CN117592300A

CN117592300A - Digital twin modeling method, device, equipment and medium for novel generator set

Info

Publication number: CN117592300A
Application number: CN202311705096.7A
Authority: CN
Inventors: 胡阳; 张效宁; 朱国雄; 牛玉广; 王庆华; 秦天牧; 杜鸣; 江凯军; 张光明; 王孝伟; 吴恒运; 陈钢; 梁庚; 邱天
Original assignee: Beijing Huairou Laboratory; North China Electric Power University
Current assignee: Beijing Huairou Laboratory; North China Electric Power University
Priority date: 2023-12-12
Filing date: 2023-12-12
Publication date: 2024-02-23

Abstract

The application discloses a novel digital twin modeling method, device, equipment and medium of a generator set, relates to the technical field of computers, and the generator set comprises a boiler system, a steam turbine system and a molten salt system, wherein the method comprises the following steps: acquiring structural relations among the boiler system, the steam turbine system and the molten salt system; according to the structural relation, carrying out structural domain division on the unit to obtain a plurality of substructures; determining corresponding operation principles of the plurality of substructures under various working conditions; and carrying out digital twin modeling on the dominant thermodynamic dynamic characteristics of the unit power generation according to the corresponding operation principles of the plurality of substructures under various working conditions. Therefore, in the method, a fused salt system is introduced to carry out digital twin modeling, and the novel generator set is subjected to structural domain division and combines with the operation principle of various working conditions, so that the model after digital twin modeling is closer to the actual condition of the generator set, and the accuracy of the model is improved.

Description

Digital twin modeling method, device, equipment and medium for novel generator set

Technical Field

The application relates to the technical field of computers, in particular to a digital twin modeling method, device, equipment and medium of a novel generator set.

Background

Digital transformation for thermal power generating units is mostly carried out by adopting methods such as white box mechanism modeling, black box data modeling, ash box composite modeling and the like in the prior art, and digital twin modeling is carried out for the power generating process of the thermal power generating unit.

The white box mechanism modeling is used for deducing a mechanism model of the thermal power unit based on an energy balance equation, a mass balance equation, working medium characteristics and a constraint equation; black box data modeling, namely, based on thermal power generation operation data, establishing a data model of a thermal power unit by adopting a machine learning algorithm such as a neural network; ash box composite modeling is carried out, and based on a thermal power generation mechanism model, the mechanism model parameter identification is carried out by combining operation data.

However, the methods of white box mechanism modeling, black box data modeling, ash box composite modeling and the like do not consider a fused salt system, and modeling analysis cannot be performed on a boiler and fused salt coupled unit.

Disclosure of Invention

The application provides a digital twin modeling method, device, equipment and medium of a novel generator set, which can perform modeling analysis on a boiler, a steam turbine and a fused salt coupled set.

In order to achieve the above purpose, the present application adopts the following technical scheme:

in a first aspect, the present application provides a digital twin modeling method for a novel generator set, the set comprising a boiler system, a turbine system and a molten salt system, the method comprising:

acquiring structural relations among the boiler system, the steam turbine system and the molten salt system;

according to the structural relation, carrying out structural domain division on the unit to obtain a plurality of substructures;

determining corresponding operation principles of the plurality of substructures under various working conditions;

and carrying out digital twin modeling on the generating characteristics of the unit according to the corresponding operation principles of the plurality of substructures under various working conditions.

Optionally, the boiler system comprises a boiler, an economizer, a molten salt heat exchanger, a boiler reheater, a boiler superheater, a first valve, a first hydraulic/electric pump, a first conduit, etc.; the turbine system comprises high-pressure water supply, a deaerator, low-pressure water supply, a low-pressure cylinder, a medium-pressure cylinder, a high-pressure cylinder, a second valve, a second hydraulic/electric pump, a second pipeline and the like; the molten salt system comprises a molten salt superheater, a molten salt evaporator, a molten salt preheater, a high-temperature molten salt tank, a steam-molten salt heat exchanger, a low-temperature molten salt tank, a third valve, a third hydraulic/electric pump, a third pipeline and the like.

Optionally, the performing structural domain division on the unit according to the structural relationship to obtain a plurality of substructures includes:

and carrying out structural domain division on the unit according to the directed graph node relation matrix corresponding to the structural relation to obtain a plurality of submatrices, wherein the submatrices are used for representing a plurality of substructures of the unit.

Optionally, the performing structural domain division on the unit according to the directed graph node relation matrix corresponding to the structural relation includes:

and processing the directed graph node relation matrix corresponding to the structural relation by utilizing a directed graph node clustering algorithm so as to divide the structural domain of the unit.

Optionally, the method further comprises:

acquiring corresponding real operation data under the various working conditions in the generating process of the unit;

the digital twin modeling of the generating characteristics of the unit is performed according to the corresponding operation principles of the plurality of substructures under various working conditions, including:

according to the corresponding operation principle and real operation data of the plurality of substructures under various working conditions, digital twin modeling is performed on the generating characteristics of the unit by adopting an algebraic or proper order differential equation mechanism model, a data-driven machine learning agent model and a multiple combination mode thereof according to different task demands.

In a second aspect, the present application provides a digital twin modeling apparatus for a novel genset, the genset comprising a boiler system, a steam turbine system and a molten salt system, the apparatus comprising:

the acquisition module is used for acquiring the structural relationship among the boiler system, the steam turbine system and the molten salt system;

the division module is used for dividing structural domains of the unit according to the structural relation to obtain a plurality of substructures;

the determining module is used for determining the corresponding operation principle of the plurality of substructures under various working conditions;

and the modeling module is used for carrying out digital twin modeling on the generating characteristics of the unit according to the corresponding operation principles of the plurality of substructures under various working conditions.

Optionally, the dividing module is specifically configured to divide the structural domain of the unit according to a directed graph node relationship matrix corresponding to the structural relationship, so as to obtain a plurality of submatrices, where the plurality of submatrices are used to characterize a plurality of substructures of the unit.

Optionally, the dividing module is specifically configured to process a directed graph node relationship matrix corresponding to the structural relationship by using a directed graph node clustering algorithm, so as to divide a structural domain of the unit.

Optionally, the obtaining module is further configured to obtain real operation data corresponding to the multiple working conditions in the generating process of the unit;

the modeling module is specifically configured to perform digital twin modeling on the generating characteristics of the unit according to the operation principles of the multiple substructures under multiple working conditions.

In a third aspect, the present application provides a computer readable storage medium for storing a computer program for performing the method according to any one of the first aspects.

In a fourth aspect, the present application provides a computer program product having a computer program stored thereon, which, when executed by a processing device, implements a method according to any of the first aspects.

According to the technical content, the technical scheme of the application has the following beneficial effects:

the application provides a digital twin modeling method of a novel generator set, wherein the generator set comprises a boiler system, a steam turbine system and a molten salt system, and the method comprises the following steps: acquiring structural relations among the boiler system, the steam turbine system and the molten salt system; according to the structural relation, carrying out structural domain division on the unit to obtain a plurality of substructures; determining corresponding operation principles of the plurality of substructures under various working conditions; and carrying out digital twin modeling on the generating characteristics of the unit according to the corresponding operation principles of the plurality of substructures under various working conditions. Therefore, in the method, a fused salt system is introduced to perform data modeling, and the unit is subjected to structural domain division and combines with the operation principle of various working conditions, so that the model subjected to digital twin modeling is more similar to the actual condition of the unit, and the accuracy of the model is improved.

It should be appreciated that the description of technical features, aspects, benefits or similar language in this application does not imply that all of the features and advantages may be realized with any single embodiment. Conversely, it should be understood that the description of features or advantages is intended to include, in at least one embodiment, the particular features, aspects, or advantages. Therefore, the description of technical features, technical solutions or advantageous effects in this specification does not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions and advantageous effects described in the present embodiment may also be combined in any appropriate manner. Those of skill in the art will appreciate that an embodiment may be implemented without one or more particular features, aspects, or benefits of a particular embodiment. In other embodiments, additional features and advantages may be recognized in certain embodiments that may not be present in all embodiments.

Drawings

Fig. 1 is a schematic diagram of a thermal power generating unit according to an embodiment of the present application;

FIG. 2 is a flowchart of a digital twin modeling method for a novel generator set according to an embodiment of the present disclosure;

FIG. 3 is a schematic diagram of digital twin modeling provided in an embodiment of the present application;

fig. 4 is a schematic diagram of a software and hardware structure for implementing a digital twin model according to an embodiment of the present application;

fig. 5 is a schematic diagram of a digital twin modeling device of a novel generator set according to an embodiment of the present application.

Detailed Description

The terms first, second, third and the like in the description and in the claims and drawings are used for distinguishing between different objects and not for limiting the specified sequence.

In the embodiments of the present application, words such as "exemplary" or "such as" are used to mean serving as examples, illustrations, or descriptions. Any embodiment or design described herein as "exemplary" or "for example" should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present related concepts in a concrete fashion.

The cleanliness of the power supply is an important characteristic of a novel power system, and the energy provided by the novel power system has an increasing duty ratio in power supply in China. The characteristic of new energy power generation operation determines the defects of poor stability and large impact on a power grid.

The demand of safe operation reliability of the power grid is more urgent, so that a thermal power generating unit with stable electric quantity supply is required to gradually change to frequency modulation, standby and emergency capacity service providers. The flexibility of the thermal power generating unit is improved, and the thermal power generating unit is an important development direction.

The existing medium-small capacity unit (300 MW level and 600MW level) has long operation time, low load efficiency and relatively poor equipment performance, is hopeful to become a main power unit for load adjustment through flexible transformation, and provides power support for a power grid; the high-capacity 1000 MW-level coal-fired unit has high efficiency and good equipment performance, can run under high load, and provides electric quantity support for a power grid.

The embodiment of the application provides a novel thermal power generating unit, as shown in fig. 1, this diagram is the schematic diagram of a thermal power generating unit that this application embodiment provided, and this thermal power generating unit includes: boiler 101, economizer 102, molten salt heat exchanger 103, boiler reheater 104, boiler superheater 105, high-feed water 106, deaerator 107, low-feed water 108, low-pressure cylinder 109, medium-pressure cylinder 110, high-pressure cylinder 111, molten salt superheater 112, molten salt evaporator 113, molten salt preheater 114, high-temperature molten salt tank 115, steam-molten salt heat exchanger 116, low-temperature molten salt tank 117.

The steam from the high pressure cylinder 111 of the steam turbine heats the molten salt from the low temperature molten salt storage tank 117 through the steam-molten salt heat exchanger 116 to achieve heat storage.

After passing through the molten salt preheater 114, the molten salt evaporator 113 and the molten salt superheater 112 in sequence, the circulating water from the high-pressure heater 106 exchanges heat with the high-temperature molten salt to obtain high-temperature high-pressure steam meeting certain conditions, and the high-temperature high-pressure steam is injected into the middle part of the high-pressure cylinder 111 of the steam turbine to perform work.

Part of steam after working from the high-pressure cylinder 111 of the steam turbine can be heated again through the fused salt reheater and then injected into the middle part of the medium-pressure cylinder 110 of the steam turbine to do work, so that the heat release of fused salt heat storage is realized, and steam is generated to improve the load of the steam turbine.

When the unit reduces load, main steam of the boiler and hearth smoke are extracted according to a rule; if the main steam A2 is extracted, the molten salt is heated by a molten salt superheater 112 and cooled in a low-temperature molten salt tank 117, and steam B2 at the outlet of the molten salt superheater 112 enters a medium-pressure cylinder 110 to do work after passing through a hearth reheater; if the hearth flue gas is extracted, the low-temperature molten salt tank 117 is heated by the hearth tail molten salt heat exchanger 103 to cool the salt.

When the unit lifts load, the high-temperature molten salt tank 115 sequentially passes through the molten salt superheater 112, the molten salt evaporator 113 and the molten salt preheater 114 to release heat, and the high-added water 106 tributaries sequentially pass through the molten salt preheater 114, the molten salt evaporator 113 and the molten salt superheater 112 to generate main steam and enter the high-pressure cylinder 111 to do work.

At present, various modeling modes exist, such as white box mechanism modeling, black box data modeling and ash box composite modeling. The white box mechanism modeling is high in model complexity, high in model parameter identification difficulty and inconsistent in model characteristics with reality, and is mostly used for fine simulation or training in the thermal power generation process; the black box data modeling only considers the external characteristics of input and output, has unknown internal mechanism characteristics, depends on modeling data and algorithms, needs updating and maintenance, has a large number of data models, and is mostly used for state monitoring and operation and maintenance in the thermal power generation process; the ash box composite modeling is generally only suitable for linear system modeling, the model order is low, the full-working-condition complex nonlinear approximation performance is limited, the optimal approximation characteristic and the actual characteristic of the model have deviation, and meanwhile, the model parameter identification difficulty is high, so that the method is mainly used for designing thermal power generation process controllers. It can be seen that the existing modeling mode only models a thermal power unit, and does not model a digital twin unit coupled with a fused salt system.

In view of this, the embodiment of the application provides a digital twin modeling method of a novel generator set, where the set to which the digital twin modeling method is directed includes a boiler system and a molten salt system, and specifically, the method includes: obtaining a structural relation between a boiler system and a molten salt system, carrying out structural domain division on a unit according to the structural relation to obtain a plurality of substructures, determining corresponding operation principles of the substructures under various working conditions, and finally carrying out digital twin modeling on the thermodynamic dynamic characteristics of the generator set power generation dominance based on the corresponding operation principles of the substructures under the various working conditions. According to the method, a fused salt system is introduced to perform data modeling, and the unit is subjected to structural domain division and combines with the operation principle of various working conditions, so that a model subjected to digital twin modeling is more similar to the actual condition of the unit, and the accuracy of the model is improved.

In order to facilitate understanding, the digital twin modeling method of the novel generator set provided by the embodiment of the application is described below with reference to the accompanying drawings.

As shown in fig. 2, the figure is a flowchart of a digital twin modeling method of a novel generator set provided in an embodiment of the present application, where the set includes a boiler system, a steam turbine system, and a molten salt system, and in some examples, the set may be as shown in fig. 1 above, and the method includes:

s201: and obtaining the structural relationship among the boiler system, the steam turbine system and the molten salt system.

The boiler system may be a system comprising the boiler 101, the economizer 102, the molten salt heat exchanger 103, the boiler reheater 104, the boiler superheater 105 in fig. 1, and the steam turbine system may comprise the high feed water 106, the deaerator 107, the low feed water 108, the low pressure cylinder 109, the medium pressure cylinder 110, the high pressure cylinder 111 in fig. 1; the molten salt system may include the molten salt superheater 112, the molten salt evaporator 113, the molten salt preheater 114, the high temperature molten salt tank 115, the steam-molten salt heat exchanger 116, the low temperature molten salt tank 117 in fig. 1. The boiler 101 may be a drum boiler or a once-through boiler, among others.

The structural relationship between the acquisition boiler system, the steam turbine system and the molten salt system may be a connection relationship between the above-described respective components. The molten salt system can be a high-temperature molten salt heat storage system.

S202: and according to the structural relation, carrying out structural domain division on the novel unit to obtain a plurality of substructures.

In some examples, after determining the structural relationships of the units, the units may be domain partitioned to obtain a plurality of substructures. Among them, the components shown in fig. 1 may be referred to as a structure, and a plurality of components may be composed into a substructure, for example, a substructure 1 may be obtained through domain division, and the substructure 1 may include a high-temperature molten salt tank 115, a steam-molten salt heat exchanger 116, and a low-temperature molten salt tank 117.

It should be noted that the above sub-structures are only exemplary, and those skilled in the art may divide the structural domain of the unit to obtain different sub-structures based on actual needs.

In some embodiments, a corresponding directed graph node relation matrix may be determined based on the above structural relation, and then the unit is subjected to domain division based on the directed graph node relation matrix to obtain a plurality of submatrices, where the plurality of submatrices are used to characterize a plurality of substructures of the unit, i.e., one submatrix characterizes one substructure.

The method comprises the steps of establishing a system directed graph of a boiler system-molten salt system coupling unit based on functional-physical combination, and dividing structural domains of node grouping based on a directed graph node relation matrix to obtain a plurality of submatrices. In some examples, a directed graph G= (V, W) of the boiler-molten salt coupling unit can be obtained by adopting directed graph nodes to establish functional and physical connection relations among subsystems (or devices) under various load conditions based on a boiler system, a molten salt system and a coupling mechanism of the boiler system and the molten salt system. Where node vector V represents a subsystem (or device) including, but not limited to, a boiler, economizer, molten salt heat exchanger, boiler reheater, boiler superheater, high feed water, deaerator, low feed water, low pressure cylinder, medium pressure cylinder, high pressure cylinder, molten salt superheater, molten salt evaporator, molten salt preheater, high temperature molten salt tank, steam-molten salt heat exchanger, low temperature molten salt tank, valve, hydraulic/electric pump, piping, etc. Initially, each node vector v= [ V ] may be set based on the operating principle of the unit and a priori knowledge ₁ ,V ₂ ,…,V _i ,…,V _n ]. The directed graph relation matrix W represents the connection relation between the nodes (for example, under the operation condition, when the connection relation exists, the value is 1, and when the connection relation does not exist, the value is 0), and the directed graph relation matrix W has:

wherein W is _ij ＝W _ji Representing the connection relationship between nodes i and j.

In some examples, a directed graph node clustering algorithm may be utilized to process a directed graph node relationship matrix corresponding to the structural relationship to domain partition the unit. The directed graph node grouping algorithm comprises, but is not limited to, a Louvain algorithm and a Tarjan algorithm. For example, the number of the cells to be processed,grouping different subsystems by using directed graph node grouping algorithm, wherein each group can be called a structural domain (or a substructure) D _p (p=1, 2, …, m; m is the number of domains divided by the coupled power generation system, i.e., how many substructures).

The initial domain division can be determined according to the operation principle and priori knowledge of the unit, after multiple iterations are applied, a domain division knowledge base can be constructed for coupled power generation systems of different types (including but not limited to natural circulation boilers, supercritical/ultra supercritical direct current boilers and the like) and different capacity grades (including but not limited to 300 megawatts, 350 megawatts, 600 megawatts and the like), and the domain division can be performed based on the domain division knowledge base.

In some examples, each domain represents a subsystem or device that is functionally or physically connected to each other under all-condition operating conditions of the coupled power generation system, capable of embodying the rationality of modeling in accordance with that domain. On the structure domain division result obtained by adopting Louvain directed graph clustering, the initial structure domain division can be determined according to the operation principle and priori knowledge of a unit, after iteration is applied for a plurality of times, a structure domain division knowledge base can be constructed for a 350MW supercritical direct-current furnace coal-fired thermal power unit-high-temperature fused salt heat storage coupling flexible power generation system, and the structure domain division can be carried out based on the structure domain division knowledge base.

For example, for a 350MW supercritical direct-current furnace coal-fired thermal power unit-high-temperature molten salt heat storage coupling flexible power generation system, the direct-current furnace unit (i.e. a boiler system) is divided into three structural domains of a pulverizing system, a boiler steam-water system, a steam turbine system and the like, and the high-temperature molten salt heat storage system (i.e. a molten salt system) is divided into structural domains of a high-temperature molten salt tank, a low-temperature molten salt tank, a hearth flue gas-molten salt heat exchanger, a molten salt transportation pipeline system, a high-temperature molten salt-steam-water system, a low-temperature molten salt-superheated steam heat exchanger, a molten salt transportation pipeline system and the like.

S203: and determining the corresponding operation principle of the plurality of substructures under various working conditions.

In some examples, model derivation may be based on parameters of subsystems or devices included within the sub-structure. For example, for subsystems or devices with a mature mechanism modeling basis (including but not limited to pulverizing systems, boiler soda systems, turbine systems, high/low temperature molten salt heat storage systems, molten salt heat storage soda systems, water or steam pipes, etc.), white-box mechanism models of appropriate order may be employed, for subsystems or devices that are complex non-linear and do not have a mature mechanism modeling basis (including but not limited to furnace combustion systems), black-box parametric models (e.g., active derived regression (ARX) models, subspace recognition models, partial least squares linear regression models, logistic regression models, piecewise affine PWA models, piecewise affine active derived regression PWAX models, etc. dynamic or static parametric models may be employed. According to the functional and physical connection relations among different subsystems or devices, the white box mechanism model and the black box parameter model adopt a multi-stage series connection, a multi-stage parallel connection or a multi-stage mixed series/parallel connection and other connection modes, and on the basis, the models of the different subsystems or devices are subjected to multi-scale dynamic integrated deduction to obtain the modularized integrated parameter model corresponding to each structural domain.

In some examples, the pulverizing system mechanism parameter model is as follows:

wherein u is _B Is a fuel quantity instruction, and the unit is kg/s; r is (r) _B ' is the actual amount of coal entering the mill in kg/s; τ is the delay time in s; r is (r) _B For entering the pulverized coal amount of the boiler, the unit is kg/s; m is M _coal The unit of the coal quantity in the mill is kg; c _B Is the basic output coefficient of the mill; f (f) _H The correction coefficient is coal grindability; f (f) _W The water content correction coefficient is the water content correction coefficient of the coal; f (f) _R The coefficient is corrected for the fineness of the pulverized coal.

The mechanism parameter model of the boiler steam-water system is as follows:

wherein,the unit is kg/m for the average point working medium density of the heated section ³ ；D _fw The unit is kg/s for water supply flow; d (D) _s The unit is kg/s for the outlet steam flow of the superheater; v _t The unit is m for the total internal volume of the economizer, the water-cooled wall and the steam-water separator ³ ；/>The unit is kJ/kg of the internal energy of the working medium at the average point of the heated section; h is a _fw The unit is kJ/kg for the enthalpy value of the water supply; h is a _s The unit of the specific enthalpy of steam at the outlet of the superheater is kJ/kg, h _s ＝u _s +p _s /ρ _s ；k ₀ The unit of the gain of heat absorption for the heated section is kJ/kg; p is p _m Steam pressure of the steam-water separator is expressed as MPa; p is p _s The unit is the outlet steam pressure of the superheater and is MPa; sigma is the resistance coefficient of the superheater pipeline; d (D) _sw The unit is kg/s for the total flow of equivalent heat-reducing water; d (D) _sw ＝D ₁ G ₂ G ₃ G ₄ +D ₂ G ₃ G ₄ +D ₃ G ₄ ，D ₁ 、D ₂ 、D ₃ Respectively representing the flow of primary, secondary and tertiary de-heating water with the unit of kg/s, G ₂ 、G ₃ 、G ₄ Respectively representing two-stage, three-stage and final-stage superheater mathematical models; d (D) _st The unit is kg/s for the steam flow of the inlet of the steam turbine, namely the main steam flow; h is a _st The unit is kJ/kg of the steam turbine inlet specific enthalpy, namely main steam specific enthalpy; p is p _s ＝p _st ，p _st The unit is MPa of steam pressure at the inlet of the steam turbine, namely main steam pressure; s is(s) ₁ 、s ₂ Is a dynamic parameter; ρ _m Is the steam density of the steam-water separator, and the unit is kg/m ³ ；u _m The unit is kJ/kg of the internal energy of steam at the steam-water separator; q (Q) ₀ Heat absorbed for the steam flow through the superheater tubes in kJ; Δp is the differential pressure created by the flow of steam through the superheater tubes, Δp=p _m -p _st The unit is MPa; z ₁ 、z ₂ Is a coefficient related to the resistance of the pipeline and the specific heat of the gas; d (D) _m0 Is the initial mass flow of steam; v ₀ For specific volume of steam, g (. Cndot.) is expressed as p, which is related to pressure and temperature of steam _m As a function of the argument.

The turbine system mechanism parameter model is as follows:

wherein ρ is _st The density of steam at the inlet of the steam turbine is kg/m ³ ；u _t The opening of the steam turbine regulating door can be represented by percentage; lambda and alpha are coefficients; n (N) _e The unit is MW for unit load; k (k) ₂ Is the gain of the steam turbine.

In some examples, a simplified low-order Bai Xiang mechanism parameter model is adopted for the structural domains of a high-temperature molten salt heat storage system (such as 5 KW) such as a high-temperature molten salt tank, a low-temperature molten salt tank, a hearth flue gas-molten salt heat exchanger, a molten salt transportation pipeline system, a high-temperature molten salt-steam-water system, a low-temperature molten salt-superheated steam heat exchanger, a molten salt transportation pipeline system and the like.

The high-temperature molten salt tank and the low-temperature molten salt tank have the same model structure, for example, a molten salt tank mechanism parameter model is as follows:

wherein m is _salt,in The unit of the mass flow is kg/s of the inlet mass flow of the molten salt heat storage tank; t (T) _salt,in The unit is the temperature of molten salt at the inlet of the molten salt heat storage tank; t (T) _salt,tank The unit is the temperature of the molten salt heat storage tank; Δt is a infinitesimal time period, and the unit is s; m is M _salt,tank The initial mass of the molten salt heat storage tank is kg.

The high-temperature molten salt-steam-water system comprises a preheater, an evaporator and a superheater, and a mechanism parameter model of the high-temperature molten salt-steam-water system is as follows:

wherein c _salt The constant pressure specific heat capacity of molten salt is kJ/(kg·DEG C); h is a _eva,in The unit of the enthalpy value of the molten salt entering the evaporator is kJ/kg; h is a _eva,out The unit of the enthalpy value of the molten salt outlet of the evaporator is kJ/kg; m is m _salt,eva The unit is kg/s for the mass flow of the molten salt in the evaporator; m is M _salt,eva The unit is kg of the mass of molten salt in the evaporator; q (Q) _salt,eva The unit of the heat transferred from the molten salt to the pipe wall is kJ; t (T) _salt,eva The temperature of the molten salt in the evaporator is given in c.

c _pipe The specific heat capacity of the evaporator pipeline is kJ/(kg. DEG C); m is M _pipe,eva The mass of the pipeline in the evaporator is kg; t (T) _pipe,eva The temperature of the inner wall of the evaporator tube is expressed as the unit of the temperature; q (Q) _s,eva The unit of heat transfer quantity from the molten salt in the evaporator to the inner wall of the pipe is J; q (Q) _w,eva The unit of heat transfer quantity from the outer wall of the evaporator tube to the water side outside the tube is J; a is that _i,eva 、A _o,eva The areas of the inner wall and the outer wall of the tube are respectively m ² ；T _w,eva The temperature of the water side of the outer wall of the evaporator pipeline is expressed as the unit of the temperature; u (U) _i,eva 、U _o,eva The heat convection coefficients between the molten salt in the evaporator and the inner wall of the tube and between the outer wall of the tube and the outside water are respectively expressed in W/m ² ·K。

M _w,eva The mass of water in the evaporator is kg; m is m _fw,eva The mass flow of the feed water in the evaporator is kg/s; m is m _cond,eva Mass flow in kg/s for the portion of the evaporator where the steam is cooled by the feedwater entering the evaporator; m is m _zq,eva The unit is kg/s for the evaporation amount of steam in the evaporator; m is m _d,eva The unit is kg/s for the mass flow of the sewage.

M _zq,eva The unit is kg of the mass of the steam in the evaporator; m is m _zq,eva,o The unit is kg/s for the outlet steam flow of the evaporator; h is a _fw,eva The enthalpy value of the feed water to the evaporator is kJ/kg; h is a _zq,eva Is steamedThe enthalpy of steam in the generator is kJ/kg; v (V) _v,eva 、V _w,eva The unit is m for the volume of steam and water in the evaporator ³ ；ρ _v,eva 、ρ _w,eva The density of steam and water in the evaporator is kg/m ³ 。

h _w,eva The enthalpy value of water in the evaporator is kJ/kg; p (P) _d For continuous blowdown, it can be characterized by percentage.

P _w,eva The unit is MPa for the water side pressure in the evaporator; p (P) _w,sup The unit is MPa for the pressure of the reclaimed water side of the superheater; zeta type toy _eva The admittance coefficient of the evaporator is related to the valve characteristics and is expressed in Pa.s ² /kg ² 。

The mechanism parameter model of the low-temperature molten salt-superheated steam heat exchanger is as follows:

wherein V is _zq,sup Is the volume of steam in the superheater, and has the unit of m ³ ；m _zq,sup,i 、m _zq,sup,o For the mass flow rate of the steam entering and exiting the superheater, the unit is kg/s; ρ _zq,sup,i 、ρ _zq,sup,o For the density of the steam entering and exiting the superheater, the unit is kg/m ³ 。

h _zq,sup,i 、h _zq,sup,o The unit of the enthalpy value of steam entering and exiting the superheater is kJ/kg; q (Q) _zq,sup The heat absorbed by the steam in the superheater by heating the pipeline is expressed in kJ.

ξ _sup Is the flow resistance or linear admittance of steam in the superheater, the unit is Pa.s ² /kg ² ；P _zq,sup,i Is the steam inlet pressure in the superheater, and has the unit of MPa; p (P) _zq,sup Is the pressure of steam in the superheater, in MPa.

c _salt The constant pressure specific heat capacity of molten salt is kJ/(kg·DEG C); m is M _salt,sup The unit is kg of molten salt mass flowing through the superheater; t (T) _salt,sup Unit for molten salt temperature through superheaterIs the temperature; m is m _salt,sup The unit is kg/s for the mass flow of molten salt through the superheater; q (Q) _salt,sup The heat released to the tube wall for the molten salt in the superheater is given in kJ.

The mechanism parameter model structure of the low-temperature fused salt-superheated steam heat exchanger can be adopted by the hearth flue gas-fused salt heat exchanger, and the description is omitted here.

In some examples, the molten salt transport conduit system employs a multi-segment conduit lumped parameter model series approach to modeling of long distance conduits. The mechanism parameter model of the single-section pipeline is as follows:

wherein C is _salt,pipe The heat capacity of molten salt in a single-section pipeline is expressed as J/°C; t (T) _salt,pipe,in 、T _{salt,pipe,out} The input temperature and the output temperature of molten salt in a single-section pipeline are given in the unit of DEG C; t (T) _am,pipe The unit is the ambient temperature of the pipeline; c _salt The specific heat capacity of the molten salt is kJ/(kg. DEG C); alpha _salt,pipe Is the convective heat transfer coefficient (J/(m) ² ·℃·s))；Q _salt,pipe The unit is kg/s for the mass flow of molten salt in the pipeline; s is S _salt,pipe Is the heat transfer area of the pipeline, and the unit is m ² 。

The modularized mechanism parameter modeling of the substructure of the powder making system, the boiler steam-water system, the steam turbine system and the like of the boiler system is introduced respectively.

In addition to the mechanism parametric model, if there is a domain that has complex nonlinear characteristics that are difficult to model, a black box parameterized model (including but not limited to an active regression ARX model, a piecewise affine PWA model, a piecewise affine active regression PWAX model, a subspace identification model, a Logistic regression model, etc.) may be used instead of the mechanism parametric model.

Based on lumped parameter modeling theory, the adopted low-order mechanism parameter model of different structural domains is a linear or nonlinear differential equation, and then the modularized integrated parameter model L corresponding to each structural domain _p (p=1, 2, …, m; m is the number of domains divided by the coupled power generation system, i.e., the number of substructures) can be uniformly expressed by the following structure:

wherein A, B, C, D is a linear or nonlinear appropriate dimension matrix; k is the timing index of the measured signal in the sampling period T; y (k) may be a single output variable, multiple output vector.

It can be seen that by means of the above model, the operation principle corresponding to the plurality of substructures can be characterized.

In some examples, the sub-structure D may also be based on _p Corresponding integrated parameter model L _p Extracting characterizable D _p Differential dynamic regression vector R (k) = [ y ] of operating characteristics under full operating conditions ^T (k-1),y ^T (k-2),…y ^T (k-n _a ),u ^T (k-1),u ^T (k-2),…,u ^T (k-n _b )] ^T Wherein n is _a 、n _b The method is characterized in that the delay orders of system output and input are respectively the convex division is carried out on the high-dimensional nonlinear operation space formed by the R (k), so that the operation principle of the substructures under various working conditions can be obtained, and then the operation principle of each substructures under various working conditions is obtained based on a similar mode.

The convex partition can adopt various methods, including but not limited to R (k) regression vector direct high-dimensional clustering partition, R (k) regression vector extraction and high-dimensional clustering partition thereof, specific duration R (k) regression matrix clustering partition and the like. For the high-dimensional clustering result, a hyperplane partitioning method (including but not limited to a support vector machine, a soft interval support vector machine and the like) can be adopted to obtain a hyperplane equation capable of representing the boundaries of different sub-operation domains. Based on regression vector R (k) and hyperplane equations, a decision can be madeAnd the moment k is coupled with the operation domain where the power generation system is positioned. Finally, substructure D _p Corresponding integrated parameter model L _p,l (l＝1,2,…,S _p ) Obtaining S _p And the sub-operation domain, namely, the operation principle of the sub-structure corresponding to the sub-structure under various working conditions is obtained.

S204: and carrying out digital twin modeling on the power generation characteristics of the novel unit according to the corresponding operation principle of the plurality of substructures under various working conditions.

When digital twin modeling is performed, modeling can be performed by adopting one or more combination modes of algebraic or differential equation mechanism models and data-driven machine learning proxy models.

In some examples, after determining the corresponding operating principles of the substructure under various operating conditions, digital twin modeling may be performed on the dominant thermodynamic dynamics of the generation of the novel unit.

In other embodiments, the digital twin modeling can be performed on the thermodynamic dynamic characteristics of the novel unit power generation dominant thermodynamic characteristics based on the operation principle corresponding to the plurality of substructures under the plurality of working conditions and the real operation data.

The mixed semi-parameter modeling of the boiler system and the molten salt system mainly comprises a model L based on each sub-operation domain _p,l Parameter identification and machine learning-based deviation dynamic neural network compensation modeling of state space equations.

In some examples, the linear or nonlinear model L in the first running domain is obtained by taking the regression vector R (k) as input and y (k) as output, and adopting methods such as numerical methods (including but not limited to equation error parameter identification, gradient correction parameter identification, probability density approximation parameter identification, least squares identification and the like) or heuristic intelligent optimization (including but not limited to genetic algorithm, particle swarm algorithm and simulated annealing algorithm) and the like _p,l Parameters, establishing a model of each sub-operation domain.

For example, define(/>For model output value), bias-compensated regression vector R _e (k)＝[e ^T (k-1),e ^T (k-2),…,e ^T (k-n _e ),y ^T (k-1),y ^T (k-2),…,y ^T (k-n _a ),u ^T (k-1),u ^T (k-2),…,u ^T (k-n _b )] ^T (n _e An autoregressive order of e (k). By R _e (k) For input, e (k) is output, e is established using a machine learning algorithm (including but not limited to a recurrent neural network, a self-encoder-recurrent neural network, etc. deep neural network machine learning method) _y (k) Multiple input-multiple output bias dynamic compensation model of (2) which can realize the dynamic compensation of all S in the p-th structural domain _p Unified compensation of individual run domain models. If care is required for the measurable state x _ob X can be _ob As an output, at e _y (k) The deviation dynamic compensation of the measurable state is realized, and the deviation dynamic compensation is fed back to a state equation x (k+1); setting the bias compensation term of the undetectable state to 0, introducing from e _y (k) Feedback measurable state x _ob Dynamic compensation term e of deviation of (2) _xob (k) Form the dynamic compensation term e of the deviation of the state equation _x (k)。

Wherein l=1, 2, …, S _p 。

Then, aiming at the first operation domain of the p-th structure domain, each sub-operation domain model L with deviation dynamic compensation is obtained _p,l,DC The method has arbitrary approximation capability to actual running dynamics. Integrated S _p Model L corresponding to sub-operation domain _p,l,DC Obtaining a mixed semi-parametric model L of the p-th modularized structural domain _p,DC I.e. a digital twin model of the p-th modular domain.

Next, p (p=1, 2, …, m) hybrid semi-parametric models L are generated according to the functional or physical connection of p (p=1, 2, …, m) modular domains _p,DC And carrying out multi-scale dynamic integration deduction to obtain an integral digital twin body model of the power generation system coupled with the boiler system and the molten salt system. The physical meaning of part or even all of the state quantity of the model can be known, and the model is helpful for control design, state monitoring and the like.

As shown in fig. 3, the figure is a digital twin modeling schematic provided in an embodiment of the present application.

In the method, the structural relation between a boiler system and a molten salt system and the respective operation mechanism are determined, then the directed graph nodes of the subsystem or equipment are determined, clustering and domain division are carried out, then a white box mechanism parameter model and a black box parameter model are introduced to obtain a domain modularized parameter model, then a differential dynamic regression vector is defined, then a differential dynamic space is divided convexly, then the domain modularized model parameter optimization identification and the performance evaluation are carried out, the different domain modularized models are carried out, and finally the digital twin model of the boiler system-the molten salt system is obtained.

Fig. 4 is a schematic diagram of a software and hardware structure for implementing a digital twin model according to an embodiment of the present application.

The boiler system-molten salt system coupling unit comprises an intelligent operation management system 401, a DCS operation control system 402 and an intelligent sensing and data acquisition and transmission 403. The digital twin system software architecture comprises a storage module 404, a data thread management 405, a data twin model library 406, a service application and visualization interface 407, a full excitation simulation system and a DCS control system 408.

In some embodiments, the software architecture of the digital twin system may also include a real-time data acquisition module, a storage module, a digital twin model creation module, a digital twin model update module, a digital twin model library management module, a digital twin model fusion module, a multi-scenario digital twin operation module, and a business application module facing operation control or management.

The real-time data acquisition module is mainly used for real-time operation data acquisition, communication and treatment of the power generation system. The storage module is used for storing and calling, and is convenient for the access of the digital twin modeling to the operation data. The digital twin model creation module and the digital twin model update module are mainly used for digital twin modeling, adaptive updating and maintenance of a physical subsystem or equipment. The digital twin body model library management module is mainly used for the management of classified retrieval, calling and the like of different subsystem models. The digital twin model fusion module and the multi-scene digital twin operation module are mainly used for interaction, interconnection and fusion of digital threads of the multi-subsystem digital twin model facing the typical scene to form a large system-level integrated digital twin model of the typical scene. The business application module facing to operation control or management mainly refers to carrying out business applications such as state monitoring, diagnosis, prediction, decision optimization and the like by taking large-system-level operation control or management optimization as task guidance based on the established large-system-level integrated digital twin body model.

In some embodiments, data transmission and communication may employ a variety of communication modes including, but not limited to, 5G, fiber optic cable, etc., and support a variety of communication protocols including, but not limited to, TCP/IP, http, webSocket, MQTT, UDP, IEC, etc. Data governance and storage support timing databases (including but not limited to databases such as TimescaleDB), relational databases (including but not limited to databases such as PgSQL), distributed file systems (including but not limited to file systems such as Minio), data stream governance (including but not limited to governance modes such as Spark Streaming framework), and distributed computing (including but not limited to Flink computing engines).

The digital twin platform adopts a Web architecture, development languages comprise, but are not limited to, golong, python, javaScript, C ++, and the like, an operating system comprises, but is not limited to, linux+ Docker, windows +docker, and the like, a model deep learning environment comprises, but is not limited to, tensorFlow, pytorch, and the like, a service monitoring system comprises, but is not limited to, prometheus, supports 3D, virtual reality, augmented reality, and the like, and data visualization development tools and languages comprise, but are not limited to Grafana, javaScript, react, c #, units, and lua. The digital twin platform interface supports authority management such as account number and password management, business operation with timeliness token and the like, interactive security management such as WAF, supervisr monitoring and the like, and load balancing management such as LVS, DNS polling and the like.

It should be noted that the modules and the technology can be developed and deployed in a large server of a station-type large data center in a centralized manner, and can also be deployed in a distributed manner on side or end-side computing equipment by adopting a cloud-side-end cooperative architecture.

Based on the above description, the embodiments of the present application provide a digital twin modeling method of a novel generator set, where the generator set includes a boiler system, a steam turbine system, and a molten salt system, and the method includes: acquiring structural relations among the boiler system, the steam turbine system and the molten salt system; according to the structural relation, carrying out structural domain division on the unit to obtain a plurality of substructures; determining corresponding operation principles of the plurality of substructures under various working conditions; and carrying out digital twin modeling on the dominant thermodynamic dynamic characteristics of the unit power generation according to the corresponding operation principles of the plurality of substructures under various working conditions. Therefore, in the method, a fused salt system is introduced to perform data modeling, and the unit is subjected to structural domain division and combines with the operation principle of various working conditions, so that the model subjected to digital twin modeling is more similar to the actual condition of the unit, and the accuracy of the model is improved.

As shown in fig. 5, the figure is a schematic diagram of a digital twin modeling device of a novel generator set provided in an embodiment of the present application, where the set includes a boiler system, a steam turbine system, and a molten salt system, and the device includes:

an acquisition module 501 for acquiring structural relationships among the boiler system, the steam turbine system and the molten salt system;

the division module 502 performs structural domain division on the unit according to the structural relationship to obtain a plurality of substructures;

a determining module 503, configured to determine operation principles of the plurality of substructures corresponding to a plurality of working conditions;

the modeling module 504 performs digital twin modeling on the generating characteristics of the unit by adopting one or more of an algebraic or differential equation mechanism model and a data-driven machine learning agent model according to the corresponding operation principles of the plurality of substructures under various working conditions.

Optionally, the boiler system comprises a boiler, an economizer, a molten salt heat exchanger, a boiler reheater and a boiler superheater; the steam turbine system comprises high-pressure water supply, a deaerator, low-pressure water supply, a low-pressure cylinder, a medium-pressure cylinder and a high-pressure cylinder, and the molten salt system comprises a molten salt superheater, a molten salt evaporator, a molten salt preheater, a high-temperature molten salt tank, a steam-molten salt heat exchanger and a low-temperature molten salt tank.

Optionally, the dividing module 502 is specifically configured to divide the structural domain of the unit according to the directed graph node relationship matrix corresponding to the structural relationship to obtain a plurality of sub-matrices, where the plurality of sub-matrices are used to characterize a plurality of sub-structures of the unit.

Optionally, the dividing module 502 is specifically configured to process a directed graph node relationship matrix corresponding to the structural relationship by using a directed graph node clustering algorithm, so as to divide a structural domain of the unit.

Optionally, the obtaining module 501 is further configured to obtain corresponding real operation data under the multiple working conditions during the generating process of the unit;

the modeling module 504 is specifically configured to perform digital twin modeling on the generating characteristics of the unit according to the operation principles of the plurality of substructures under the plurality of working conditions.

From the foregoing description of the embodiments, it will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of functional modules is illustrated, and in practical application, the above-described functional allocation may be implemented by different functional modules according to needs, i.e. the internal structure of the apparatus is divided into different functional modules to implement all or part of the functions described above. The specific working processes of the above-described systems, devices and units may refer to the corresponding processes in the foregoing method embodiments, which are not described herein.

In the several embodiments provided in this embodiment, it should be understood that the disclosed processing apparatus and method may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, e.g., the division of the modules or units is merely a logical functional division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present embodiment may be integrated in one processing unit, each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present embodiment may be essentially or a part contributing to the prior art or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium, including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the method described in the respective embodiments. And the aforementioned storage medium includes: flash memory, removable hard disk, read-only memory, random access memory, magnetic or optical disk, and the like.

Embodiments of the present application also provide a computer readable storage medium for storing a computer program for performing a method according to any one of the above method embodiments.

The present application further provides a computer program product having a computer program stored thereon, which, when executed by a processing device, implements a method according to any of the above method embodiments.

The foregoing is merely a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, and any changes or substitutions within the technical scope of the present disclosure should be covered in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A digital twin modeling method of a novel generator set, wherein the set comprises a boiler system, a steam turbine system and a molten salt system, the method comprising:

2. The method of claim 1, wherein the boiler system comprises a boiler, an economizer, a molten salt heat exchanger, a boiler reheater, a boiler superheater, a first valve, a first hydraulic/electric pump, a first conduit; the steam turbine system comprises high-pressure water supply, a deaerator, low-pressure water supply, a low-pressure cylinder, a medium-pressure cylinder, a high-pressure cylinder, a second valve, a second hydraulic/electric pump and a second pipeline; the molten salt system comprises a molten salt superheater, a molten salt evaporator, a molten salt preheater, a high-temperature molten salt tank, a steam-molten salt heat exchanger, a low-temperature molten salt tank, a third valve, a third hydraulic/electric pump and a third pipeline.

3. The method according to claim 1, wherein the performing domain division on the unit according to the structural relationship to obtain a plurality of substructures includes:

4. A method according to claim 3, wherein the performing structural domain division on the unit according to the directed graph node relation matrix corresponding to the structural relation comprises:

5. The method according to claim 1, wherein the method further comprises:

and carrying out digital twin modeling on the generating characteristics of the unit according to the corresponding operating principles and real operating data of the plurality of substructures under various working conditions.

6. A digital twin modeling device for a novel generator set, the set comprising a boiler system, a turbine system and a molten salt system, the device comprising:

and the modeling module is used for digitally twinning and modeling the generating characteristics of the unit by adopting one or more of an algebraic or differential equation mechanism model and a data-driven machine learning agent model according to the corresponding operation principles of the plurality of substructures under various working conditions.

7. The apparatus of claim 6, wherein the boiler system comprises a boiler, an economizer, a molten salt heat exchanger, a boiler reheater, a boiler superheater, a first valve, a first hydraulic/electric pump, a first conduit; the steam turbine system comprises high-pressure water supply, a deaerator, low-pressure water supply, a low-pressure cylinder, a medium-pressure cylinder, a high-pressure cylinder, a second valve, a second hydraulic/electric pump and a second pipeline; the steam turbine system comprises a molten salt superheater, a molten salt evaporator, a molten salt preheater, a high-temperature molten salt tank, a steam-molten salt heat exchanger, a low-temperature molten salt tank, a third valve, a third hydraulic/electric pump and a third pipeline.

8. The device of claim 6, wherein the dividing module is specifically configured to divide the structural domain of the unit according to a node relation matrix of the directed graph corresponding to the structural relation, so as to obtain a plurality of submatrices, where the plurality of submatrices are used to characterize a plurality of substructures of the unit.

9. A computer readable storage medium, characterized in that the computer readable storage medium is adapted to store a computer program adapted to perform the method of any of claims 1-5.

10. A computer program product having a computer program stored thereon, characterized in that the program, when being executed by a processing means, implements the method of any of claims 1 to 5.