function coverage rate are achieved.