705 results about "Voltage stability" patented technology

The present invention relates to the technical field of power distribution networks. Disclosed are a multi-energy storage power distribution network voltage regulation method and apparatus, and a device. The method comprises: establishing a voltage optimization model for a power distribution network, converting an energy storage constraint in the voltage optimization model of the power distribution network on the basis of physical information of an energy storage system, establishing a security model of the energy storage system, forming a training guide model of the energy storage system on the basis of a Markov decision model, training each energy storage system by using a TD3 algorithm, sharing network parameters obtained through training between different energy storage systems, and adjusting the voltage of the power distribution network. By means of optimizing an operation strategy of an energy storage system, the present invention realizes efficient and sustainable operation of the energy storage system. A charging and discharging strategy of the energy storage system is optimized by means of a DRL algorithm having shared parameters, improving voltage stability of the power distribution network, reducing energy loss, improving the overall efficiency of the system, enhancing the collaboration performance of multiple energy storage systems, and optimizing control of the energy storage system in an active power distribution network.

The invention discloses a phase modifier unit state monitoring and load control method and system for reactive power compensation of a power grid, belongs to the technical field of automatic control of power systems, and mainly solves the problem that service life loss is aggravated or a control target is single due to the fact that the health state of equipment is not fully considered when an existing phase modifier unit responds to reactive power requirements of the power grid. The method comprises the following steps of: calculating reactive vacancy and extracting vibration frequency band energy characteristics by acquiring voltage and frequency of a power grid as well as vibration, temperature and exciting current signals of a unit in real time, and constructing a health state index HSI in combination with self-adaptive wavelet basis selection; taking minimum equipment life loss and optimal power grid voltage stability as multiple objectives, dynamically generating a load control instruction by adopting a particle swarm algorithm, realizing weight adaptive adjustment based on HSI, and introducing an online life loss model as a constraint condition; finally, reactive power output is adjusted through an excitation system, and cooperative control of power grid stability and equipment life extension is achieved.

The invention relates to a power distribution network voltage partition control method and system based on a hierarchical K-means clustering algorithm, and belongs to the technical field of power distribution system voltage partition optimization. According to the technical scheme, power distribution network node parameters are collected in a self-adaptive partition mode, a node sensitivity coefficient and an eigenvector of an electrical distance are constructed, and then a K-means algorithm is used for fine region division; selecting a dominant node: solving a Jacobi matrix through load flow calculation, extracting a voltage sensitivity coefficient, and selecting a node with the maximum sensitivity as a dominant node in each partition; multi-objective optimization: constructing an optimization model with minimum network loss, minimum voltage deviation and highest voltage stability as objectives; and partition cooperative control: accessing wind power and photovoltaic power to the dominant node, adjusting reactive power output in real time according to an optimization result, and realizing partition autonomy and global cooperation. According to the invention, voltage fluctuation and out-of-limit are inhibited, system network loss is reduced, control efficiency and economy are improved, and the method is suitable for complex topology and high permeability scenes.

The invention discloses a protocol conversion and protocol self-learning method and system for transformer area optical storage and charging cooperation, and the method comprises the steps: constructing a digital twinborn body of a transformer area optical storage and charging system, carrying out the parallel operation of a protocol agent and a protocol agent in a virtual environment, and achieving the cooperative training of the protocol agent and the protocol agent through a hierarchical reinforcement learning architecture, the protocol agent is responsible for learning a dynamic priority scheduling and compression strategy for heterogeneous protocol messages such as Modbus, CAN and IEC 104, the protocol agent is responsible for learning a power balance and voltage stability control strategy based on photovoltaic output, energy storage SOC and charging load, and the two agents realize cross-domain collaborative optimization through a reward function mutual coupling mechanism. And finally, safely deploying the collaborative strategy obtained by training to the edge control equipment of the physical transformer area. According to the method, the problems of disjunction of protocol conversion and cooperative control, protocol strategy solidification, insufficient cross-domain cooperation and the like in the prior art are solved, and the operation efficiency and the self-adaptive capability of the transformer area optical storage and charging system are improved.

The invention discloses a hydroelectric generating set speed regulation control method considering excitation coordination. The method comprises the steps that 1, system state information is collected and preprocessed; step 2, on-line identification of excitation speed regulation coupling characteristics; 3, decomposing a multi-time-scale control target; step 4, designing a self-adaptive multivariable controller; 5, working condition self-adaptive parameter optimization; step 6, intelligently inhibiting the water attack effect; 7, multi-machine cooperative reactive power optimization control is carried out; 8, optimizing and executing a control strategy in real time; the technical problems that excitation speed regulation dynamic coupling, multi-time scale response mismatch, multi-machine cooperative control and the like in a hydroelectric generating set control system cannot be comprehensively solved in the prior art are solved, and the regulation response speed and the system operation safety are improved while the frequency and voltage stability is ensured.

The invention discloses a power grid transient stability margin prediction method based on steady-state power flow and anticipated faults, and the method comprises the steps: carrying out the sample generation based on different operation conditions of a power system, an AC power flow, operation modes under different time sequences, and faults under N-1 and N-2 scenes; labeling the generated sample according to two transient stability margin evaluation indexes of power angle stability and voltage stability; the labeled sample data and the anticipated fault data after the embedded coding are combined to serve as input features to be input into a deep learning model and trained, and a trained deep learning model is obtained; and predicting the transient stability margin of the power grid according to the trained deep learning model, and judging the stable state of the power system. The transient stability evaluation sample generation method is provided, the steady-state power flow and the expected fault before the fault are considered, the transient stability margin prediction of the power grid is realized by using the deep learning model, and the transient stability margin prediction speed, precision and reliability are improved.

The invention relates to the technical field of power electronics and power system control, and discloses an intelligent control method and system for a multi-network converter, and the method comprises the steps: collecting the frequency deviation, the power change rate and the voltage amplitude of a power grid in real time, carrying out the filtering and feature extraction, and analyzing the coupling relation, thereby obtaining the parameter coupling strength. And if the intensity exceeds the threshold value, an inertia adjusting signal is generated through a fuzzy logic controller. And calculating a deviation correction value by combining the state evaluation response performance of the converter, and generating an optimization instruction by fusing the frequency deviation. And solving a voltage stability threshold based on the instruction, updating a coupling index, and issuing and executing after verification. And finally, updating the fuzzy rule base by adopting neural network iterative learning according to feedback data, and forming an inertia support frame with self-learning capability. According to the method, the inertia response speed and the operation stability of the multi-network converter system under high-proportion new energy access can be effectively improved, and adaptive cooperative control is realized.

The invention provides a photovoltaic access power distribution network voltage stability evaluation method, which comprises the following steps: acquiring ship docking time, staying duration and crane working frequency in a port area in real time, fusing obtained meteorological information, identifying a photovoltaic power generation output power change trend, and obtaining a daily power load peak-valley time period and a load fluctuation amplitude; according to the adjusted voltage value of each node, generating a switch operation instruction and a power adjustment instruction of the shore power supply equipment, identifying load abrupt change by executing the instructions, and determining a real-time power dispatching scheme by combining the power adjustment instruction and the ship arrival time; and according to the target voltage control value, determining the power utilization safety state of the port power distribution network, and evaluating the voltage stability of the photovoltaic access power distribution network through the operation data of the photovoltaic power generation access point.

The invention relates to the technical field of battery management, and particularly discloses a dynamic control method for an intelligent balance management system of a lithium battery module, which comprises the following steps of: acquiring load current of the battery module and end voltage signals of each monomer in real time; calculating instantaneous internal resistance and polarization voltage parameters of each single battery in real time; the terminal voltage is compensated in real time, and a purified open-circuit voltage signal is generated; fusing the charge state reference value with the current integral result, dynamically calculating a confidence index according to a current standing condition and voltage stability so as to perform adaptive weighting, and outputting a smooth charge state estimation value; calculating a balance urgency degree index by combining the estimated value, the real-time temperature and the inconsistency change trend, and generating a differentiated balance control instruction; an energy transfer path is established and controlled according to the instruction, the rate is dynamically adjusted in the process, and operation is terminated when the duration or state consistency condition is met; according to the method, the stability of state estimation and the accuracy, safety and efficiency of balance control under the dynamic working condition are improved.

The invention provides a voltage stability control method for a multi-output switching power supply, which comprises the following steps of: sampling a main output channel in real time to obtain a voltage signal; the voltage signal is processed through a wide bandwidth error amplifier, a main channel adjusting signal is generated, and the main channel adjusting signal is used for driving a main power switch device; receiving an input voltage fluctuation frequency band of an auxiliary channel, a load current transient change gradient and a frequency spectrum coupling characteristic of a main channel adjusting signal through a dynamic prediction model, and generating an auxiliary channel adjusting signal with a phase lead compensation characteristic; a cross coupling active offset mechanism is constructed, time domain-frequency domain characteristics of coupling interference are extracted through a high-precision differential detection circuit, amplitude-frequency adjustable reverse correction signals are generated in real time through an adaptive compensation algorithm and injected into control nodes of an auxiliary output channel, and dynamic decoupling and ripple suppression are carried out. According to the method, high-stability control of multi-channel voltage under all working conditions is realized, and the dynamic response speed and the anti-interference capability are improved.

The invention is suitable for the technical field of power electronics and power transmission, and provides a low voltage ride-through control method and system for a network construction type converter, and the method comprises the steps: carrying out the conventional VSG control according to a VSG control strategy, and continuously monitoring the difference between the voltage of a power grid and a set reference value, so as to judge whether a fault occurs or not; when it is detected that the power grid breaks down, homodyne control is carried out on an active power control ring and a reactive power control ring of the grid-forming type converter so as to fix the phase and the amplitude of the output voltage of the converter; calculating and injecting a positive sequence compensation voltage and a negative sequence compensation voltage according to the drop degree of the power grid voltage; and according to the predicted impact current value, dynamically calculating and inputting adaptive virtual impedance. The power angle and the voltage in the fault period can be ensured to be stable, transient overcurrent is effectively restrained, and large power fluctuation is not caused.

The invention relates to the technical field of bidirectional PCS, and discloses a bidirectional PCS converter multi-target power balance optimization method and system based on model predictive control. According to the method, real-time operation parameters are analyzed by using a first-order lag initial model, prediction of power disturbance in a plurality of control periods in the future is realized, power balance optimization is converted from passive response to active pre-judgment, and a linear time-varying prediction model is constructed, so that the real-time performance of power balance optimization is improved. According to the method, the evolution trajectory of the direct current bus voltage in a period of time in the future is predicted, the power balance optimization behavior is continuous and prospective through a rolling optimization mechanism of prediction, evaluation and preferential selection, and meanwhile, the three targets of ensuring voltage stability, reducing switching loss and suppressing current harmonic content are fused into power balance optimization, so that the power balance optimization efficiency is improved. The problems of control hysteresis and target singleness existing in a traditional control method are solved, and quick and accurate power balance control of the bidirectional PCS is realized.

The invention provides a high-low voltage valve group heterogeneous flexible direct current power transmission system and a direct current fault clearing method, and belongs to the field of power transmission and distribution of electric power systems, and the system comprises an alternating current collection network which is used for collecting alternating current electric energy of a new energy base; each of the sending end converter station and the receiving end converter station comprises a negative-voltage type modular multilevel converter and a half-bridge type modular multilevel converter; the direct-current overhead line is electrically connected between the sending-end converter station and the receiving-end converter station, and direct-current quick switches are arranged on the two sides of the direct-current overhead line respectively; the sending end converter station adopts cooperative network construction control and is used for maintaining the voltage stability of the alternating current collection network; and the receiving end converter station adopts direct-current voltage control and is used for maintaining the voltage stability of the direct-current electric energy of the direct-current overhead line. According to the invention, large-scale long-distance new energy supply is realized, and the construction cost and the operation loss are reduced.

The invention discloses a new energy power grid frequency voltage collaborative support method, which comprises the following steps: acquiring electrical quantity signals of a plurality of units, and constructing multi-unit power angle trajectory data; regarding the instantaneous nonlinear coupling characteristic as a space curve in differential geometry, calculating a local curvature and a torsion rate of the instantaneous nonlinear coupling characteristic, establishing a three-dimensional dynamic coupling model capable of quantifying the instantaneous nonlinear coupling characteristic, and forming a three-dimensional coupling state matrix; dynamically evaluating the supporting capacity of the unit by using the matrix and combining a Lyapunov index to obtain a dynamic supporting capacity matrix; according to the dynamic support capability matrix, modeling a cooperative control problem as a non-cooperative game model and solving Nash equilibrium, performing conflict decoupling of a multi-machine cooperative strategy, and determining a cooperative control weight matrix; and performing multi-objective collaborative optimization based on the weight matrix to generate a collaborative support control instruction sequence. Transient coupling can be accurately quantified, dynamic self-adaption of a control strategy is achieved, and the frequency and voltage stability of a high-proportion new energy power grid is improved.

The invention discloses a reactive compensation optimal configuration method and system of a new energy base, equipment and a medium, the method performs steady-state configuration of initial reactive capacity based on load flow calculation, can obtain effective steady-state reactive compensation configuration in a full output range, improves the robustness of steady-state reactive compensation, and improves the optimal configuration of the reactive compensation of the new energy base. A global coupling optimization architecture crossing steady state and transient state is constructed for the first time, closed-loop convergence of three optimization configuration stages is realized, so that global optimal configuration of reactive compensation is realized, and meanwhile, reasonable division and matching between two types of dynamic reactive compensation equipment, namely a phase modifier and an SVG / STATCOM, are realized; therefore, the new energy base not only can solve the transient voltage problem, but also has high power grid support strength, and the total configuration cost is greatly reduced while the requirements of standards, short-circuit ratio and transient voltage stability are met.

The invention relates to the technical field of power grid systems, and discloses a unified control method and system for multi-target collaborative optimization of a network construction type energy storage PCS. According to the method, the voltage waveform data and the power scheduling target are subjected to conjoint analysis to generate the voltage stability priority coefficient and the power tracking priority coefficient, so that the physical conflict between meeting the power grid voltage support requirement and meeting the response power output instruction is converted into an executable quantitative basis, and the power scheduling efficiency is improved. The problem that a traditional control method can only carry out rigid switching between a voltage stabilization mode and a power tracking mode is solved, and the battery life loss rate and the system efficiency index are converted into constraint conditions in an optimization problem. A control instruction automatically tends to change towards two objectives of reducing battery performance loss and improving energy storage PCS conversion efficiency, so that the energy storage PCS can perform collaborative optimization on the objectives of power adjustment, power grid voltage stabilization, battery life prolonging and conversion efficiency optimization.

The invention discloses a coordination control method and related equipment for a network-construction-type flexible direct current converter station. The method comprises the following steps: firstly, determining the maximum capacitive reactive power capacity of the network-construction-type flexible direct current converter station; simulating a fault through time domain simulation, and calculating dynamic reactive power required for guaranteeing the stability of the transient voltage of the hub bus; selecting a parallel capacitor bank close to a grid-connected point to form a collaborative resource set, and initializing and distributing the input capacity of the collaborative resource set; taking the flexible direct current reference voltage, the active power and the capacitor input capacity as boundaries, taking maximization of a static voltage stability index as a target, executing continuous power flow optimization, calculating reactive power reserve capacity, and sending a control target after conditions are met. According to the invention, excessive consumption of flexible direct-current reactive power reserve is avoided by determining the capacity reference, and the dynamic reactive power demand is accurately calculated to ensure fault support; the cooperative capacitor bank supplements insufficient static resources and reduces flexible DC burden; through continuous power flow optimization, overall planning of steady-state and transient-state voltage stability is achieved, and the comprehensive supporting capacity of the network-constructing flexible direct current power supply to the voltage stability of the power grid is effectively improved.

The invention relates to the technical field of power distribution automation and intelligent power grid equipment, in particular to an FTU-based intelligent voltage regulation protection bypass cooperative control system, which comprises a state acquisition module, a state identification module, a self-adaptive disturbance judgment module, a voltage regulation cooperative module and a state reconstruction module, the system establishes an operation state vector model by synchronously sampling the voltage and current of a distribution line and the states of a circuit breaker and a bypass switch; judging line operation conditions according to vector state classification, and generating corresponding action priorities and time windows; unified scheduling and interlocking control are carried out on the voltage regulation unit, the protection unit and the bypass unit through the cooperative control execution module, and logic cooperation and state closed-loop updating of multiple control objects are achieved; according to the invention, time sequence coordination and intelligent linkage of voltage regulation, protection and bypass operation can be realized under complex working conditions, the voltage stability and the operation reliability of the distribution line are improved, the fault conflict probability is reduced, and the operation and maintenance intervention is reduced.

The present application discloses a distributed power supply configuration method and apparatus for an energy storage system, and a device. The method comprises: on the basis of an optimization objective, constructing a configuration optimization model of an energy storage system; performing objective unification processing on the configuration optimization model to generate a fuzzy optimization model; and acquiring distributed power supply location information of the energy storage system, and performing calculation on the distributed power supply location information by means of the fuzzy optimization model, so as to perform distributed power supply configuration on the energy storage system. When constructing the configuration optimization model, a plurality of optimization objectives are considered, comprising a minimum grid loss value, a maximum system voltage stability margin value, and a minimum voltage deviation value, thereby helping implement comprehensive optimization of an energy storage system, and improving the operating efficiency and stability of a power grid. Objective unification processing is performed on the configuration optimization model by means of fuzzy set theory, thereby preventing conflicts among sub-objectives. Power supply configuration is performed on the basis of a calculation result, so that power flow distribution of a power grid can be optimized, thereby reducing grid loss and voltage deviation.

The invention relates to the technical field of micro-grids, and discloses a control method of a distributed power supply in a plant micro-grid, and the method comprises the steps: S1, constructing a multi-source collaborative micro-grid system; s2, executing real-time power dynamic allocation; s3, working condition power balance control; s4, active regulation of bidirectional power flow and the like. Based on a real-time AGC instruction and plant load data, a hierarchical coordination mechanism is adopted, charging and discharging power is dynamically adjusted by means of stored energy to track the AGC instruction, and constant power output of wind power is maintained in combination with a wind power prediction model; when a thermal power generating unit is subjected to deep peak regulation, shutdown or fault, the system absorbs excess wind power through energy storage charging, synergistically improves the hydrogen production load power, guarantees dynamic matching of the load and power generation, and actively adjusts energy storage charging and discharging and hydrogen production loads by monitoring the power flow directions of a thermal power generating unit section and a desulfurization section in real time according to voltage fluctuation caused by bidirectional power flow. And the voltage deviation is dynamically compensated, and the system voltage stability is greatly improved.

The invention discloses a switching control method based on SPWM soft start and a pulse type load LLC resonant converter. The converter mainly comprises an LLC resonant converter, an H-bridge inverter circuit and a load of the H-bridge inverter circuit, and the control method comprises the following steps: setting k-level SPWM modulation in a starting stage to realize smooth soft start of output voltage (k is any positive integer); the load state is accurately judged through an H-bridge switching logic unit, and Burst intermittent control and linear frequency control are adopted for no-load and full-load in a steady state respectively; during load switching, a nonlinear SOTC optimal trajectory controller is adopted to perform two-pulse optimal control, so that the state is transited to a new mode along the shortest trajectory, and a dynamic response optimizer is introduced to compensate the switched frequency according to the state radius error so as to suppress secondary voltage fluctuation. The problems of slow dynamic response, large output voltage fluctuation and the like when traditional linear control faces a pulse type load LLC resonant converter are solved, and the stability and the anti-interference capability of the output voltage of the system are improved.

The invention provides a day-ahead reactive power optimization method based on a graph convolutional network and a conditional value-at-risk strategy. The method comprises the following steps: firstly, utilizing Monte Carlo simulation to generate a plurality of joint scenes of wind power and load uncertainty, extracting a typical operation scene through a K-means clustering method, and establishing a multi-target random reactive power optimization model taking active power loss and node voltage deviation as optimization targets; secondly, based on the optimization model, constructing a weighted adjacency matrix fused with power grid physical parameters, and enhancing the recognition capability of the graph convolutional network on the relationship between the nodes; then, a graph convolutional network model is trained to establish efficient mapping between a power system operation state and a reactive power regulation scheme, and rapid strategy generation is realized; and finally, performing cross-scene evaluation on the performance of each scene strategy and the voltage stability risk based on the conditional value-at-risk, and screening out a day-ahead optimization strategy with the best robustness. According to the method, the calculation efficiency and strategy robustness of day-ahead reactive power optimization are effectively improved, and the method is suitable for a power system with high-proportion new energy access.

The invention discloses a flexible AC / DC hybrid power distribution network fault recovery optimization method and system based on multi-agent deep reinforcement learning, a computer device and a non-transient computer readable storage medium, and the method comprises the steps: firstly constructing an AC / DC power distribution network model considering the operation characteristics of a flexible switch and a voltage source converter; then, flexible equipment is modeled as an independent agent, a multi-agent training architecture is established, a composite reward function fusing a load recovery rate, voltage stability and operation safety is designed, and a mechanism combining centralized training and distributed execution is adopted to realize multi-point cooperative control between the flexible equipment; and finally, designing a recovery optimization process for rapid decision and control execution after the fault. The method can realize higher load recovery rate and better voltage control performance in various fault scenes, has the advantages of strong decision real-time performance, good system expandability, strong strategy generalization ability and the like, and is suitable for autonomous recovery and intelligent regulation and control of the novel flexible AC / DC hybrid power distribution network.

The invention discloses an optimized operation method and system for a wind-solar storage micro-grid. According to the method, a three-layer architecture of top-layer optimization, middle-layer cooperation and bottom-layer execution is adopted, and an intelligent closed loop of'prediction-planning-coordination-execution-feedback 'is formed. The top layer carries out global power flow congestion identification and dispersion based on LSTM prediction and a traffic congestion algorithm, carries out multi-target refined optimization in a feasible interval by using an improved particle swarm algorithm, and generates an optimal operation instruction; the middle layer dynamically adjusts and optimizes the target weight according to the operation scene, and decomposes an instruction into a multi-device cooperation strategy; and the bottom layer executes a customized control algorithm to realize accurate regulation and control and millisecond-level rapid protection of the equipment. The problems of micro-grid multi-target collaborative optimization, predictive control disjunction, operation risk look-ahead protection and the like are effectively solved, and the voltage stability, economical efficiency, dynamic response speed, safety and reliability of system operation are remarkably improved.

The invention discloses a no-load state monitoring method and system for an aircraft power supply system, and relates to the technical field of electrical parameter measurement and aircraft power supply system monitoring. Initializing a monitoring system, calibrating a voltage sensor, a current sensor, a frequency sensor and an impedance sensor, setting a no-load state judgment threshold system, and importing historical no-load data as a comparison reference; no-load voltage, current, frequency and impedance signals are synchronously acquired through sampling frequency of the sensor group, and the signals are transmitted to the data processing unit through the shielding cable; median filtering and low-pass filtering are adopted to remove noise interference, normalization processing is carried out on the signals, and basic electrical parameters are calculated; calculating core characteristic values of the voltage stability, the frequency fluctuation coefficient, the impedance stability and the harmonic distortion rate; comparing the characteristic value with a preset threshold value, judging whether the state is normal or not by combining historical data, and identifying an abnormal type; and outputting a judgment result in real time, and starting acousto-optic early warning in case of abnormality. Abnormal traceability and trend prediction are realized, and safe and stable operation of an aircraft power supply system is guaranteed.

The invention discloses a day-ahead power generation plan generation method, system and device considering transient voltage stability constraint and a medium, and the method comprises the steps: constructing a security constraint unit combination model, and solving a day-ahead power generation pre-plan and a corresponding conventional unit initial start-stop scheme; based on a day-ahead power generation pre-plan, generating an alternating current power flow state of the reactive compensation equipment in a preset time period, and for an expected fault set, performing transient voltage stability evaluation on the power flow state of the power grid in each time period; for the units which do not meet the transient voltage stability evaluation standard, constructing an optimization model of a newly-added starting-up unit with the minimum cost, and solving through an improved branch and bound method to obtain a newly-added starting-up scheme; and combining the initial start-stop scheme and the newly added start-up scheme, constructing a security constraint economic dispatching model, and solving to obtain a day-ahead power generation plan. According to the method, the voltage stability of the power grid under high-proportion new energy access can be remarkably improved; the dynamic voltage supporting capability of the power grid is enhanced; and the grid voltage linkage instability risk is avoided.

The invention relates to the technical field of motor driving, and discloses a motor driving direction fault detection and compensation method and device. The method comprises the following steps: collecting state data of a target vehicle and constructing a sensor fusion vector; calculating a compensation torque value based on the state data and the sensor fusion vector; based on the compensation torque value, a boost module and a super capacitor combination in a two-stage voltage stabilizing circuit are driven to carry out voltage pre-adjustment, and a stable bus voltage is obtained; and generating a motor control signal based on the compensation torque value and the stable bus voltage, and sending the motor control signal to a motor controller to execute torque compensation and voltage stable adjustment operation. Self-adaptive compensation torque calculation under different working conditions is achieved, predictive adjustment on instantaneous power requirements is achieved through coordinated work of the boost module and the super capacitor combination, voltage fluctuation in the torque compensation process is effectively restrained, and then effective execution of torque compensation and voltage stability adjustment is ensured.

The invention relates to a multi-agent reinforcement learning voltage control method considering transient voltage stability margin evaluation. The method comprises the following steps: constructing a power grid multi-period operation scene set and a regional stability margin index set; training a constructed multi-layer perceptron model by using the power grid multi-period operation scene set and the regional stability margin index set to obtain a transient voltage stability margin evaluation model; a power system voltage control problem is converted into a Markov decision process, a power grid partition-based multi-agent model is constructed to solve the Markov decision process so as to obtain an optimal control strategy, and the optimal control strategy is that according to actual power grid operation characteristics, the voltage control multi-agent model outputs a voltage adjustment instruction, and the voltage adjustment instruction is output by the power grid partition-based multi-agent model. The voltage regulation instruction maximizes the total reward and punishment function. The voltage control method provided by the invention can maintain enough stability margin to ensure the stability of transient voltage while maximizing economic operation.

The invention discloses a power system transient voltage stability evaluation method, device and equipment and a medium, and the method comprises the steps: collecting real-time electrical data of a target power system, and continuously inputting the real-time electrical data into a preset double-branch network, so that the double-branch network carries out the feature extraction of the real-time electrical data, obtaining local dynamic features and global dependency features; wherein the dual-branch network comprises an LSTM (Long Short Term Memory) branch and a Transform branch; performing feature fusion on the local dynamic features and the global dependency features through a preset variational fusion mechanism to obtain a fusion feature vector; wherein the variational fusion mechanism is characterized in that feature fusion is carried out through Gaussian mixture distribution and a derivable re-parameterization strategy; and generating a stability probability value according to the fusion feature vector, and obtaining a transient voltage stability evaluation result of the target power system according to the stability probability value. According to the invention, the real-time performance and accuracy of transient voltage stability evaluation of the power system can be improved.

The invention relates to the technical field of energy storage, in particular to a network construction type energy storage power balance control method and system based on a hierarchical collaboration strategy. According to the invention, a three-level layered cooperative control architecture including a bottom unit control layer, a middle region coordination layer and a top system optimization layer is introduced, and the bottom unit control layer establishes a quick-response unit-level local control mechanism by collecting power output deviation and voltage fluctuation of an energy storage unit in real time; the middle region coordination layer carries out coordination correction on power instructions among a plurality of energy storage units on the basis of comprehensively considering a line impedance difference coefficient and a region load change rate, and the top system optimization layer carries out global optimization decision by combining a system total power demand, an energy storage residual capacity difference and a power grid frequency deviation. Therefore, dynamic power distribution and voltage stability control among the energy storage units can be realized under different operation conditions, and the problem of failure of traditional droop control under the condition of low impedance difference is remarkably improved.