23 results about "Multi machine" patented technology

The invention discloses a multi-machine multi-card distributed training optimization system and method oriented to a credential heterogeneous environment, and relates to the technical field of distributed training. In order to solve the defects existing in distributed training in a credential and credential heterogeneous environment, the system comprises an application layer for receiving a training task; the core system layer comprises a self-adaptive task scheduling module, a hybrid communication topological optimization module and a dynamic check point and elastic recovery control module, and the scheduling module generates a performance portrait and an optimal scheme based on a training task and adjusts the scheme according to a node fault; the optimization module executes the scheme, firstly detects node topology, constructs a hierarchical path, loads and initializes an optimal communication rear end for target equipment, and synchronizes the path to the target equipment; the control module periodically stores a complete training context and realizes reliable recovery of a training process; the credential hardware abstraction layer is used for providing a uniform interface for realizing underlying resource access and data acquisition; and the hardware resource layer comprises heterogeneous computing nodes and an RoCE network switch.

The invention relates to the technical field of power system energy storage equipment testing, in particular to a small and medium-sized industrial and commercial energy storage cabinet twin-trawling testing system which comprises a main loop, a secondary control loop and an anti-countercurrent detection unit, the main loop comprises double testing loops and four sets of alternating current contactors, and multi-machine twin-trawling testing frameworks such as one-to-one, one-trawling-two and one-trawling-three can be flexibly combined; independent switching between grid connection and grid disconnection is supported, the secondary control loop comprises a button with a lamp and a grouping reset button, and the countercurrent prevention detection unit detects the energy flow direction in real time through an intelligent electric meter and a current sensor and has a mains supply charging function. According to the test system, through a double-loop parallel test architecture and a multi-machine flexible combination design, the system can be synchronously butted with multiple energy storage cabinets to carry out parallel test, the test efficiency can be improved, and a safety system with anti-countercurrent detection, rapid mode switching and precise control protection guarantees the test safety; the arrangement of the standardized terminal and the button with the lamp simplifies the operation difficulty, and reduces the test threshold.

The application discloses a kind of multi-machine cooperation SLAM methods based on active deep reinforcement learning.The method comprises the following steps: running ORB-SLAM2 program to robot, and the initial motion trajectory pose graph of multiple machines is obtained by pose estimation of image acquisition through camera;Based on the obtained robot motion trajectory pose graph, more accurate pose is obtained by using deep reinforcement learning TD3 algorithm training to optimize trajectory;On the basis of reinforcement learning algorithm, active perception strategy is introduced to optimize the pose of multiple machines simultaneously, and according to the real-time SLAM estimation probability value P, the corresponding robot is selected to optimize the pose information by TD3 algorithm;The pose information and actual distance information of each robot are transmitted between robots, and the back-end optimization of SLAM trajectory is carried out using TD3 algorithm, to eliminate the cumulative error effect.The application can effectively eliminate the error accumulation in SLAM system, improve the positioning and mapping accuracy of SLAM, and there is no loop restriction, which increases the robustness of SLAM system.

The application discloses a kind of multi-machine frequency response analysis method, device, equipment, medium and program product, it is related to the multi-machine frequency response modeling analysis technical field of enhanced SVG and distributed phase modifier, including: inputting preset disturbance power in the multi-machine frequency response model of target multi-machine grid-connected system, obtains the grid frequency variation of target multi-machine grid-connected system;Wherein, multi-machine frequency response model is based on the first equivalent frequency response model of phase modifier cluster access grid and the second equivalent frequency response model of enhanced SVG access grid is constructed, and is configured to reflect the functional relationship between disturbance power injected into grid and grid frequency variation, phase modifier cluster includes multiple distributed phase modifier;Based on grid frequency variation, the frequency stability of target multi-machine grid-connected system is analyzed.The application can realize the systematic frequency response analysis of enhanced SVG and distributed phase modifier multi-machine grid connection.

The application provides a concrete vibration table parameter determination method based on a same-direction multi-machine self-synchronization, and specifically comprises the following steps: S1: obtaining kinetic energy, potential energy and a generalized force equation of the concrete vibration table according to a dynamic model of the concrete vibration table; S2: obtaining a motion differential equation of the concrete vibration table in combination with a Lagrange equation; S3: substituting an average phase angle, an angular velocity and an angular acceleration of an exciter into the motion differential equation of the concrete vibration table to obtain a dimensionless coupling equation of the concrete vibration table; S4: obtaining a synchronization condition in combination with the dimensionless coupling equation of the concrete vibration table and a limitation condition of an electromagnetic torque; and S5: obtaining a stability condition in combination with the dimensionless coupling equation of the concrete vibration table under the condition of meeting the synchronization condition. The method increases the vibration intensity of the multiple exciters in the concrete vibration table, thereby ensuring the vibration intensity of the concrete vibration table and improving the quality of the concrete prefabricated part.

The invention discloses a multi-machine grid-connected and off-grid control method for energy storage converters, and relates to the field of power grid control, and the method comprises the steps: dividing m energy storage converters into n blocks according to the spatial position, setting a common energy storage converter between the same grid-connected and off-grid instruction blocks, and forming a block chain; setting a first energy storage converter and a second energy storage converter in the same block according to the spatial positions of the energy storage converters; according to the actual grid-connected and off-grid requirements and the fault state of the energy storage converter, the block division coverage range is adjusted; the energy storage converter in the block chain is adjusted to compete again to generate a new public energy storage converter; the energy storage converters in the blocks are adjusted to compete again to generate a new first energy storage converter and a new second energy storage converter; and the energy storage converter of the adjusting block part receives the grid-connected and off-grid instruction to perform grid-connected and off-grid so as to realize multi-machine grid-connected and off-grid control of a dynamic region, and the stability and efficiency of multi-machine grid-connected and off-grid instruction transfer are improved.

The invention discloses a Cesium-based multi-machine multi-mode remote sensing monitoring and three-dimensional visualization system and method, and the system comprises a data collection and communication module, a rear-end data processing module, a data management and release module, a space analysis module, a monitoring module, a video stream synchronization module, and a front-end display module. The system accesses the multi-aircraft flight state and load data in real time, realizes state data fusion through time alignment and exception elimination, and constructs a continuous smooth track; lightweight video frame splicing is realized, and a spliced image is projected to a three-dimensional earth surface in combination with the pose of the unmanned aerial vehicle. The method supports spatial analysis to carry out auxiliary decision making of unmanned aerial vehicle tasks, realizes automatic release and multivariate fusion display of multi-modal remote sensing data, realizes integrated visualization and linkage analysis of multi-source remote sensing information, can improve situation awareness, spatial positioning precision and data processing real-time performance of multi-aircraft cooperative tasks, and has a good application prospect. The method has the characteristics of high real-time performance, high data fusion degree and excellent three-dimensional display effect.

The application belongs to the technical field of energy storage systems, and relates to a multi-machine parallel network energy storage converter fault ride-through method. θ ref The internal potential E output by the reactive power ring is combined to obtain a modulation wave; during normal operation, the switch is connected to the 1 / s integral link, and the system frequency is stabilized in cooperation with the active power ring; when a fault occurs, the switch is disconnected and does not access any link, and the system operates at a fixed frequency; when the fault is recovered, the switch is connected to the k d The proportional link and the 1 / s integral link are connected in parallel branches, power and frequency oscillations are quickly suppressed, and the system gradually recovers to normal; the application can ensure that the multi-machine parallel energy storage converter system realizes frequency synchronization, power balance and stable output during the whole fault ride-through process, greatly improves the fault ride-through capability and grid connection adaptability of the system, and provides a reliable technical guarantee for stable operation of a large-scale energy storage substation.

This invention relates to the field of data management technology, specifically to a multi-machine, multi-process synchronous data management system and method based on block tasks. It includes a technical support unit, a multi-process synchronization unit, a data synchronization unit, and a data management unit. The technical support unit supports system connection and operation; the multi-process synchronization unit manages the synchronization of multiple concurrent processes; the data synchronization unit implements data synchronization within the system; and the data management unit manages the synchronized data. The system designed in this invention can improve system parallel efficiency, quickly achieve massive data synchronization, and then process and provide the data for user access and application. Its method can reduce the workload of data transmission, lower the probability of data loss, improve data synchronization efficiency, reduce user workload, lower the probability of data corruption, ensure business operation progress, effectively improve resource utilization, and increase system throughput.

The application discloses an inverter single-phase / two-phase island detection method and device based on negative sequence components and belongs to the technical field of inverter island detection, which comprises the following steps: injecting periodic reactive disturbance current into a power grid through each inverter, including four disturbance modes in one disturbance period, each disturbance mode lasts N rated grid periods, and the amplitudes of positive and negative reactive disturbance currents are not equal; for each inverter, the following steps are performed: (S1) obtaining the voltage negative sequence component vector of a common coupling point in the current rated grid period, and judging whether disturbance reset is needed; (S2) updating the number of rated grid periods AI_CNT of the current disturbance mode, if AI_CNT is not equal to 0, turning to (S1), otherwise, turning to (S3); (S3) point multiplying the voltage negative sequence component vector change amount of the current and the previous disturbance mode; (S4) judging whether the island detection criterion is met according to the point multiplication result. The application can improve the accuracy of single-phase or two-phase island detection and can be applied to multi-machine grid-connected working conditions.

The invention discloses a multi-machine co-control power distribution network distributed protection terminal self-healing collaborative test method. The method comprises the following steps: generating diversified fault scene test cases according to preset topology and power distribution network distributed protection self-healing terminal configuration; the multi-terminal synchronous control unit is used for synchronously triggering action instructions of a plurality of distributed protection terminals, and simulating an information interaction and collaborative decision process between the terminals; action time, logic judgment results, fault isolation and power supply recovery path indexes of each terminal are recorded in real time through a data acquisition and analysis module; according to a preset self-healing function evaluation system, performing quantitative evaluation on the accuracy, timeliness and coordination of the multi-terminal coordination action, and generating a test report; a multi-machine simultaneous control test system with a three-level architecture of a cloud platform layer, an edge control layer and an equipment layer is constructed, and a standardized test process and a quantitative evaluation system are combined, so that comprehensive verification of the collaborative self-healing function of the distributed protection terminal of the power distribution network is realized.

The invention discloses a reactor and multi-machine leading phase collaborative optimization method and system based on an improved whale algorithm, and relates to the technical field of power system voltage stability control, and the method comprises the steps: obtaining power flow distribution and small disturbance characteristic data, determining a generator set of leading phase operation according to the participation degree of a leading electromechanical mode, and recording the maximum allowable leading phase depth; establishing a multi-target optimization model with voltage deviation minimization and reactive power dispatching cost minimization as targets by combining tidal current power data, voltage measurement data and reactor operation information under the constraint of the maximum allowable leading phase depth; and an improved hybrid adaptive whale algorithm is adopted for solving, so that synchronous optimization of the leading phase depth of the generator and the switching state of the reactor is realized. According to the method, cooperative control of multiple units and the reactor can be realized, the voltage distribution balance is effectively improved, the reactive power dispatching cost is reduced, and the system stability and economy are enhanced.

This invention discloses a transient stability analysis method and system for a multi-machine grid-connected system. First, a transient model of the system is established, and the stable equilibrium point before the fault is determined based on the power balance equation and the eigenvalues ​​of the Jacobian matrix. Using the stable equilibrium point as initial values, the fault trajectory is calculated, and an interpolation correction algorithm is introduced to eliminate discrete simulation step size errors and accurately locate the point on the fault trajectory where the potential energy derivative changes from positive to negative. Using this point as an initial value, the system iteratively searches for the minimum point of power imbalance along the gradient descent path. Solving the power balance equation using the minimum point as an initial value yields the dominant unstable equilibrium point. Finally, the energy of the dominant unstable equilibrium point is used as the critical energy to determine whether the system is stable after the fault is cleared. The method provided by this invention reduces the conservatism of the traditional energy function method for stability assessment in multi-machine systems, effectively reduces the multi-saddle point interference problem in multi-machine systems, and accurately locates the dominant unstable equilibrium point related to the fault path.

This invention discloses a multi-aircraft, multi-modal remote sensing monitoring and 3D visualization system and method based on Cesium. The system includes: a data acquisition and communication module, a backend data processing module, a data management and publishing module, a spatial analysis module, a monitoring module, a video stream synchronization module, and a frontend display module. The system receives multi-aircraft flight status and payload data in real time, achieves status data fusion through time alignment and anomaly removal, and constructs continuous and smooth flight paths; it achieves lightweight video frame stitching, and projects the stitched image onto a 3D surface in combination with the UAV's pose. This invention supports spatial analysis for auxiliary decision-making in UAV missions, and realizes automatic publishing and multi-source fusion display of multi-modal remote sensing data. It achieves integrated visualization and linkage analysis of multi-source remote sensing information, which can improve the situational awareness, spatial positioning accuracy, and real-time data processing of multi-aircraft collaborative missions. It has the characteristics of strong real-time performance, high data fusion degree, and excellent 3D display effect.

The invention discloses a multi-machine frequency response analysis method and device, equipment, a medium and a program product, and relates to the technical field of multi-machine frequency response modeling analysis of an enhanced SVG and a distributed phase modifier, and the method comprises the steps: inputting preset disturbance power into a multi-machine frequency response model of a target multi-machine grid-connected system, obtaining the power grid frequency variation of the target multi-machine grid-connected system; wherein the multi-machine frequency response model is constructed based on a first equivalent frequency response model of a phase modifier cluster access power grid and a second equivalent frequency response model of an enhanced SVG access power grid, and is configured to reflect a function relationship between disturbance power injected into the power grid and power grid frequency variation; the phase modifier cluster comprises a plurality of distributed phase modifiers; and based on the power grid frequency variation, analyzing the frequency stability of the target multi-machine grid-connected system. According to the invention, systematic frequency response analysis of multi-machine grid connection of the enhanced SVG and the distributed phase modifier can be realized.

The invention provides a new energy station multi-machine grid-connected stability margin online evaluation and control design method, which comprises the following steps: step 010, station multi-machine model construction: establishing a control port extension type unit model, and considering an electric network to form a station multi-machine closed-loop model of an explicit retention control loop structure; 020, stability margin online evaluation: putting forward an oscillation mode numerical search algorithm, and respectively substituting local / global information to realize unit and station margin online evaluation; 030, margin feedback control design: constructing a unit autonomous control law to realize zero-difference adjustment and quick response control of margin reference, and constructing a station correction control law to improve the accuracy of unit control; and 040, online refreshing of the margin control parameters: establishing an increment relationship between the damping control quantity and the stability margin based on the time-frequency sensitivity, and performing online calculation of the margin control parameters by controlling discretization. According to the method, the broadband dynamic and multi-machine coupling influence of the new energy field station can be considered, and the accuracy and rapidity of stability margin control can be considered.

This disclosure provides a multi-terminal flexible DC adaptive grid-connection control system, method, equipment, and medium, belonging to the field of flexible DC transmission technology. The system sets the power setpoints of the first, second, third, and fourth adaptive grid-connection converters based on the DC line losses; constructs an adaptive grid-connection control model; determines the AC voltage command based on the adaptive grid-connection control model; establishes a current inner-loop constraint control strategy based on virtual admittance and current clamping; and determines the current inner-loop control command based on the current inner-loop constraint control strategy. Through multi-machine peer-to-peer grid-connection control of the multi-terminal flexible DC converters, the multi-terminal flexible DC achieves multi-terminal active synchronization and flexible access of new energy sources, activating the active support capability of flexible DC for both the receiving and sending grids, and realizing the adaptive, active, synchronous, and stable transmission of new energy sources without conventional power sources.

The invention belongs to the technical field of energy storage power station control, and discloses a battery energy storage power station power control method, system and device and a medium. The method comprises the following steps: collecting power grid frequency, power grid voltage, total power demand of a battery energy storage power station and real-time operation state data of each energy storage unit; calculating the total power demand of the peak regulation area and the total power demand of the active support area through a fuzzy adaptive low-pass filter; and calculating a power instruction of the energy storage unit in each area based on the total power demand of the peak regulation area and the total power demand of the active support area, and issuing the power instruction to the corresponding energy storage converter for execution. Power distribution is carried out based on the fuzzy self-adaptive low-pass filter, the cooperative control level of the multi-machine parallel energy storage unit of the battery energy storage power station is improved, the influence of frequent charging and discharging fluctuation of the energy storage unit on the performance of the energy storage system is relieved, the SOC consistency level of the energy storage unit is improved, and the overall charging and discharging capacity of the energy storage power station is improved.

The embodiment of the application provides a kind of based on the method, system, equipment and medium of multi-modal oscillation suppression of damping coordination control, belong to the field of coordination control new energy station inside different output characteristic inverter maintain system transient stability.It comprises: the multiple-node admittance matrix of multi-machine grid-connected system is constructed, one or more oscillation modes of multi-machine grid-connected system is identified based on multiple-node admittance matrix, and the participation factor of each unit under each oscillation mode is determined;According to its quantitative contribution degree of each unit to corresponding oscillation mode, determine the responsibility weight of each unit in damping coordination control based on it;Reconstruction damping distribution coefficient, and it is mapped to the additional damping control parameter of each unit;Adjust the additional signal of injection inverter control loop, realize the coordinated suppression of multi-modal oscillation of multi-machine grid-connected system.The precise quantification and differentiated distribution of the damping responsibility of each unit are realized, so that it dynamically coordinates the damping resource of each unit, suppresses wideband multi-modal oscillation under multiple working conditions.

The utility model relates to an auxiliary device for honing an inner hole of a valve body. The auxiliary device comprises a bottom plate, a supporting plate, a baffle, a guide plate, a guide sleeve and a core fixing seat, a core fixing base penetrates through the center of the bottom plate, a supporting plate is arranged at the top of the bottom plate and is of a cuboid structure, one side wall of the supporting plate is open, a baffle is arranged on the outer side of the side wall opposite to the open side wall, a guide plate is arranged at the top of the supporting plate, and a guide sleeve penetrates through the center of the guide plate. The distance from the bottom surface of the guide plate to the bottom of the guide sleeve is greater than that from the top surface of the bottom plate to the top of the fixed core seat; according to the valve body inner hole honing auxiliary device, the clamp time and the clamping time are saved; the tool is simple and convenient to operate and practical, the machining efficiency is improved, the labor intensity of workers is relieved, multiple valve body valve holes are honed at the same time through the tool, a workpiece can be rapidly and accurately positioned, and the machining precision of the honing holes of the honing valve bodies is guaranteed; and a worker can easily master the operation of multiple machines by one person.

The application provides a group black start method and device of a multi-machine parallel energy storage system and an energy storage system, and belongs to the field of energy storage. The method comprises the following steps: obtaining positive sequence feedback current average, negative sequence feedback current average and group black start given voltage of the energy storage system; performing current sharing control based on the positive sequence feedback current average and the negative sequence feedback current average to obtain positive sequence average current control quantity and negative sequence average current control quantity; performing positive sequence voltage control based on the positive sequence average current control quantity and the group black start given voltage to obtain given output current; performing current control based on the negative sequence average current control quantity and the given output current to determine a target control quantity; and controlling the corresponding energy storage converter to perform synchronous black start based on the target control quantity. The application can realize synchronous black start of each energy storage converter, and can save a large amount of time without starting in sequence, while ensuring that the power of the energy storage system can reach the expected output level.

The application provides a kind of parallel and off-grid switching method of multiple machine parallel energy storage system, controller and energy storage system, belongs to the field of energy storage.Energy storage system includes multiple parallel energy storage converters, the method includes: when energy storage system works in off-grid mode, if it is detected that power grid recovers normal, then control each energy storage converter to track grid voltage, so that the off-grid voltage of each energy storage converter is kept synchronous with grid voltage;When it is detected that the off-grid voltage of each energy storage converter is kept synchronous with grid voltage, control grid-connected switch to close;Each energy storage converter is connected in parallel, and is connected with grid through grid-connected switch;The working loop of each energy storage converter is switched from off-grid loop to grid-connected loop.The application can realize the automatic switching from off-grid to grid-connected when multiple energy storage converters are connected in parallel, and can realize the non-power-down switching from off-grid to grid-connected.

The application discloses a current sharing control method and system for a multi-machine parallel operation energy storage system, comprising: collecting three-phase voltage and AC side synchronous phase angle of the energy storage system; through positive sequence d-q transformation of the three-phase voltage, and according to the AC side synchronous phase angle, an output control phase angle of the energy storage system in off-grid operation is obtained to control the energy storage system to start; based on the obtained active power and reactive power of the energy storage system, after comparison according to average power output by the energy storage system, the output voltage of the energy storage system is corrected by obtaining the output of the power outer ring of the energy storage system, to inhibit the circulating current between the converters of the energy storage system; the application can inhibit the circulating current between the converters, ensure stable operation of the energy storage system, and overcome the defects of unstable system operation caused by unbalanced load by judging the balance of the load, thereby providing technical support for safe operation of the multi-machine parallel energy storage system.