44results about How to "Eliminate static errors" patented technology

The invention relates to a control method of a grid-connected inverter of a large-scale grid-connected photovoltaic power station. The control method is characterized by comprising the following steps: establishing a grid-connected inverter model under a d-q coordinate system; realizing active and passive power decoupling control by adopting a state feedback exact linearization nonlinearity control strategy; and eliminating static control errors generated by inaccurate system parameter estimation by utilizing cascaded PI (Proportion Integration) adjustment. By utilizing the method of the invention, modeling and grid-connected dynamic analysis are carried out on the grid-connected inverter of the large-scale grid-connected photovoltaic power station to fully reflect the validity of the control method under the conditions of considering different power factors, intensities of illumination, faults, and the like. Compared with a traditional control method, the control method disclosed by the invention is scientific, simple to adjust parameters, can eliminate static errors and has the advantages of favorable dynamic characteristics, stronger robustness, strong suitability, high practical application values, and the like.

The present invention provides a rolling optimization method and system for model predictive control of modularized multilevel converters, which respectively calculate the predicted value of the total voltage of the bridge arm and the output phase voltage according to the phase current and the reference value of the internal current of the bridge arm; The current and the sampling value of the internal current of the bridge arm calculate the prediction error of the total voltage of the bridge arm and the output phase voltage; perform rolling optimization on the total voltage of the bridge arm and the output phase voltage according to the prediction error of the total voltage of the bridge arm and the output phase voltage, and obtain the final The total voltage of the bridge arm and the predicted value of the output phase voltage; according to the predicted value of the total voltage of the bridge arm and the output phase voltage, calculate the number of sub-modules that need to be opened for each bridge arm; according to the current direction of the bridge arm and the voltage of the sub-module Sorting, selecting the corresponding sub-module as the input sub-module, and issuing a trigger command to control the switching of the input sub-module and other sub-modules. The invention can eliminate the static error caused by the low accuracy of system parameters, and improve the steady-state accuracy of the system.

A direct torque control method based on space voltage vector prediction, a new voltage vector calculation method is obtained through theoretical derivation, the number of PI regulators is reduced, and the flux linkage at (k+1) time is obtained by predictive control algorithm And torque value, in order to eliminate the problem of one beat delay. It effectively reduces the workload when setting PI parameters, eliminates a one-beat delay, and makes the system have smaller torque fluctuations. With the help of the prediction algorithm to eliminate the influence of one beat delay, the x-axis voltage component obtained by the calculation method can theoretically be the stator flux linkage to achieve the dead-beat control effect, and further reduce the torque ripple, and the y-axis voltage component is adjusted by PI Obtained by the controller, it can ensure that the system achieves the effect of no static difference in torque during operation, so that the static error problem caused by the change of motor parameters can be eliminated.

A Reinforcement Learning-Based Hierarchical Control Method for Salamander Robotic Path Following. Aiming at the path tracking problem of the salamander bionic robot, a layered control framework is established, including the upper controller based on reinforcement learning and the lower controller based on inverse kinematics, which realizes the tracking target path of the salamander bionic robot. Specifically, for the upper controller, the state space representation, action space representation and reward function are designed based on the softActor‑Critic (actor‑critic) algorithm, which can improve tracking accuracy and eliminate static errors. For the underlying controllers, a leg controller and a spine controller based on inverse kinematics are established. Finally, the controller trained by the robot in the simulation environment is migrated to the real environment to verify the feasibility and generalization ability of the algorithm. Experimental results show that the present invention can better complete the control target, and shows a better control effect in terms of migration and generalization from simulation to reality.

The invention discloses a method for controlling CPU utilization rate. The method comprises: starting a work process, wherein the work process and a target CPU have binding relation; obtaining current utilization rate and target utilization rate of the target CPU, and obtaining an error value between the current utilization rate and the target utilization rate; according to the error value, generating work time of the work process; and according to the work time of the work process, controlling the work process. The method can make CPU utilization rate control achieve characteristics of stable effect, high real-time property, and high accuracy. The invention also discloses a device for controlling CPU utilization rate.