Search

CN-121980841-A - Nuclear power cooling tower structure safety analysis intelligent body operation and maintenance method based on industrial physical AI engine

CN121980841ACN 121980841 ACN121980841 ACN 121980841ACN-121980841-A

Abstract

The invention discloses a nuclear power cooling tower structure safety analysis intelligent body operation and maintenance method based on an industrial physical AI engine, which relates to the technical field of nuclear power, and provides a method for respectively simulating concrete and double-layer bidirectional reinforcement layers by adopting a multi-layer shell unit with common nodes, simultaneously considering normal bias of each reinforcement layer relative to the middle surface of a tower wall, replacing the traditional method for modeling the three-dimensional entity of a reinforced concrete structure, reducing the number of degrees of freedom of a model on the premise of reflecting the composition of a real structure, in the aspect of material nonlinearity, respectively adopting a nonlinear damage model and a linear hardening model to respectively simulate the mechanical behaviors of concrete and steel bars, and realizing the integral of an independent nonlinear constitutive model of the reinforced concrete, the superposition calculation of a total tangential stiffness matrix and the separation and extraction of a reinforced concrete stress result, thereby improving the solving efficiency and the solving precision, accessing a nuclear power cooling tower AI intelligent operation and maintenance service flow by using a more efficient finite element industrial engine, and intuitively predicting the stability and damage condition of the cooling tower.

Inventors

  • TAN YU
  • ZHOU LU
  • Min Jiesheng

Assignees

  • 杭州摩斯智工科技有限公司

Dates

Publication Date
20260505
Application Date
20251217

Claims (10)

  1. 1. A nuclear power cooling tower structure safety analysis intelligent body operation and maintenance method based on an industrial physical AI engine is characterized by comprising the following steps: Step S1, drawing a geometric model comprising a ring base, a supporting structure and a tower wall according to a cooling tower body parameter or a design drawing, dividing a first-order grid, copying surface units at the same position of the tower wall for simulating a reinforcing steel bar layer, converting original surface units into second-order and adding a center node for simulating concrete; s2, selecting the least adverse working condition of the cooling tower, and directly calculating or adopting a finite element method to solve the temperature distribution of the inner surface and the outer surface of the wall of the tower; s3, taking the thickness change of the cylinder wall, the spacing and arrangement of the reinforcing steel bars and the nonlinear material characteristics into consideration, respectively carrying out constitutive model integration on a concrete nonlinear damage model and a reinforcing steel bar linear hardening model, solving a nonlinear thermosetting coupling balance equation of a reinforced concrete multi-layer shell model by adopting a Newton iteration method under the actions of dead weight, wind pressure load and temperature, and calculating a reinforced concrete stress field and a concrete damage factor of a cooling tower; Step S4, calculating a local stability coefficient according to the stress result, gradually increasing wind pressure load, iteratively carrying out the solving and calculating in the step S3 until the calculation diverges, obtaining critical wind pressure and wind speed, and calculating the overall stability coefficient of the cooling tower; And S5, packaging the physical simulation engine into a server in a Model Context Protocol (MCP) mode, and integrating the server into an artificial intelligent structure safety early warning module of the digital twin system of the cooling tower.
  2. 2. The nuclear power cooling tower structure safety analysis intelligent body operation and maintenance method based on the industrial physical AI engine of claim 1, wherein the step S1 specifically comprises the following steps: s101, drawing a tower cylinder wall, a supporting structure and a ring base geometric model according to a cooling tower body parameter or a design drawing; s102, dividing a finite element grid of a cooling tower, specifically, dividing a first-order grid for all geometric structures, copying 4 layers of surface units which are overlapped with original units and are in common nodes at the same position of the wall of the tower to simulate a reinforcing steel bar layer, converting the original surface units into second-order units, and adding a center node to simulate concrete.
  3. 3. The nuclear power cooling tower structure safety analysis intelligent body operation and maintenance method based on the industrial physical AI engine of claim 1, wherein the step S2 specifically comprises the following steps: S201, selecting the least adverse working condition of the cooling tower, calculating the temperature distribution of the inner surface and the outer surface of the wall of the tower, and acquiring the temperature distribution of the inner surface and the outer surface of the tower according to the actual condition of engineering from low to high according to simulation precision in three modes, wherein the temperature distribution is specifically as follows: Calculating according to the design air temperature distribution regulation and the internal and external surface temperature difference calculation formula of the inner side of a ventilation tube of a natural ventilation cooling tower in the current design specification; fitting the temperature distribution of the inner surface and the outer surface by adopting a least square method according to the actual monitoring data of the temperature measuring points arranged on the inner surface and the outer surface of the cooling tower barrel; and thirdly, solving a concrete structure temperature field of the cooling tower based on a three-dimensional shell unit steady-state heat transfer model by adopting a finite element method according to the internal and external air temperature and the convective heat transfer coefficient of the cooling tower.
  4. 4. The nuclear power cooling tower structure safety analysis intelligent body operation and maintenance method based on the industrial physical AI engine of claim 2, wherein the step S3 comprises the following steps: s301, calculating a load right end term of a balance equation of the wall of the reinforced concrete tower of the cooling tower under the action of dead weight and wind pressure; s302, calculating a temperature load right-end item according to the concrete temperature field obtained in the step S1, and using the temperature load right-end item for subsequent displacement prediction calculation; S303, carrying out displacement prediction based on an elastic hypothesis to obtain Newton iteration initial displacement; S304, simulating concrete by adopting a shell unit, simulating a double-layer bidirectional reinforcement layer by adopting a bias film unit, and respectively integrating the concrete nonlinear damage model and the reinforcement linear hardening model by adopting an analytic method in consideration of the temperature effect; S305, superposing the stress fields of the concrete and the steel bar layers, updating the displacement field, and calculating to be converged based on the Newton iteration method to obtain the displacement field of the wall of the cooling tower, the stress field of the concrete and the damage distribution result.
  5. 5. The method for safely analyzing the operation and maintenance of the intelligent body based on the nuclear power cooling tower structure of the industrial physical AI engine as set forth in claim 4, wherein said step S4 comprises the steps of: S401, calculating a local stability coefficient of the cooling tower according to the current design specification, wherein when the local stability coefficient is calculated, the circumferential stress of the cooling tower is the sum of the circumferential stress of concrete and the circumferential stress of the reinforcing steel bar layers on the inner surface and the outer surface, and the radial stress of the cooling tower is the sum of the radial stress of the concrete and the radial stress of the reinforcing steel bar layers on the inner surface and the outer surface; s402, amplifying the wind pressure function corresponding to the equivalent wind load standard value in the step S3 Multiplying, and solving nonlinear finite elements according to the method of the step S3; S403, increasing the amplification factor by step Halving, returning to step S401 until Less than a preset threshold value, and finally obtaining The overall stability coefficient of the cooling tower is obtained.
  6. 6. The method for safely analyzing the operation and maintenance of the intelligent agent based on the nuclear power cooling tower structure of the industrial physical AI engine as set forth in claim 5, wherein said step S5 comprises the steps of: S501, packaging the cooling tower stability and damage analysis physical simulation engine in the steps S2 to S4 into a server in an MCP mode, and exposing a callable interface for inputting parameters, executing nonlinear finite element simulation analysis and returning result data; S502, in a nuclear power cooling tower twin system, a LANGCHAIN application program development framework is adopted to build a structural safety analysis and early warning sub-module formed by a plurality of large language model intelligent bodies, so that an industrial AI engine for cooling tower stability and damage analysis is formed.
  7. 7. The method for safely analyzing the operation and maintenance of the intelligent agent based on the nuclear power cooling tower structure of the industrial physical AI engine as set forth in claim 6, wherein the step S5 comprises the following steps: the intelligent agent 1 is used for collecting and analyzing the cooling tower environment and load monitoring data, and is triggered by a question or a periodic instruction of a power plant manager; The intelligent agent 2 is used for analyzing nonlinear stability and damage of the cooling tower, receiving the environmental temperature and load information transmitted by the intelligent agent 1 and converting the environmental temperature and load information into physical simulation engine input parameters; and 3, carrying out safety evaluation and early warning on the cooling tower structure, and receiving the nonlinear stability and damage analysis result of the cooling tower transmitted by the intelligent body 2.
  8. 8. The nuclear power cooling tower structure safety analysis intelligent body operation and maintenance method based on the industrial physical AI engine of claim 7, wherein the intelligent body 1 further comprises a self-learning multi-source perception optimization module for carrying out real-time fusion and dynamic weight adjustment on multi-source monitoring data such as temperature, wind pressure, humidity, meteorological parameters and the like of the inner surface and the outer surface of a cooling tower barrel; The module identifies the drifting, abnormal fluctuation and data loss conditions of the measuring points by comparing the time sequence of the historical data and the current measuring results of each sensor, and automatically corrects the confidence weight of the corresponding sensor.
  9. 9. The nuclear power cooling tower structure safety analysis intelligent body operation and maintenance method based on the industrial physical AI engine of claim 7, wherein the intelligent body 2 further comprises an AI-physical collaborative solving module, and the module combines an approximate analysis method based on deep learning with a traditional finite element nonlinear computing method to form a hybrid solving mechanism with physical constraint and self-adaptive optimization capability.
  10. 10. The nuclear power cooling tower structure safety analysis intelligent agent operation and maintenance method based on the industrial physical AI engine of claim 7, wherein the intelligent agent 3 further comprises a group intelligent reinforcement learning collaborative optimization module for establishing a distributed decision and collaborative control mechanism among a plurality of cooling towers of a nuclear power plant so as to realize global dynamic optimization of structure safety.

Description

Nuclear power cooling tower structure safety analysis intelligent body operation and maintenance method based on industrial physical AI engine Technical Field The invention relates to the technical field of nuclear power, in particular to a nuclear power cooling tower structure safety analysis intelligent body operation and maintenance method based on an industrial physical AI engine. Background The natural ventilation cooling tower is an important equipment facility of a nuclear power plant, is mainly of a hyperbolic cylinder type reinforced concrete structure and is used for conducting waste heat of the nuclear power plant to the atmosphere in a hot air mode through heat exchange on the surface of a filler, compared with a seawater direct-current circulating cooling scheme, the natural ventilation cooling tower has the remarkable advantages of stable efficiency, simplicity in maintenance, capability of greatly reducing sea surface area, warm water drainage waste heat and the like, and is not applicable any more along with the evolution of a nuclear power stack, the improvement of the power generation efficiency of the nuclear power plant and the importance of China on environmental protection requirements, the established and constructed nuclear power cooling tower gradually develops to an ultra-large scale, namely, the water spraying area exceeds ten thousand square meters, and because the cooling tower belongs to a thin shell structure, the overall stability of the cooling tower under the wind load and the dead weight is a key evaluation index of the design operation and maintenance of the cooling tower, the calculation is generally performed on the basis of a linear elasticity assumption, and the empirical formula in the GB/T50102 industrial circulating water cooling design specification is adopted, and the empirical formula is not applicable any more along with the continuous improvement of the height and the diameter of the cooling tower, so that the industrial limited element stability analysis industrial simulation engine can take the nonlinear structural material of the practical reinforced concrete structure into consideration has important significance in the design and operation and maintenance monitoring of the ultra-large cooling tower. The tower body structure of the nuclear power cooling tower is generally composed of reinforced concrete, double-layer bidirectional reinforcing steel bar layers are arranged on the inner surface and the outer surface, and the reinforcing steel bar layers improve the ductility, the tensile strength and the compressive strength of the concrete through sharing and transferring external loads, so that the concrete is prevented from being pulled to crack. In the patent with the publication number CN118940563A, the wall thickness optimization calculation method of the hyperbolic natural ventilation cooling tower is disclosed, and the method comprises the steps of obtaining initial wall thickness data, a shell curve equation, structural material parameters and basic wind pressure of the cooling tower to be optimized, optimizing the initial wall thickness data according to the shell curve equation, the structural material parameters and the basic wind pressure to obtain middle wall thickness data, carrying out structure finite element calculation according to the middle wall thickness data to obtain shell internal force data, and optimizing the middle wall thickness data according to the shell internal force data to obtain target wall thickness data; In the aspect of nonlinear simulation analysis of a reinforced concrete structure of a cooling tower, the traditional simulation method is mainly divided into two modes of entity unit modeling and single-layer shell modeling, wherein the entity unit modeling is a separated method, namely, concrete is simulated by adopting entity units, and a common node surface unit or independently divided entity units are adopted to simulate a reinforced bar layer, so that the local mechanical behaviors such as structural stress, reinforced concrete interface friction, slippage and the like can be calculated more accurately, but the defects of difficult grid division and low calculation efficiency exist, the single-layer shell modeling mode usually simplifies reinforced concrete into a single-layer isotropic shell unit, the grid division is simple, the calculation efficiency is high, but the stress state of the reinforced concrete or the concrete cannot be distinguished independently, and the real nonlinear mechanical behavior of the reinforced concrete is difficult to simulate, so that the traditional finite element simulation method has great limitation in the design operation and maintenance of the ultra-large cooling tower or the intelligent integral stability monitoring system based on digital twin. Disclosure of Invention The invention provides a nuclear power cooling tower structure safety analysis intelligent body operation and