Search

CN-122001007-A - Wind-solar energy storage cooperative control system based on source-load matching

CN122001007ACN 122001007 ACN122001007 ACN 122001007ACN-122001007-A

Abstract

The invention discloses a wind-solar energy storage cooperative control system based on source-load matching, which is used for aligning and credibly evaluating a source-load storage network and communication information, predicting wind-solar energy and load and giving uncertainty level, calculating matching degree and short-plate hot spots, combining network side hard constraints such as voltage, tide, back-feeding and the like to obtain an acceptable boundary, identifying residual energy/lack energy and reasons when the matching is qualified and residual energy, rapidly checking movable margin and loss by edges, generating a routing plan containing holding time and action budget, rolling and scheduling energy storage and flexible load, and simultaneously writing back risk budget and reserve according to feasibility, prediction uncertainty and communication quality, and realizing gray release and automatic rollback by matching with degradation and playback score, thereby improving consumption and traceability.

Inventors

  • LEI MINGYU
  • CHEN FEN

Assignees

  • 北京垒实科技有限公司

Dates

Publication Date
20260508
Application Date
20251230

Claims (10)

  1. 1. The wind-solar energy storage cooperative control system based on source-load matching is characterized by comprising a scene of source-load storage for a power distribution network, a micro-grid or a park, wherein the system comprises the following components in communication connection with each other: (1) The data processing unit is used for performing time alignment and credibility evaluation on source side output, load side load, energy storage state, network side operation measurement and communication state, and outputting a data quality mark Q_score and communication quality comm_quality; (2) The prediction unit is used for outputting a prediction sequence of wind-light output and load in a prediction time domain H and an uncertainty grade CI_level; (3) The matching evaluation unit is used for calculating the comprehensive matching degree S_total according to the prediction sequence and outputting a short-board hot spot hotspot; (4) The network side acceptance calculating unit is used for calculating an upper limit P_acceptance_max and a permissible feedback power P_export_allowed according to voltage, line load, section power flow, feedback, unbalance and electric energy quality constraint; (5) The residual energy/energy lack judging unit is used for outputting residual energy power P_sum and energy lack power P_deficits when the trigger criterion is met, and outputting a space constraint reason code constraint_cause and an energy lack Value factor Value; (6) The edge feasibility checking unit is used for outputting a migratable margin TRANSFERMARGIN (i-j), an incremental loss estimation loss_est and a feasibility score feasilty_score aiming at the candidate residual energy point i and the defect energy point j; (7) The routing and scheduling unit is used for generating an energy routing plan RoutePlan when the feasibility_score meets a threshold value, generating and issuing a control plan comprising energy storage charging and discharging, flexible load adjustment, reactive support, limit sending and RoutePlan in a rolling way in the day and on a real-time scale; (8) The risk budget write-back unit is used for distributing opportunity constraint risk budget epsilon_k and reserve_req, and executing write-back update on epsilon_k and reserve_req according to the feassibility_score, CI_level and comm_quality; (9) The arbitration and degradation unit is used for performing hierarchical arbitration and switching among the multi-level degradation levels when resources conflict or data/communication is abnormal, and performing write-back triggering recalculation on degradation, current limiting or interlocking events; (10) The playback verification and audit unit is used for forming traceable records of data, models, plans, instructions and receipts, and executing strategy A/B comparison, gray release and automatic rollback based on playback consistency scores replay _score; The system meets the requirements that prediction residual errors, constraint violations, action times and communication quality in the running process generate statistics through the playback verification and audit unit and are used for driving the risk budget write-back unit to adaptively update epsilon_k and reserve_req so as to realize closed-loop balance between constraint feasibility and action vibration suppression at the network side.
  2. 2. The wind-solar hybrid control system based on source load matching according to claim 1, wherein the comprehensive matching degree s_total is at least synthesized by a time shape matching sub-index s_shape, a frequency domain matching sub-index s_freq, a flexible coverage sub-index s_flex and a network side coverage sub-index s_grid according to weights, wherein s_freq performs ratio matching on a payload fluctuation amplitude amp_net (b_k) and a resource tracking capability cap_following (b_k) based on a plurality of frequency bands b_k, and outputs a hot spot hotspot indicating a short-board item.
  3. 3. The wind-solar energy storage cooperative control system based on source load matching according to claim 1, wherein the residual energy/energy shortage judging unit starts a residual energy routing link when the following trigger criteria are simultaneously met, wherein S_total is more than or equal to S_th, P_surplus (t) is more than or equal to P_th in a prediction time domain H, and constraint_cause belongs to a space constraint set; the space constraint set at least comprises a voltage upper limit, a feedback limit, a section congestion, an unbalance constraint and an electric energy quality constraint, and the constraint_cause is used for screening of a defect point or selecting of a routing strategy.
  4. 4. The wind-solar energy storage cooperative control system based on source load matching according to claim 1, wherein the network side receiving computing unit computes p_accept_max in a hard constraint intersection mode, namely, p_accept_max is a minimum value of a plurality of constraint upper limits including at least a voltage margin upper limit, a line load margin upper limit, a section power flow margin upper limit, a feedback margin upper limit, an unbalance margin upper limit and an electric energy quality margin upper limit, and p_export_allowed is a minimum value of a feedback power limit and a feedback margin obtained by mapping a section margin or a voltage margin.
  5. 5. The wind-solar hybrid control system based on source load matching according to claim 1, wherein the edge feasibility checking unit performs constraint checking item by item on candidate migration power based on a sensitivity matrix or linearization trend approximation to obtain a power upper limit p_limit_ corresponding to each constraint, and makes TRANSFERMARGIN (i→j) =min { p_limit_ }; and only allowing writing RoutePlan when the feassibility_score is greater than or equal to F_th and the loss_est is less than or equal to loss_th, otherwise, degrading the space migration into a time migration or limited issue strategy, wherein F_th and loss_th are configurable thresholds.
  6. 6. The collaborative control system for wind and solar energy storage based on source load matching according to claim 1, wherein RoutePlan is managed in a state machine mode and at least comprises an initialization state INIT, an activation state ACTIVE and a RELEASE state RELEASE, and the entry condition of ACTIVE at least comprises the trigger criterion and the feasibility_score not less than F_th; The entry conditions of RELEASE include at least constraint touchdown, comm_quality below threshold, device current limiting/interlocking, revenue drop exceeding threshold or ttl expiration, and state switching frequency limited by minimum hold time t_hold and hysteresis threshold; Wherein the RoutePlan includes an action budget ActionBudget, the ActionBudget defines that the number of switch operations within the sliding window t_win does not exceed n_sw_max, the number of route switches does not exceed n_route_max, and the rate of change of interconnect power does not exceed (dP/dt) _max; spatial migration is suppressed and temporal migration is preferentially employed when ActionBudget is insufficient.
  7. 7. The wind-solar energy storage cooperative control system based on source load matching according to claim 1, wherein the risk budget write-back unit performs write-back on opportunity constraint risk budget epsilon_k according to the following rule that epsilon_k is limited to epsilon_min and epsilon_max, and the variation of each rolling period does not exceed delta epsilon_max; tightening epsilon_k and increasing standby demand reserve_req when feasilty_score decreases, uncertainty level CI_level increases or comm_quality decreases; and relaxing epsilon_k and reducing reserved_req when the key network side constraint is not violated and comm_quality is not lower than a threshold value in N consecutive rolling periods, wherein N, epsilon_min, epsilon_max and delta epsilon_max are configurable parameters.
  8. 8. The wind-solar hybrid control system based on source load matching according to claim 1, wherein the arbitration and degradation unit arbitrates according to priority levels of network-side safety constraint over equipment protection, equipment protection over energy storage safety belts, energy storage safety belts over critical load guarantee, and critical load guarantee over economy; And the multi-level degradation levels include at least a first level that allows spatial migration and a second level that prohibits spatial migration that allows only temporal migration, triggering write back recalculation of RoutePlan and the risk budget upon entry into the second level.
  9. 9. The wind-solar hybrid control system based on source load matching according to claim 1, wherein the playback consistency score replay _score at least comprises constraint violation cost terms, deviation cost terms and action cost terms, and is used for gray release and automatic rollback; when replay _score deteriorates beyond a threshold Δr_th within a sliding window t_eval, where t_eval and Δr_th are configurable parameters, a rollback to the historical policy version is triggered and at the same time a risk budget tightening and scheduling recalculation is triggered.
  10. 10. The method for collaborative control of wind and solar energy storage based on source load matching according to any one of claims 1-9, wherein the method is performed by the system of any one of claims 1-9 and comprises at least time alignment and credibility assessment of source load storage network and communication data; outputting a wind-light and load prediction sequence and an uncertainty level; calculating comprehensive matching degree and outputting hot spots; calculating the network side acceptable upper limit and the allowable reverse power according to the hard constraint intersection; outputting residual energy power, energy deficiency power and space constraint reason codes when the triggering criterion is met, and checking edge feasibility; Generating RoutePlan containing ttl, t_hold, action budget and switching cost when the feasibility meets the threshold value, and performing rolling scheduling in the day before, day in and real-time scale; Updating the risk budget and the standby requirement according to the feasibility score, the uncertainty level and the communication quality write-back; performing degradation arbitration and write-back triggering recalculation when resources conflict or are abnormal; and performing gray level publishing and automatic rollback based on the playback consistency score and forming an audit evidence chain record.

Description

Wind-solar energy storage cooperative control system based on source-load matching Technical Field The invention relates to the technical field of automatic control of a power system and grid-connected operation of new energy, in particular to cooperative control of wind power, photovoltaics, energy storage, flexible interconnection devices and flexible loads in a power distribution network, a micro-grid and a park source network load storage scene. More particularly, the invention relates to a wind-solar energy storage cooperative control system and a control method thereof, wherein the wind-solar energy storage cooperative control system takes 'source-load matching evaluation' as a drive, takes network side hard constraint feasibility as a boundary and takes risk budget write-back and playback consistency verification as a closed-loop treatment means. Background With the large-scale access of distributed photovoltaic, distributed wind power, user side energy storage and adjustable load on the power distribution network side, the power distribution network operates to present the characteristics of strong new energy output fluctuation, obvious prediction error, increased risk of power flow reversal and node voltage out-of-limit, three-phase unbalance, increased electric energy quality problem and the like. In the prior art, a scheduling mode of 'day-ahead planning + intra-day correction + real-time control' is generally adopted, or a control strategy with energy storage smoothing and peak clipping and valley filling as cores is adopted, but the following defects still exist in engineering landing: Firstly, source load matching evaluation often stays in 'account matching' of an energy layer or a power layer, lacks unified computable indexes for different frequency band fluctuation, climbing capacity, flexible resource coverage and network side admittability, and is difficult to quickly position a short plate and a hot spot. Secondly, even if the policy shows that the matching is qualified in the time dimension, the matching is qualified and the residual energy is abandoned or lost due to space constraint (such as node voltage upper limit, feedback power limit, section congestion, three-phase imbalance, electric energy quality limit and the like), namely, an engineering pain point that the matching is qualified and still wastes is generated. Third, the prior art relies on deterministic control or single safety margin, lacks the risk budget closed loop linked with the prediction uncertainty, communication quality and feasibility verification result, and causes either overserved and reduced consumption or frequent out-of-limit and unstable control. Fourth, policy issuing, conflict arbitration and abnormal degradation mechanisms of a master station-edge-device layer often lack verifiable closed loops, sand box playback, A/B comparison, gray level release, automatic rollback and other treatment capabilities, so that policy online risks are high, and audit tracing is difficult. Therefore, a cooperative control system capable of opening the multi-dimensional matching evaluation, network side hard constraint, residual energy/energy shortage routing, risk budget write-back, trusted degradation, playback verification audit is needed, so that new energy consumption is improved and power discarding/loss is reduced on the premise of ensuring that constraint is feasible. Disclosure of Invention The invention aims to provide a wind-solar energy storage cooperative control system based on source-load matching, which can identify and treat a scene which is qualified in matching and still remains energy under the conditions of prediction uncertainty and network side hard constraint, realize cooperative control of time migration priority and space migration supplement through an edge quick feasibility check and routing state machine, and realize policy self-adaptive convergence and verifiable treatment through risk budget write-back and playback consistency scoring, thereby improving the in-situ capacity of new energy, reducing power abandoning and loss, and improving the operation safety and traceability of a power distribution network. In order to achieve the above purpose, the present invention constructs a closed loop with the "source-load matching drive" as a main line: Data credibility, multi-time scale prediction, multi-dimensional matching evaluation, network side admissible boundary calculation, residual energy/energy shortage identification and reason code, edge feasibility check, routePlan routing and scheduling, risk budget write-back and standby linkage, arbitration and degradation, receipt and playback verification, audit evidence chain solidification and write-back convergence. The method comprises the steps of evaluating supply and demand relations of a time domain, combining frequency domain fluctuation, flexible resource coverage and network side admittability, forming a unified executable boundary by a hard