Search

CN-122026629-A - Cloud-edge cooperation-based intelligent operation and maintenance scheduling optimization method for energy storage power station

CN122026629ACN 122026629 ACN122026629 ACN 122026629ACN-122026629-A

Abstract

The invention relates to the technical field of intelligent operation and maintenance of an energy storage power station, and discloses an intelligent operation and maintenance scheduling optimization method of the energy storage power station based on cloud edge cooperation, which comprises the steps of applying flow velocity pulse excitation to a circulating pump at a station end and collecting fluid responses of all branches, constructing electrolyte residence time distribution and extracting statistics reflecting long tail residence, and simultaneously calculating bypass shunt strength by combining with potential gradient data of a common liquid supply manifold to form a coupling index reflecting the attenuation degree of effective capacity; the edge monitoring end calculates the module capacity loss rate according to the module capacity loss rate and compares the module capacity loss rate with an allowable degradation threshold issued by the cloud end to generate a normal or abnormal operation and maintenance scheduling result, and outputs branch risk sequencing and operation and maintenance work orders under abnormal conditions to realize online identification and refined operation and maintenance management of hidden efficiency degradation problems of the energy storage power station.

Inventors

  • ZHANG ZHIDONG

Assignees

  • 能拓能源股份有限公司

Dates

Publication Date
20260512
Application Date
20260413

Claims (7)

  1. 1. The intelligent operation and maintenance scheduling optimization method for the energy storage power station based on cloud edge cooperation is characterized by comprising the following steps of: The edge monitoring end controls the circulating pump to execute electrolyte flow pulse excitation, collects fluid response data of each galvanic pile branch to construct electrolyte hydraulic transport delay distribution, and extracts transport long tail deviation values representing asymmetry of the hydraulic transport delay distribution; determining manifold bypass shunt strength reflecting that the conductive electrolyte forms a leakage path in the manifold according to the along-path potential gradient data of the common liquid supply manifold; constructing a shunt coupling long tail loss degree representing a nonlinear coupling relation between the shunt strength of the manifold bypass and the transport long tail deviation value so as to quantify effective capacity attenuation caused by long tail retention of electrolyte in a galvanic pile branch; And receiving an allowable efficiency degradation threshold value issued by a cloud management and control platform, carrying out normalized comparison on the module effective capacity loss rate calculated based on the split coupling long tail loss degree and the allowable efficiency degradation threshold value, and outputting an operation and maintenance scheduling instruction containing a priority investigation strategy for the high-bias branch according to a state mark generated by comparison.
  2. 2. The cloud-edge collaboration-based intelligent operation and maintenance scheduling optimization method for the energy storage power station, according to claim 1, is characterized in that an edge monitoring end controls a circulating pump to execute electrolyte flow pulse excitation, fluid response data of each pile branch is collected to construct electrolyte hydraulic transport delay distribution, and the method comprises the following steps: The method comprises the steps of superposing square wave pulses with fixed amplitude and fixed duration on the basis of steady-state pump speed to serve as the pulse excitation of the electrolyte flow velocity, synchronously collecting fluid response data of each pile branch in a preset observation window, removing a steady-state baseline to obtain a net impulse response component, performing full-time domain integration on the net impulse response component, and performing normalization processing on the net impulse response component by utilizing an integration result to obtain the electrolyte hydraulic transport delay distribution.
  3. 3. The cloud-edge-synergy-based intelligent operation and maintenance scheduling optimization method for the energy storage power station, according to claim 2, is characterized by extracting a transportation long tail skewness value representing asymmetry of the hydraulic transportation delay distribution, and comprises the following steps: the method comprises the steps of carrying out time weighted integration on the electrolyte hydraulic transport delay distribution, calculating to obtain average hydraulic retention time of the electrolyte in a branch, calculating third-order central moment of the electrolyte hydraulic transport delay distribution based on the average hydraulic retention time, and taking the third-order central moment as the transport long tail deviation value for quantifying the fluid retention tailing degree so as to represent the asymmetric distribution characteristics of the electrolyte in the pipeline transmission process.
  4. 4. The cloud-edge collaboration-based intelligent operation and maintenance scheduling optimization method for the energy storage power station, according to claim 1, is characterized in that determining manifold bypass shunt strength reflecting formation of a leakage path of conductive electrolyte in a manifold according to along-path potential gradient data of a common liquid supply manifold, and comprises the following steps: The method comprises the steps of calculating a root mean square value of potential difference of adjacent potential sampling points on a common liquid supply manifold, taking the root mean square value as a potential gradient index of driving shunt, calculating manifold equivalent resistance according to geometric dimension parameters of the common liquid supply manifold and collected electrolyte real-time conductivity, obtaining effective retention volume of conductive liquid based on real-time flow and average hydraulic retention time of each pile branch by weighting and summarizing, and dividing the potential gradient index by a product of the manifold equivalent resistance, theoretical charge density of unit volume and the effective retention volume of the conductive liquid to obtain shunt strength of the manifold bypass.
  5. 5. The cloud-edge-synergy-based intelligent operation and maintenance scheduling optimization method for the energy storage power station, according to claim 1, is characterized by constructing a split coupling long-tail loss degree representing a nonlinear coupling relation between the manifold bypass split strength and the transportation long-tail deviation value, and comprises the following steps: And multiplying the cubic value with the transportation long tail deflection value corresponding to the electric pile branch, and dividing the operation result by a fixed coefficient to obtain the split coupling long tail loss degree for quantifying the phagocytosis degree of the electric pile branch by split current due to long tail retention.
  6. 6. The cloud-edge-collaboration-based intelligent operation and maintenance scheduling optimization method for the energy storage power station, according to claim 5, is characterized in that an allowable efficiency degradation threshold issued by a cloud management and control platform is received, a module effective capacity loss rate calculated based on the split coupling long tail loss degree is normalized and compared with the allowable efficiency degradation threshold, and according to a state sign generated by comparison, the method comprises the following steps: The method comprises the steps of utilizing the manifold bypass distribution strength, average hydraulic retention time, second order central moment of hydraulic transport delay distribution and the distribution coupling long tail loss degree, approximately calculating effective charge survival fractions of all pile branches through accumulation quantity expansion, taking a value obtained by subtracting the effective charge survival fractions as a branch loss rate, carrying out weighted summation on the branch loss rates based on flow duty ratios of all pile branches to obtain the module effective capacity loss rate, calculating the ratio of the module effective capacity loss rate to the allowable efficiency degradation threshold, and setting the state mark to be an abnormal value if the difference value obtained by subtracting the ratio is larger than or equal to zero, otherwise setting the state mark to be a normal value.
  7. 7. The cloud-edge collaboration-based intelligent operation and maintenance scheduling optimization method for the energy storage power station, according to claim 6, is characterized in that outputting an operation and maintenance scheduling instruction containing a priority investigation strategy for a high-bias branch, and comprises the following steps: The method comprises the steps of constructing a linear combination model comprising a normal strategy vector and an abnormal strategy vector, utilizing the state mark as a weight coefficient to select and output the operation and maintenance scheduling instruction, and executing the abnormal strategy vector when the state mark indicates abnormality, wherein the abnormal strategy vector comprises the steps of sequencing all galvanic pile branches according to the sequence from large to small of the split coupling long tail loss degree to generate a priority checking sequence for a high-bias branch, and sequentially executing checking for filter blockage and pipeline retention cavities on the galvanic pile branches which are sequenced to the front according to the priority checking sequence, and executing insulation and equivalent resistance retesting on a public liquid supply manifold.

Description

Cloud-edge cooperation-based intelligent operation and maintenance scheduling optimization method for energy storage power station Technical Field The invention relates to the technical field of intelligent operation and maintenance of energy storage power stations, in particular to an intelligent operation and maintenance scheduling optimization method of an energy storage power station based on cloud edge cooperation. Background All-Vanadium Redox Flow Battery (VRFB) has become one of the core technical routes of gigawatt-level large-scale long-term energy storage power stations by virtue of the advantages of intrinsic safety, long cycle life, decoupling capacity power and the like. In order to increase energy density and simplify peripheral auxiliary facilities when constructing flow battery energy storage systems above megawatt level, the industry generally adopts a design architecture in which a plurality of galvanic pile modules share an electrolyte pipeline. In such architecture, several to tens of galvanic pile monomers are connected in parallel through a common liquid supply manifold, and electrolyte is distributed to each reactor core through a complex pipe network under the drive of a circulating pump so as to realize continuous charge-discharge circulation. However, the operating mechanism of flow batteries dictates that they face unique fluid and electrical coupling challenges. Since the vanadium electrolyte has good ionic conductivity, when it flows in a common manifold and branch connecting nodes of different potentials, a conductive loop is naturally formed, creating a bypass shunt current (i.e., leakage current). Meanwhile, the flow speed and the residence time of the electrolyte in each branch are not uniform due to the processing tolerance, the difference of the laying lengths of the pipelines and the difference of local flow resistance of the filter and the valve. The current monitoring means mainly rely on a Battery Management System (BMS) to collect macroscopic parameters such as total flow, inlet and outlet pressure difference, average voltage of a galvanic pile and the like, and an average hydraulic model or an average electric leakage model is generally adopted to estimate the efficiency loss of the system. This prior art based on mean statistics has significant monitoring dead zones. In practical engineering, fluid transportation of a pipe network often shows obvious asymmetry, namely, a long tail phenomenon that part of electrolyte stays in a pipeline for too long time exists. Since the consumption of the effective active material by the split current follows a first order decay law, the electrolyte in the long tail retention region is subject to exponentially amplified charge dissipation. The existing averaging monitoring method can not sense the hidden capacity phagocytosis caused by nonlinear coupling of fluid long tail retention and electric field shunt effect, so that the effective capacity and coulomb efficiency are irreversibly degraded under the condition that the overall parameters of the system are seemingly normal, and an operation and maintenance person is difficult to position a specific hydraulic bottleneck causing the attenuation through the existing data dimension. Disclosure of Invention The invention provides an intelligent operation and maintenance scheduling optimization method for an energy storage power station based on cloud edge cooperation, which solves the technical problems in the background technology. The invention provides an intelligent operation and maintenance scheduling optimization method of an energy storage power station based on cloud edge cooperation, which comprises the following steps: The edge monitoring end controls the circulating pump to execute electrolyte flow pulse excitation, collects fluid response data of each galvanic pile branch to construct electrolyte hydraulic transport delay distribution, and extracts transport long tail deviation values representing asymmetry of the hydraulic transport delay distribution; determining manifold bypass shunt strength reflecting that the conductive electrolyte forms a leakage path in the manifold according to the along-path potential gradient data of the common liquid supply manifold; constructing a shunt coupling long tail loss degree representing a nonlinear coupling relation between the shunt strength of the manifold bypass and the transport long tail deviation value so as to quantify effective capacity attenuation caused by long tail retention of electrolyte in a galvanic pile branch; And receiving an allowable efficiency degradation threshold value issued by a cloud management and control platform, carrying out normalized comparison on the module effective capacity loss rate calculated based on the split coupling long tail loss degree and the allowable efficiency degradation threshold value, and outputting an operation and maintenance scheduling instruction containing a priority investigat