CN111046018B

CN111046018B - Multi-objective optimization-based power generation equipment operation condition library construction method and device

Info

Publication number: CN111046018B
Application number: CN201911075866.8A
Authority: CN
Inventors: 刘双白; 朱龙飞; 张德利; 张春雷; 周卫庆
Original assignee: State Grid Corp of China SGCC; North China Electric Power Research Institute Co Ltd; Nanjing Institute of Technology
Current assignee: State Grid Corp of China SGCC; North China Electric Power Research Institute Co Ltd; Nanjing Institute of Technology
Priority date: 2019-11-06
Filing date: 2019-11-06
Publication date: 2024-02-27
Anticipated expiration: 2039-11-06
Also published as: CN111046018A

Abstract

The invention provides a method and a device for constructing a power generation equipment operation condition library based on multi-objective optimization, wherein the method comprises the following steps: acquiring historical working condition data of a power generation device; establishing a multi-objective optimization model by taking minimum estimation error and shortest estimation time as targets when the operating condition library of the power generation equipment is utilized to carry out multi-element state estimation; the multi-objective genetic algorithm is utilized to solve the multi-objective optimization model to obtain the optimal cluster class number, and then the power generation equipment operation condition library is obtained, the power generation equipment operation condition library is obtained by utilizing the data clustering algorithm to extract typical operation conditions of the historical operation condition data based on the optimal cluster class number, the number of the typical operation conditions in the power generation equipment operation condition library is equal to the optimal cluster class number, the multi-objective optimization model is constructed, the multi-objective genetic algorithm is utilized to conduct optimization, the extraction of the operation condition library can be completed from the multi-objective optimization angle, and the estimation precision is high and the operation time is short when the multi-element state estimation is conducted by utilizing the operation condition library.

Description

Multi-objective optimization-based power generation equipment operation condition library construction method and device

Technical Field

The invention relates to the technical field of electric power, in particular to a method and a device for constructing a power generation equipment operation condition library based on multi-objective optimization.

Background

Along with the continuous improvement of the informatization automation degree of a power plant, massive historical operation data of the unit are stored in a historical/real-time database, the application of processing the massive historical operation data of the unit in unit monitoring, diagnosis and optimization by utilizing a big data modeling technology is also widely researched, and the research on the front edge is more and more focused on the accuracy and reliability aspects of a data modeling method.

The multi-state estimation method based on the historical typical working conditions is called a research hot spot in recent years, and the characteristics of simple structure and easy realization are particularly suitable for engineering application, and the acquisition of a stable and reliable working condition library is the key of whether the method can be successfully applied.

At present, the following method is mainly adopted in the aspect of construction and research of a typical working condition library: the Min-Max method is that the historical operation data of the unit are extracted by manually setting a certain interval, and the maximum and minimum values are stored; the data clustering method is used for classifying historical operation data of the unit by utilizing a data clustering algorithm, and finally determining typical operation conditions; and providing a specific index according to the information quantity of the data reaction, and searching enough data to form a typical working condition library.

However, the working condition library constructed by the method cannot be used for both quantity and precision, so that the estimation precision is poor or the operation time is too long when the working condition library is used for multi-element state estimation, and popularization and application are not facilitated.

Disclosure of Invention

Aiming at the problems in the prior art, the invention provides a method and a device for constructing a power generation equipment operation condition library based on multi-objective optimization, electronic equipment and a computer readable storage medium, which can at least partially solve the problems in the prior art.

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

in a first aspect, a method for constructing a power generation equipment operation condition library based on multi-objective optimization is provided, including:

acquiring historical working condition data of a power generation device;

establishing a multi-objective optimization model by taking minimum estimation error and shortest estimation time as targets when the operating condition library of the power generation equipment is utilized to carry out multi-element state estimation;

and solving the multi-objective optimization model by utilizing a multi-objective genetic algorithm to obtain an optimal cluster class number, and further obtaining a power generation equipment operation condition library, wherein the power generation equipment operation condition library is obtained by extracting typical operation conditions of the historical operation condition data based on the optimal cluster class number by utilizing a data clustering algorithm, and the number of the typical operation conditions in the power generation equipment operation condition library is equal to the optimal cluster class number.

Further, the multi-objective optimization model is:

wherein F is _RMSE (x) And (3) representing an estimation error, T (x) representing estimation time, x representing a clustering class number, L representing the minimum working condition number contained in the power generation equipment operation working condition library, and H representing the maximum working condition number contained in the power generation equipment operation working condition library.

Further, the data clustering algorithm includes: density clustering-based method, LDA clustering algorithm or K-means clustering algorithm, fuzzy C-means clustering algorithm.

Further, the method for solving the multi-objective optimization model by utilizing the multi-objective genetic algorithm to obtain the optimal cluster class number comprises the following steps:

generating an initial population, the initial population comprising a plurality of individuals;

clustering the historical working condition data by taking each individual as a clustering class number to obtain a plurality of corresponding alternative working condition libraries;

carrying out regression estimation on test data by using an alternative working condition library to obtain a corresponding estimated value;

acquiring an estimation error and an estimation time of the alternative working condition library according to the estimation value;

acquiring the adaptability of the working condition library according to the estimation error and the estimation time;

selecting the next generation of individuals according to the fitness, and generating a new population after crossing and mutation;

and carrying out iterative computation according to the new population until reaching a termination condition to obtain the optimal cluster class number.

Further, when regression estimation is performed on test data by using an alternative working condition library to obtain a corresponding estimated value, the following formula is adopted:

wherein,represents the estimated value corresponding to the candidate working condition library i, D _i Representing an alternative working condition library i,/>And the test data of the application alternative working condition library i is represented.

Further, when the estimation error and the estimation time of the candidate working condition library are obtained according to the estimation value, the following formula is adopted:

RMSE _i and the estimation error corresponding to the alternative working condition library i is represented, namely the root mean square error, i represents the number of the test data sets, and the estimation time is directly obtained by timing.

Further, the obtaining the adaptability of the working condition library according to the estimation error and the estimation time comprises the following steps:

judging whether the estimated time is smaller than a first preset threshold value or not;

if so, taking the estimated error as the adaptability of the working condition library.

Further, the method for obtaining the adaptability of the working condition library according to the estimation error and the estimation time further comprises the following steps:

if not, judging whether the estimation time is smaller than a second preset threshold value;

and when the estimated time is smaller than a second preset threshold value, calculating the adaptability of the working condition library by adopting a adaptability calculation formula.

Further, the fitness calculation formula is:

g(x)＝F _RMSE (x)+λT(x)

Wherein g (x) represents fitness, F _RMSE (x) And (3) representing an estimation error, wherein T (x) represents estimation time, x represents a cluster class number, and lambda is a preset parameter.

Further, the method for generating a new population after crossing and mutation by selecting the next generation of individuals according to the fitness comprises the following steps:

sequencing the alternative working condition libraries according to the order from small to large in adaptability;

selecting a preset number of individuals with the front ranking as a super parent population;

performing genetic operation on the rest of the population to generate a new sub-population;

and merging the excellent parent population and the new child population to obtain a new population.

Further, before the clustering is performed on the historical working condition data to obtain the working condition library of the power generation equipment, the method further comprises the following steps:

preprocessing the historical working condition data.

Further, the preprocessing the historical operating condition data includes:

and detecting and deleting the mutation points in the historical working condition data.

Further, the historical operating condition data includes a plurality of mechanical performance monitoring parameter values for the power plant;

the preprocessing of the historical operating mode data comprises:

acquiring the fluctuation rate of a mechanical performance monitoring parameter value in the history working condition data within a preset time interval;

judging whether the fluctuation rate is larger than a pre-acquired standard deviation;

If not, deleting the historical working condition data in the preset time interval.

In a second aspect, a power generation equipment operation condition library construction device based on multi-objective optimization is provided, including:

the historical working condition data acquisition module is used for acquiring historical working condition data of a power generation device;

the modeling module is used for establishing a multi-objective optimization model by taking minimum estimation error and shortest estimation time as targets when the power generation equipment operation working condition library is utilized to carry out multi-element state estimation;

and the optimizing module is used for solving the multi-target optimizing model by utilizing a multi-target genetic algorithm to obtain the optimal cluster class number, so as to obtain a power generation equipment operation condition library, wherein the power generation equipment operation condition library is obtained by extracting typical operation conditions of the historical operation condition data based on the optimal cluster class number by utilizing a data clustering algorithm, and the number of the typical operation conditions in the power generation equipment operation condition library is equal to the optimal cluster class number.

Further, the optimizing module includes:

an initial population generation unit that generates an initial population including a plurality of individuals;

The clustering unit clusters the historical working condition data by taking each individual as a clustering class number to obtain a plurality of corresponding alternative working condition libraries;

the regression estimation unit carries out regression estimation on test data by using an alternative working condition library to obtain a corresponding estimated value;

the first calculation unit is used for obtaining the estimation error and the estimation time of the alternative working condition library according to the estimation value;

the second calculation unit is used for acquiring the adaptability of the working condition library according to the estimation error and the estimation time;

the population updating unit is used for selecting the next generation of individuals according to the fitness, and generating a new population after crossing and mutation;

and the iterative calculation unit performs iterative calculation according to the new population until the termination condition is reached, so as to obtain the optimal cluster class number.

Further, the second calculation unit includes:

a first judging subunit for judging whether the estimation time is smaller than a first preset threshold value;

and the first fitness computing subunit takes the estimation error as the fitness of the working condition library if the estimation time is smaller than a first preset threshold value.

Further, the second computing unit further includes:

a second judging subunit for judging whether the estimated time is less than a second preset threshold value if the estimated time is not less than the first preset threshold value;

And the second fitness calculating subunit is used for calculating the fitness of the working condition library by adopting a fitness calculating formula when the estimated time is smaller than a second preset threshold value.

Further, the population updating unit includes:

the sequencing subunit sequences the candidate working condition libraries according to the order from small to large of the fitness;

selecting a subunit from the optimal parent population, and selecting a preset number of individuals with the top ranking as the optimal parent population;

a sub population generation subunit, which performs genetic operation on the rest of the population to generate a new sub population;

and a population updating subunit, wherein the excellent parent population and the new child population are combined to obtain a new population.

Further, the method further comprises the following steps:

and the data preprocessing module is used for preprocessing the historical working condition data.

Further, the data preprocessing module includes:

and the mutation point removing unit is used for detecting and deleting mutation points in the historical working condition data.

the data preprocessing module comprises:

the fluctuation rate acquisition unit is used for acquiring the fluctuation rate of a mechanical performance monitoring parameter value in the history working condition data within a preset time interval;

a fluctuation ratio judgment unit that judges whether or not the fluctuation ratio is larger than a pre-acquired standard deviation;

And the fluctuation data deleting unit is used for deleting the history working condition data in the preset time interval if the fluctuation rate is not greater than the pre-acquired standard deviation.

In a third aspect, there is provided an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing when the program:

acquiring historical working condition data of a power generation device;

In a fourth aspect, a computer readable storage medium is provided, having stored thereon a computer program which, when executed by a processor, implements:

acquiring historical working condition data of a power generation device;

The invention provides a method and a device for constructing a power generation equipment operation condition library based on multi-objective optimization, electronic equipment and a computer readable storage medium, wherein the method comprises the following steps: acquiring historical working condition data of a power generation device; establishing a multi-objective optimization model by taking minimum estimation error and shortest estimation time as targets when the operating condition library of the power generation equipment is utilized to carry out multi-element state estimation; and solving the multi-objective optimization model by utilizing a multi-objective genetic algorithm to obtain an optimal cluster class number, and further obtaining a power generation equipment operation condition library, wherein the power generation equipment operation condition library is obtained by extracting typical operation conditions of the historical operation condition data based on the optimal cluster class number by utilizing a data clustering algorithm, and the number of the typical operation conditions in the power generation equipment operation condition library is equal to the optimal cluster class number. The multi-objective optimization model is constructed, the multi-objective genetic algorithm is utilized to perform optimization, the operation condition library can be extracted from the multi-objective optimization angle, the obtained condition library can be used for state monitoring, modeling and other purposes, the application requirements of the condition library are met, and the problems that the quantity and the precision cannot be considered, the estimation precision is poor or the operation time is too long when the condition library is utilized to perform multi-element state estimation, and popularization and application are not facilitated are solved.

The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular description of preferred embodiments, as illustrated in the accompanying drawings.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:

fig. 1 is a schematic diagram of an architecture between a server S1 and a client device B1 according to an embodiment of the present invention;

fig. 2 is a schematic diagram of an architecture among a server S1, a client device B1 and a database server S2 according to an embodiment of the present invention;

FIG. 3 is a flowchart of a method for building a power plant operating condition library based on multi-objective optimization in accordance with an embodiment of the present invention;

fig. 4 shows a specific step of step S300 in fig. 3;

fig. 5 shows a specific step of step S350 in fig. 4;

fig. 6 shows a specific step of step S370 in fig. 4;

FIG. 7 is a graph comparing an evaluation value and an actual measurement value of a power generation equipment operation condition library constructed by the power generation equipment operation condition library construction method based on multi-objective optimization according to the embodiment of the present invention after evaluating test data;

FIG. 8 is a second flow chart of a method for building a power plant operating condition library based on multi-objective optimization in an embodiment of the present invention;

fig. 9 shows a specific step of step S10 in fig. 8;

FIG. 10 is a block diagram of a multi-objective optimization-based power plant operating condition library construction device in accordance with an embodiment of the present invention;

FIG. 11 is a block diagram II of a multi-objective optimization-based power plant operating condition library construction apparatus in an embodiment of the present invention;

fig. 12 is a block diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.

It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

It should be noted that the terms "comprises" and "comprising," and any variations thereof, in the description and claims of the present application and in the foregoing figures, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.

It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

The working condition library constructed in the prior art cannot give consideration to quantity and precision, so that the estimation precision is poor or the operation time is too long when the working condition library is used for multi-element state estimation, and popularization and application are not facilitated.

In order to solve the technical problems in the prior art, the embodiment of the invention provides a method for constructing a power generation equipment operation working condition library based on multi-objective optimization, which is characterized in that a multi-objective optimization model is constructed, and a multi-objective genetic algorithm is utilized for optimizing, so that the operation working condition library can be extracted from the multi-objective optimization angle, the obtained working condition library can be used for purposes such as state monitoring and modeling, the application requirements of the working condition library are met, and the problems that the quantity and the precision cannot be considered, the estimation precision is poor or the operation time is too long when the multi-objective optimization working condition library is utilized to estimate the multi-element state, and popularization and application are not facilitated are solved.

In view of this, the present application provides a power generation device operation condition library construction device based on multi-objective optimization, which may be a server S1, see fig. 1, where the server S1 may be communicatively connected to at least one client device B1, the client device B1 may send condition data to the server S1, and the server S1 may receive the condition data online. The server S1 can preprocess the acquired working condition data online or offline, and aims at minimum estimation error and shortest estimation time when the power generation equipment operation working condition library is utilized to carry out multi-element state estimation, so as to establish a multi-objective optimization model; and solving the multi-objective optimization model by utilizing a multi-objective genetic algorithm to obtain an optimal cluster class number, and further obtaining a power generation equipment operation condition library, wherein the power generation equipment operation condition library is obtained by extracting typical operation conditions of the condition data based on the optimal cluster class number by utilizing a data clustering algorithm, and the number of the typical operation conditions in the power generation equipment operation condition library is equal to the optimal cluster class number.

In addition, referring to fig. 2, the server S1 may be further communicatively connected to at least one database server S2, where the database server S2 is configured to store historical operating condition data, and the client device B1 sets the required parameters. The database server S2 transmits the historical working condition data to the server S1 on line, the server S1 can receive the historical working condition data on line, and then a power generation equipment operation working condition library is constructed according to the historical working condition data.

It is understood that the client device B1 may include a temperature sensor, a voltmeter, an ammeter, etc. detection device.

Any suitable network protocol may be used for communication between the server and the client device, including those not yet developed at the filing date of this application. The network protocols may include, for example, TCP/IP protocol, UDP/IP protocol, HTTP protocol, HTTPS protocol, etc. Of course, the network protocol may also include, for example, RPC protocol (Remote Procedure Call Protocol ), REST protocol (Representational State Transfer, representational state transfer protocol), etc. used above the above-described protocol.

FIG. 3 is a flowchart of a method for building a power plant operating condition library based on multi-objective optimization in accordance with an embodiment of the present invention. As shown in fig. 3, the method for constructing the operation condition library of the power generation equipment based on the multi-objective optimization can include the following steps:

Step S100: historical operating condition data of a power generation device is obtained.

According to the operation principle and characteristics of the power generation equipment, measuring points of the power generation equipment are arranged, state vectors of a plurality of parameters related to the operation characteristics of the power generation equipment are selected to form equipment, for example, for a coal mill, parameters related to mechanical operation of the coal mill mainly comprise bearing temperature of a reduction gearbox, lubricating oil temperature, lubricating oil pressure, motor current and the like, and the parameters can be combined into mechanical performance monitoring parameters of the coal mill to serve as working condition library parameters representing mechanical performance of the coal mill.

And then, collecting historical working condition data of continuous operation of the equipment in a period of time or a plurality of periods of time by taking the determined working condition library parameters as the basis, wherein the principle of historical data collection is that all working conditions must cover the whole working condition range of the operation of the machine set. For example, for coal mills, the most important criteria to choose are the unit load and the mill motor current, i.e. for a coal mill, the unit load and the mill current must cover their maximum and minimum operating values.

Step S200: establishing a multi-objective optimization model by taking minimum estimation error and shortest estimation time as targets when the operating condition library of the power generation equipment is utilized to carry out multi-element state estimation;

Wherein two optimization objectives are determined: 1. the estimation effect is best, generally, the more the number of working conditions contained in the working condition library is, the wider the coverage range is, the larger the probability of searching a new working condition to be similar to the new working condition is, and the better the corresponding estimation effect is; 2. the shortest calculation time is, the fewer the number of working condition vectors contained in the working condition library in the actual calculation process is, the fewer the number of working conditions participating in calculation is, and the faster the corresponding calculation speed is.

Specifically, the number of working conditions (or the class of working conditions) contained in the finally determined operation working condition library of the power generation equipment is used as an optimization variable x, the estimation effect is represented by a root mean square RMSE of errors, the determined optimization targets are the root mean square minimum of the errors and the estimation time Tminimum, and the multi-target optimization model is as follows:

wherein F is _RMSE (x) And (3) representing an estimation error, T (x) representing estimation time, x representing a clustering class number, L representing the minimum working condition number contained in the power generation equipment operation working condition library, and H representing the maximum working condition number contained in the power generation equipment operation working condition library. Wherein H and L are preset values.

Step S300: and solving the multi-objective optimization model by utilizing a multi-objective genetic algorithm to obtain an optimal cluster class number, and further obtaining a power generation equipment operation condition library, wherein the power generation equipment operation condition library is obtained by extracting typical operation conditions of the historical operation condition data based on the optimal cluster class number by utilizing a data clustering algorithm, and the number of the typical operation conditions in the power generation equipment operation condition library is equal to the optimal cluster class number.

The multi-objective genetic algorithm HSGA_II is adopted for solving, the algorithm can be specifically realized by adopting existing open source software, such as MATLAB, or by adopting a python genetic algorithm toolbox, and a user only needs to input data and set iteration times.

The data clustering algorithm comprises the following steps: density clustering-based methods, LDA clustering algorithms or K-means clustering algorithms, fuzzy C-means clustering algorithms and the like.

According to the method for constructing the operation working condition library of the power generation equipment based on the multi-objective optimization, the multi-objective optimization model is constructed, the multi-objective genetic algorithm is utilized for optimizing, the operation working condition library can be extracted from the multi-objective optimization angle, the obtained working condition library can be used for purposes such as state monitoring and modeling, the application requirements of the working condition library are met, and the problems that the estimation precision is poor or the operation time is too long when the multi-objective optimization model is utilized for carrying out multi-element state estimation and popularization and application are not facilitated are solved.

Fig. 4 shows a specific step of step S300 in fig. 3. As shown in fig. 4, this step S300 may include the following:

step S310: an initial population is generated, the initial population comprising a plurality of individuals.

The number of individuals is selected according to the application requirements, for example, 30 to 120, such as 50, 80, 100, etc. can be selected.

Step S320: and clustering the historical working condition data by taking each individual as a clustering class number to obtain a plurality of corresponding alternative working condition libraries.

Wherein a density-based clustering method or other clustering algorithms may be employed.

Step S330: and carrying out regression estimation on the test data by using an alternative working condition library to obtain a corresponding estimated value.

Specifically, the following formula is employed:

Step S340: and obtaining the estimation error and the estimation time of the alternative working condition library according to the estimation value.

Specifically, the following formula is adopted:

Step S350: and acquiring the adaptability of the working condition library according to the estimation error and the estimation time.

Step S360: and judging whether a termination condition is reached.

If yes, go to step S380, otherwise go to step S370.

The termination condition may be the iteration number, that is, it is determined whether the iteration number reaches the preset iteration number, if yes, the genetic algorithm is terminated, the current obtained optimal solution is used as the optimal cluster class number, if no, the total group is updated, and then steps S320 to S370 are repeatedly executed until the termination condition is reached, that is: and carrying out iterative computation according to the new population until reaching a termination condition to obtain the optimal cluster class number.

The iteration number is set by a user according to actual demands, the higher the iteration number is, the better the accuracy of the obtained working condition library is, but the longer the processing time is, the lower the iteration number is, the shorter the processing time is, but the accuracy of the obtained working condition library is poor, and the iteration number can be set to 8 to 50 times, such as 10, 20, 30, 40 and the like, in terms of distance.

Step S370: and selecting the next generation of individuals according to the fitness, and generating a new population after crossing and mutation.

Step S380: and obtaining the optimal cluster class number according to the adaptability.

By optimizing the multi-objective optimization model by adopting the method, the optimal solution can be obtained in the shortest time, and the efficiency of the method is improved.

Fig. 5 shows a specific step of step S350 in fig. 4. Referring to fig. 5, this step S350 may include the following:

step S351: judging whether the estimated time is smaller than a first preset threshold value or not;

if yes, go to step S352, if no, go to step S353.

Specifically, the first preset threshold is set by the user according to the accuracy requirement, for example, may be between 0.5S and 3S, such as 1S, 2S, 2.5S, and so on.

Step S352: and taking the estimated error as the adaptability of the working condition library.

Step S353: judging whether the estimated time is smaller than a second preset threshold value or not;

if yes, go to step S354, if no, go to step S355.

Specifically, the second preset threshold is set by the user according to the accuracy requirement, for example, may be between 4S and 7S, such as 5S, 6S, 5.5S, and the like.

Step S354: and calculating the adaptability of the working condition library by adopting a fitness calculation formula.

The fitness calculation formula is:

g(x)＝F _RMSE (x)+λT(x)

wherein g (x) represents fitness, F _RMSE (x) And (3) representing an estimation error, wherein T (x) represents the estimation time, x represents the clustering class number, lambda is a preset parameter, and the parameter is set by a user.

Step S355: the initial population is regenerated and returns to step S320.

Fig. 6 shows a specific step of step S370 in fig. 4. Referring to fig. 6, this step S370 specifically includes the following:

step S371: and sequencing the candidate working condition libraries according to the order of the adaptability from small to large.

Step S372: and selecting a preset number of individuals with the top ranking as a superparent population.

Step S373: genetic manipulation of the remaining population generates a new sub-population.

Step S374: and merging the excellent parent population and the new child population to obtain a new population.

The following describes, by way of example, a process of solving the multi-objective optimization model by using a multi-objective genetic algorithm to obtain an optimal cluster class number:

(1) Generating an initial population, assuming that n individuals are included, denoted as x ₁ ,x ₂ ,…,x _n ；

(2) Selecting a density-based clustering method, and integrating each x _i Clustering the historical working condition data as a clustering class number to obtain n alternative working condition libraries D ₁ ,…,D _n ；

(3) Using each working-condition library D according to the estimation principle _i Regression estimation is carried out on the test data, and a computational formula of multi-element state estimation is adopted as followsAnd calculating the corresponding root mean square error RMSEi and the corresponding estimation time ti of each working condition library.

(4) The Pareto front solution set { D _j ,RMSE _j ,t _j ,x _j Fitness calculation evaluation was performed for j=1, …, p } using the following principles:

(1) if t _j Less than or equal to 1s, and RMSE is added _j As the fitness of the working condition library;

(2) if 1s<t _j <5s, calculating the adaptability of the working condition library in a weighted mode;

specifically, if 1s<t _j Less than or equal to 2, the requirement on model precision in the calculation time period is considered to be higher, and lambda=0.1 is taken; if 2s<t _j Less than or equal to 3, lambda=0.2; if 3s<t _j And less than or equal to 5, the specific gravity of the calculation time is considered to be increased in the calculation time period, and lambda=0.3 is taken.

(3) If t _j If the time is more than or equal to 5s, the working condition library is considered unsuitable, and the working condition library is not taken as a candidate working condition library;

in addition, when t corresponds to all initial working condition libraries _j And if the initial population is more than or equal to 5 seconds, the initial population is considered to be required to be adjusted, the initial population is regenerated again, and the steps are executed again. This is an extreme case, where all D matrices are too large resulting in too long a computation time, regardless of the optimal solution falling within this range.

(5) Sorting the Pareto front solution sets according to the fitness;

(6) Selecting individuals with the top 50% of the ranks as the super parent population, and carrying out genetic operation on the rest populations, wherein the genetic operation comprises selection, crossing and mutation to generate a new child population, and the new generation population is the sum of the super parent population and the child population;

(7) Judging whether a termination condition is reached, and outputting an optimal solution in the current population if the termination condition is reached; and if the termination condition is not met, repeating the steps 2-6.

Specifically, a set of parameters are used for representing the running state of a cold end system of a certain unit, and 21 parameters including condenser vacuum, circulating water pump motor current, circulating water pump outlet pressure, condenser end difference, condensing water pump motor current, condensing water pump outlet pressure and the like are selected. And 20000 groups of normal operation history data are selected, the front 18000 groups of data are used for constructing a system working condition library, and the rear 2000 groups of data are used for verification calculation.

The method has the advantages that the proper parameters are selected and optimized and calculated by using the NSGA-II genetic algorithm, the obtained results are shown in the table 1 and the figure 7, and the power generation equipment operation condition library constructed by the power generation equipment operation condition library construction method based on the multi-objective optimization provided by the embodiment of the invention has very close evaluation values and actual measurement values after evaluating the test data, so that the power generation equipment operation condition library constructed by the power generation equipment operation condition library construction method based on the multi-objective optimization provided by the embodiment of the invention is verified to have high evaluation accuracy and short operation time.

Table 1 optimization results

FIG. 8 is a second flow chart of a method for building a power plant operating condition library based on multi-objective optimization in an embodiment of the present invention; as shown in fig. 8, the method for constructing the operation condition library of the power generation equipment based on the multi-objective optimization may further include, on the basis of including the steps of the operation condition library constructing method of the power generation equipment based on the multi-objective optimization shown in fig. 3:

step S10: and preprocessing the historical working condition data.

The method comprises the steps of preprocessing historical working condition data, including detection and rejection of data mutation points and detection of data fluctuation rate, wherein once the data mutation points are detected, the data mutation points directly reject the group of data, and the data fluctuation rate is based on standard deviation measured by each parameter, namely fluctuation of certain parameter data in a certain period of time is smaller than the standard deviation, the data is considered to be abnormal data, and all data vectors in the continuous period of time are rejected.

Specifically, referring to fig. 9, the historical operating mode data includes various mechanical performance monitoring parameter values of the power generation device, and the step S10 may include the following technical contents:

step S11: and acquiring the fluctuation rate of a mechanical performance monitoring parameter value in the history working condition data within a preset time interval.

The fluctuation rate is obtained by dividing the change value of the mechanical performance monitoring parameter value in the preset time interval by the time.

Step S12: judging whether the fluctuation rate is larger than a pre-acquired standard deviation;

if so, the preprocessing flow is ended, and if not, step S13 is executed.

Step S13: and deleting the historical working condition data in the preset time interval.

In an alternative embodiment, the step S10 may further include: and detecting and deleting the mutation points in the historical working condition data.

The historical working condition data are preprocessed, abnormal data are removed, and the accuracy of the follow-up flow can be effectively improved.

Based on the same inventive concept, the embodiment of the application also provides a power generation equipment operation condition library construction device based on multi-objective optimization, which can be used for realizing the method described in the embodiment, and the embodiment is described below. Because the principle of solving the problem of the power generation equipment operation condition library construction device based on the multi-objective optimization is similar to that of the method, the implementation of the power generation equipment operation condition library construction device based on the multi-objective optimization can be referred to the implementation of the method, and the repetition is omitted. As used below, the term "unit" or "module" may be a combination of software and/or hardware that implements the intended function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

FIG. 10 is a block diagram of a multi-objective optimization-based power plant operating condition library construction device in accordance with an embodiment of the present invention. As shown in fig. 10, the power generation equipment operation condition library construction device based on multi-objective optimization includes: the system comprises a historical working condition data acquisition module 10, a modeling module 20 and an optimizing module 30.

The historical operating condition data acquisition module 10 acquires historical operating condition data of a power generation device.

The modeling module 20 builds a multi-objective optimization model with the objective of minimizing the estimation error and minimizing the estimation time using the power plant operating condition library for multi-element state estimation.

The optimizing module 30 utilizes a multi-objective genetic algorithm to solve the multi-objective optimizing model to obtain an optimal cluster class number, and further obtains a power generation equipment operation condition library, wherein the power generation equipment operation condition library is obtained by extracting typical operation conditions of the historical operation condition data based on the optimal cluster class number by utilizing a data clustering algorithm, and the number of the typical operation conditions in the power generation equipment operation condition library is equal to the optimal cluster class number.

According to the power generation equipment operation working condition library construction device based on multi-objective optimization, the multi-objective optimization model is constructed, the multi-objective genetic algorithm is utilized for optimizing, the operation working condition library can be extracted from the multi-objective optimization angle, the obtained working condition library can be used for purposes of state monitoring, modeling and the like, the application requirements of the working condition library are met, and the problems that the number and the precision cannot be considered, the estimation precision is poor or the operation time is too long when the working condition library is utilized for multi-element state estimation, and popularization and application are not facilitated are solved.

In an alternative embodiment, the optimizing module 30 may include: the device comprises an initial population generation unit, a clustering unit, a regression estimation unit, a first calculation unit, a second calculation unit, a population updating unit and an iterative calculation unit.

The method comprises the steps that an initial population generation unit generates an initial population, wherein the initial population comprises a plurality of individuals;

specifically, the following formula is employed:

The first calculation unit obtains the estimation error and the estimation time of the alternative working condition library according to the estimation value;

specifically, the following formula is adopted:

The second calculation unit obtains the adaptability of the working condition library according to the estimation error and the estimation time;

the population updating unit selects the next generation of individuals according to the fitness, and generates a new population after crossing and mutation;

and the iterative computation unit performs iterative computation according to the new population until the termination condition is reached, so as to obtain the optimal cluster class number.

The termination condition may be iteration number, that is, whether the iteration number reaches the preset iteration number is judged, if yes, the genetic algorithm is terminated, the current obtained optimal solution is used as the optimal cluster class number, if not, the total group is updated, and then each module repeatedly works until reaching the termination condition, that is: and carrying out iterative computation according to the new population until reaching a termination condition to obtain the optimal cluster class number.

It will be appreciated by those skilled in the art that by optimizing the multi-objective optimization model using the method described above, the optimal solution can be obtained in the shortest time, improving the efficiency of the method.

In an alternative embodiment, the second computing unit comprises: the system comprises a first judging subunit, a second judging subunit, a first fitness computing subunit and a second fitness computing subunit.

The first judging subunit judges whether the estimation time is smaller than a first preset threshold value or not;

And if the estimated time is smaller than a first preset threshold value, the first fitness computing subunit takes the estimated error as the fitness of the working condition library.

The second judging subunit judges whether the estimated time is smaller than a second preset threshold value if the estimated time is not smaller than the first preset threshold value;

And when the estimated time is smaller than a second preset threshold value, the second fitness calculating subunit calculates the fitness of the working condition library by adopting a fitness calculating formula.

The fitness calculation formula is:

g(x)＝F _RMSE (x)+λT(x)

In an alternative embodiment, the population updating unit comprises: the method comprises the steps of sorting sub-units, selecting sub-units of a super parent population, generating sub-units of a sub-population and updating sub-units of the population.

The sequencing subunit sequences the candidate working condition libraries according to the order from small to large in adaptability;

the parent-parent population selection subunit selects a preset number of individuals with the top ranking as a parent-parent population;

the sub-population generation subunit performs genetic operation on the rest of the population to generate a new sub-population;

and the population updating subunit merges the excellent parent population and the new child population to obtain a new population.

FIG. 11 is a block diagram II of a power plant operating condition library construction apparatus based on multi-objective optimization in an embodiment of the present invention. As shown in fig. 11, the power generation equipment operation condition library construction device based on multi-objective optimization may further include, on the basis of including the modules shown in fig. 10: a data preprocessing module 1.

The data preprocessing module 1 preprocesses the historical working condition data.

In an alternative embodiment, the data preprocessing module includes: and the mutation point eliminating unit is used for detecting and deleting mutation points in the historical working condition data. The data preprocessing module comprises: a fluctuation rate acquisition unit, a fluctuation rate judgment unit, and a fluctuation data deletion unit.

The fluctuation rate acquisition unit acquires the fluctuation rate of a mechanical performance monitoring parameter value in the history working condition data within a preset time interval;

The fluctuation rate judging unit judges whether the fluctuation rate is larger than a pre-acquired standard deviation;

and if the fluctuation rate is not greater than the pre-acquired standard deviation, the fluctuation data deleting unit deletes the history working condition data in the preset time interval.

In an alternative embodiment, the historical operating condition data includes a plurality of mechanical performance monitoring parameter values for the power plant;

The apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. A typical implementation device is an electronic device, which may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

In a typical example the electronic device comprises in particular a memory, a processor and a computer program stored on the memory and executable on the processor, said processor implementing the following steps when said program is executed:

acquiring historical working condition data of a power generation device;

From the above description, it can be known that the electronic device provided by the embodiment of the invention can be used for constructing the operation working condition library of the power generation device, optimizing by utilizing a multi-objective genetic algorithm through constructing a multi-objective optimization model, extracting the operation working condition library from the multi-objective optimization angle, and the obtained working condition library can be used for purposes such as state monitoring and modeling, so as to meet the application requirements of the working condition library, solve the problems that the number and the precision cannot be considered, and the estimation precision is poor or the operation time is too long when the multi-objective working condition library is utilized for multi-element state estimation, which is unfavorable for popularization and application.

Referring now to fig. 12, a schematic diagram of an electronic device 600 suitable for use in implementing embodiments of the present application is shown.

As shown in fig. 12, the electronic apparatus 600 includes a Central Processing Unit (CPU) 601, which can perform various appropriate works and processes according to a program stored in a Read Only Memory (ROM) 602 or a program loaded from a storage section 608 into a Random Access Memory (RAM)) 603. In the RAM603, various programs and data required for the operation of the system 600 are also stored. The CPU601, ROM602, and RAM603 are connected to each other through a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.

The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, mouse, etc.; an output portion 607 including a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, a speaker, and the like; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. The drive 610 is also connected to the I/O interface 605 as needed. Removable media 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on drive 610 as needed, so that a computer program read therefrom is mounted as needed as storage section 608.

In particular, according to embodiments of the present invention, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, an embodiment of the invention includes a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:

acquiring historical working condition data of a power generation device;

As can be seen from the above description, the computer readable storage medium provided by the embodiment of the present invention may be used to construct an operation condition library of a power generation device, by constructing a multi-objective optimization model and optimizing by using a multi-objective genetic algorithm, so that the operation condition library can be extracted from the multi-objective optimization, and the obtained condition library can be used for purposes such as state monitoring and modeling, so as to meet the application requirements of the condition library, and solve the problems that the number and the precision cannot be considered, and the estimation precision is poor or the operation time is too long when the condition library is used for performing multi-element state estimation, which is not beneficial to popularization and application.

In such an embodiment, the computer program may be downloaded and installed from a network through the communication portion 609, and/or installed from the removable medium 611.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

For convenience of description, the above devices are described as being functionally divided into various units, respectively. Of course, the functions of each element may be implemented in one or more software and/or hardware elements when implemented in the present application.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.

Claims

1. The utility model provides a power generation equipment operation condition library construction method based on multi-objective optimization, which is characterized by comprising the following steps:

acquiring historical working condition data of a power generation device;

the multi-objective optimization model is as follows:

wherein F is _RMSE (x) Representing an estimation error, wherein T (x) represents estimation time, x represents a clustering class number, L represents the minimum working condition number contained in the power generation equipment operation working condition library, and H represents the maximum working condition number contained in the power generation equipment operation working condition library;

solving the multi-objective optimization model by utilizing a multi-objective genetic algorithm to obtain an optimal cluster class number, and further obtaining a power generation equipment operation condition library, wherein the power generation equipment operation condition library is obtained by extracting typical operation conditions of the history condition data based on the optimal cluster class number by utilizing a data clustering algorithm, and the number of the typical operation conditions in the power generation equipment operation condition library is equal to the optimal cluster class number;

the method for solving the multi-objective optimization model by utilizing the multi-objective genetic algorithm to obtain the optimal cluster class number comprises the following steps:

performing iterative computation according to the new population until reaching a termination condition to obtain an optimal cluster class number;

the step of obtaining the adaptability of the working condition library according to the estimation error and the estimation time comprises the following steps:

if yes, taking the estimated error as the adaptability of the working condition library;

2. The multi-objective optimization-based power generation equipment operation condition library construction method according to claim 1, wherein the data clustering algorithm comprises: density clustering-based method, LDA clustering algorithm or K-means clustering algorithm, fuzzy C-means clustering algorithm.

3. The method for constructing the operation condition library of the power generation equipment based on the multi-objective optimization according to claim 1, wherein when regression estimation is performed on test data by using an alternative condition library to obtain a corresponding estimated value, the following formula is adopted:

wherein,represents the estimated value corresponding to the candidate working condition library i, D _i Representing an alternative working condition library i,/>Indicating that it shouldTest data of the alternative working condition library i are used.

4. The method for constructing the operation condition library of the power generation equipment based on the multi-objective optimization according to claim 3, wherein when the estimation error and the estimation of the candidate condition library are obtained according to the estimation value, the following formula is adopted:

5. The method for constructing the operation condition library of the power generation equipment based on the multi-objective optimization according to claim 1, wherein the fitness calculation formula is as follows:

g(x)＝F _RMSE (x)+λT(x)

6. The method for constructing the operation condition library of the power generation equipment based on the multi-objective optimization according to claim 3, wherein the steps of selecting the next generation of individuals according to the fitness, crossing and mutating to generate a new population comprise the following steps:

7. The method for constructing a power generation device operation condition library based on multi-objective optimization according to claim 1, wherein before clustering the historical operation condition data to obtain the power generation device operation condition library, the method further comprises:

and preprocessing the historical working condition data.

8. The method for constructing a power generation equipment operation condition library based on multi-objective optimization according to claim 7, wherein the preprocessing the historical operation condition data comprises:

9. The multi-objective optimization-based power plant operating condition library construction method of claim 7, wherein the historical operating condition data comprises a plurality of mechanical performance monitoring parameter values of the power plant;

the preprocessing the historical working condition data comprises the following steps:

10. The utility model provides a power generation facility operation condition storehouse construction equipment based on multi-objective optimization which characterized in that includes:

the multi-objective optimization model is as follows:

the optimizing module is used for solving the multi-objective optimizing model by utilizing a multi-objective genetic algorithm to obtain an optimal cluster class number, so as to obtain a power generation equipment operation condition library, wherein the power generation equipment operation condition library is obtained by extracting typical operation conditions of the history condition data based on the optimal cluster class number by utilizing a data clustering algorithm, and the number of the typical operation conditions in the power generation equipment operation condition library is equal to the optimal cluster class number;

The optimizing module comprises:

the iterative calculation unit performs iterative calculation according to the new population until reaching a termination condition to obtain the optimal cluster class number;

the second calculation unit includes:

a first judging subunit for judging whether the estimation time is smaller than a first preset threshold;

the first fitness computing subunit takes the estimation error as the fitness of the working condition library if the estimation time is smaller than a first preset threshold;

A second judging subunit, configured to judge whether the estimated time is less than a second preset threshold value if the estimated time is not less than the first preset threshold value;

11. The multi-objective optimization-based power generation equipment operation condition library construction device according to claim 10, wherein the data clustering algorithm comprises: density clustering-based method, LDA clustering algorithm or K-means clustering algorithm, fuzzy C-means clustering algorithm.

12. The multi-objective optimization-based power plant operating condition library construction device according to claim 10, wherein the population updating unit comprises:

13. The multi-objective optimization-based power plant operating condition library construction device of claim 10, further comprising:

14. The multi-objective optimization-based power plant operating condition library construction device of claim 13, wherein the data preprocessing module comprises:

and the mutation point eliminating unit is used for detecting and deleting mutation points in the historical working condition data.

15. The multi-objective optimization based power plant operating condition library construction device of claim 14, wherein the historical operating condition data comprises a plurality of mechanical performance monitoring parameter values for the power plant;

the data preprocessing module comprises:

a fluctuation ratio judgment unit which judges whether the fluctuation ratio is larger than a pre-acquired standard deviation;

16. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the multi-objective optimization based power plant operating condition library construction method of any one of claims 1 to 9 when the program is executed by the processor.

17. A computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of the multi-objective optimization based power plant operating condition library construction method of any one of claims 1 to 9.