CN117592804A - Liquefied compressed air energy storage liquefaction rate characterization method, system and electronic equipment - Google Patents
Liquefied compressed air energy storage liquefaction rate characterization method, system and electronic equipment Download PDFInfo
- Publication number
- CN117592804A CN117592804A CN202311548257.6A CN202311548257A CN117592804A CN 117592804 A CN117592804 A CN 117592804A CN 202311548257 A CN202311548257 A CN 202311548257A CN 117592804 A CN117592804 A CN 117592804A
- Authority
- CN
- China
- Prior art keywords
- liquefaction rate
- liquefaction
- working conditions
- compressed air
- data set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012512 characterization method Methods 0.000 title claims abstract description 70
- 238000004146 energy storage Methods 0.000 title claims abstract description 39
- 238000012360 testing method Methods 0.000 claims abstract description 53
- 238000012549 training Methods 0.000 claims abstract description 38
- 238000007781 pre-processing Methods 0.000 claims abstract description 15
- 230000007246 mechanism Effects 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 47
- 239000011159 matrix material Substances 0.000 claims description 23
- 238000000354 decomposition reaction Methods 0.000 claims description 12
- 238000011156 evaluation Methods 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 9
- 230000009466 transformation Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 5
- 230000006870 function Effects 0.000 description 12
- 230000008569 process Effects 0.000 description 12
- 230000002159 abnormal effect Effects 0.000 description 8
- 238000009825 accumulation Methods 0.000 description 6
- 239000007788 liquid Substances 0.000 description 5
- 238000010276 construction Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Computation (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Evolutionary Biology (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Marketing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Educational Administration (AREA)
- Public Health (AREA)
- Separation By Low-Temperature Treatments (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
Abstract
The invention relates to a liquefied compressed air energy storage liquefaction rate characterization method, a system and electronic equipment, which comprise the following steps: determining influencing factors related to the liquefaction rate according to the liquefaction compressed air energy storage operation mechanism; collecting the influence factors related to the load, the liquefaction rate and the liquefaction rate of the power station as historical data, and preprocessing, classifying and decomposing the historical data to obtain training data sets and test data sets under different working conditions; constructing liquefaction rate characterization models under different working conditions according to the training data set; according to the power station load, the test data set is subjected to working condition division, the test data set is decomposed and input into the liquefaction rate characterization model corresponding to different working conditions, the liquefaction rate characterization model is used for evaluating the accuracy of the liquefaction rate characterization model, the liquefaction effect can be monitored at any time, and a guarantee measure is provided for improving the efficiency of the liquefied compressed air energy storage power station.
Description
Technical Field
The invention belongs to the technical field of energy storage, and particularly relates to a liquefied compressed air energy storage liquefaction rate characterization method, a system and electronic equipment.
Background
Under the dual-carbon background, the new energy duty ratio of China is rapidly improved, the influence of the instability of the new energy on the power grid is gradually improved, and the construction of large-scale long-time energy storage is needed to provide key support for the stable operation of a novel power system. The compressed air is used as a long-term energy storage, has the advantages of short construction period, high safety, environmental protection and the like, and has wide application in auxiliary services such as peak regulation, frequency modulation, peak clipping and valley filling. The liquefied compressed air energy storage is used as an important branch of compressed air energy storage, the energy storage density of the liquefied air energy storage is high, the energy storage is carried out in the form of liquefied air, and the occupied area of the storage device is small. Therefore, whether the air can be completely liquefied is critical to the efficiency of the system.
Whether the liquefied compressed air energy storage air medium can be completely liquefied is generally characterized by a liquefaction rate, which is the rate or proportion of converting gaseous compressed air energy storage into a liquid state, and a high liquefaction rate means that more compressed air is converted into a liquid state for storage, thereby improving the energy storage efficiency. Meanwhile, the liquefaction rate also affects the response time of the system, and the high liquefaction rate can quickly convert gaseous air into liquid air for storage, so that the energy storage and release process is accelerated. The characterization method of the liquefaction rate generally adopts a mechanism modeling method, namely, a thermodynamic principle and system parameters are adopted for modeling, factors such as heat conduction, heat transfer, material balance and the like of a system and energy loss in the system process are required to be fully considered in the modeling process, influencing factors are complex, and model accuracy is greatly influenced by a mechanism model. For this reason, developing a characterization of the data-driven liquefied compressed air storage liquefaction rate is a current urgent problem to be solved.
Disclosure of Invention
In order to overcome the problems in the prior art, the invention provides a liquefied compressed air energy storage liquefaction rate characterization method, a liquefied compressed air energy storage liquefaction rate characterization system and electronic equipment, which can monitor the liquefaction effect at any time, provide a guarantee measure for improving the efficiency of a liquefied compressed air energy storage power station and are used for solving the technical problems in the prior art.
A method for characterizing the stored energy liquefaction rate of liquefied compressed air, the method comprising the steps of:
s1, determining influence factors related to the liquefaction rate according to a liquefied compressed air energy storage operation mechanism;
s2, collecting the load, the liquefaction rate and the influence factors of the power station as first historical data, and preprocessing the first historical data to obtain second historical data;
s3, classifying the second historical data, determining different operation conditions, and obtaining third historical data under different conditions;
s4, decomposing the third historical data, and randomly dividing the decomposed data to obtain training data sets and test data sets under different working conditions;
s5, constructing liquefaction rate characterization models under different working conditions according to the training data set;
s6, dividing working conditions of the test data set, decomposing the test data set, inputting the test data set into liquefaction rate characterization models corresponding to different working conditions, and evaluating the liquefaction rate characterization models, wherein the sequence of S2 and S3 can be exchanged.
Aspects and any one of the possible implementations described above, further providing an implementation, the factors related to the liquefaction rate include a regenerator exit pressure, a regenerator exit temperature, a pre-throttle pressure, and a pre-throttle air flow.
In the aspects and any possible implementation manner described above, there is further provided an implementation manner, where the different working conditions are divided according to the collected power station load, and are divided into T working conditions, where T is a positive integer.
In the above aspect and any possible implementation manner, there is further provided an implementation manner, in S4, the decomposing adopts an empirical mode decomposition method, and after the third historical data is decomposed, random division is performed, so as to obtain a training data set and a test data set under different working conditions.
In the aspect and any possible implementation manner described above, there is further provided an implementation manner, the liquefaction rate characterization model is built according to the training data set, and is implemented by adopting a width learning method, an input layer of the liquefaction rate characterization model is a liquefaction rate influencing factor, and an output layer of the liquefaction rate characterization model is a liquefaction rate.
In the aspect and any possible implementation manner described above, there is further provided an implementation manner, the width learning method includes the following steps: mapping the data in the input training data set into a characteristic node matrix, and converting the characteristic node matrix into an enhanced node matrix through enhanced transformation; and then combining the characteristic node matrix and the enhanced node matrix into a new input, and further constructing a new relation between the input and the output.
According to the aspects and any one of possible implementation manners, an implementation manner is further provided, the test data sets are subjected to working condition division to obtain test data sets under T working conditions, the test data sets are decomposed, and the decomposed data are input into the liquefaction rate characterization model under the corresponding working conditions, so that the accuracy of the liquefaction rate characterization model under different working conditions is evaluated.
In the aspect and any possible implementation manner, there is further provided an implementation manner, wherein the evaluation is expressed by a decision coefficient, and if the decision coefficient is greater than or equal to 0.9, the characterization model is high in accuracy and can be used for subsequent evaluation; if the decision coefficient is less than 0.9, the characterization model needs to be retrained.
The invention also provides a liquefied compressed air energy storage liquefaction rate characterization system, which is used for realizing the method and comprises the following steps:
the determining module is used for determining influencing factors related to the liquefaction rate according to the liquefied compressed air energy storage operation mechanism;
the preprocessing module is used for collecting the load of the power station, the liquefaction rate and the influence factors as first historical data, and preprocessing the first historical data to obtain second historical data;
the classification module is used for classifying the second historical data, determining different operation conditions and obtaining third historical data under different conditions;
the decomposition module is used for decomposing the third historical data and randomly dividing the decomposed data to obtain training data sets and test data sets under different working conditions;
the building module is used for building liquefaction rate characterization models under different working conditions according to the training data set;
the evaluation module is used for dividing the working conditions of the test data set, decomposing the test data set, inputting the test data set into the liquefaction rate characterization model corresponding to different working conditions, and evaluating the accuracy of the liquefaction rate characterization model.
The invention also provides an electronic device, which comprises:
a memory storing executable instructions;
and a processor executing the executable instructions in the memory to implement the method.
The beneficial effects of the invention are that
Compared with the prior art, the invention has the following beneficial effects:
the invention relates to a liquefied compressed air energy storage liquefaction rate characterization method, which comprises the following steps: determining influencing factors related to the liquefaction rate according to the liquefaction compressed air energy storage operation mechanism; collecting the influence factors related to the load, the liquefaction rate and the liquefaction rate of the power station as historical data, and preprocessing, classifying and decomposing the historical data to obtain training data sets and test data sets under different working conditions; constructing liquefaction rate characterization models under different working conditions according to the training data set; according to the load of the power station, the working condition of the test data set is divided, the test data set is decomposed, and the decomposed test data set is input into the liquefaction rate characterization model corresponding to different working conditions, so that the precision of the liquefaction rate characterization model is evaluated.
Compared with the prior art, the invention has the following technical effects:
(1) The invention provides a liquefied compressed air energy storage liquefaction rate characterization method which can monitor the change process of the liquefaction rate in real time and discover the problem of abnormal liquefaction rate in time.
(2) The invention fully considers the problem of data characteristic difference under different working conditions, establishes a plurality of liquefaction rate characterization models, and the established models are more in line with the actual working conditions and have more significance for characterization of the liquefaction rate.
(3) The invention adopts the empirical mode decomposition method to preprocess the data, has good characterization effect on the time sequence data, and particularly has higher signal to noise ratio on the unstable and irregular data shown by the time sequence data after the empirical mode decomposition processing.
(4) The invention also provides a liquefaction rate characterization model for the liquefaction rate characterization, solves the problems of complex process, large calculated amount and long calculation time of the traditional deep learning method, can quickly obtain the liquefaction rate data, and ensures that the liquefied compressed air energy storage realizes high-efficiency operation.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention.
Detailed Description
For a better understanding of the present invention, the present disclosure includes, but is not limited to, the following detailed description, and similar techniques and methods should be considered as falling within the scope of the present protection. In order to make the technical problems, technical solutions and advantages to be solved more apparent, the following detailed description will be given with reference to the accompanying drawings and specific embodiments.
It should be understood that the described embodiments of the invention are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The terminology used in the embodiments of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
As shown in fig. 1, a flow chart of a method for characterizing the energy storage and liquefaction rate of liquefied compressed air according to an embodiment of the present invention specifically includes the following steps:
and step 101, determining influence factors related to the liquefaction rate according to the energy storage operation mechanism of the liquefied compressed air.
Step 201, collecting power station load, liquefaction rate influencing factors and the like as first historical data, preprocessing the first historical data to obtain high-quality second historical data, and preprocessing the first historical data in advance to realize subsequent high-precision modeling because abnormal data and missing data exist in the adopted historical data and the influence of the data on later modeling is large.
Step 301, classifying the second historical data according to the collected power station load, determining different operation conditions, and obtaining third historical data under different operation conditions, wherein the two processes of classification in this step and preprocessing in the previous step 201 can be reversed, namely, classification is performed first and preprocessing is performed again, so that the purposes are the same, and the purpose of high-precision modeling is achieved later.
Step 401, performing empirical mode decomposition on third historical data under different working conditions, and randomly dividing the decomposed data to obtain a training data set and a test data set under different working conditions. The training process uses the training data set to build a modeling model, and the testing process uses the testing data set to test the accuracy of the training process model. If the model is not accurate enough, retraining using the training dataset is required. The training process and the testing process are performed reciprocally to obtain a high-precision model.
Step 501, building a width learning model, namely a liquefaction rate characterization model, under different working conditions according to the training data set. The function of building the learning model is to use the model to predict the liquefaction rate in the later period, and the liquefaction effect can be monitored in real time. The model is built by adopting the factors of the outlet pressure of the cold accumulation device, the outlet temperature of the cold accumulation device, the temperature before throttling, the pressure before throttling and the air flow before throttling. The purpose of the plant load is to classify the historical data, with the liquefaction rate data as the target value. Modeling requires the construction of [ liquefaction rate influencing factors liquefaction rate ] data pairs.
And 601, dividing the working condition of the test data set according to the load of the power station, carrying out empirical mode decomposition on the test data set, inputting the test data set into the liquefaction rate characterization model under the corresponding working condition, and evaluating the accuracy of the liquefaction rate characterization model, namely evaluating how accurate the established liquefaction rate characterization model can be used or not by using the test data set.
Specifically, the factors related to the liquefaction rate mainly include the cold storage device outlet pressure, the cold storage device outlet temperature, the pre-throttling pressure, and the pre-throttling air flow.
Specifically, the historical data are collected from the DCS system, the collected data comprise running time, power station load, cold accumulation device outlet pressure, cold accumulation device outlet temperature, pre-throttling pressure, pre-throttling air flow and liquefaction rate data displayed by the DCS system, the collection time is past running data of one year before the current running time, and if the historical data of one year are to be collected, the running time is 2022 years 11 month 6 days 00:00 to 2023 years 11 months 6 days 00:00. The data preprocessing adopts an interpolation method to replace abnormal data, the abnormal data is automatically generated by a DCS system, and in the running process of a power station, the abnormal data can be generated due to various factors, and the abnormal data can influence the modeling at the back, so that interpolation replacement is needed by utilizing peripheral data points of the abnormal data.
Specifically, the different working conditions are divided into T working conditions according to the collected power station load data, historical data under the T different working conditions are obtained, and T is a positive integer.
Specifically, the empirical mode decomposition specifically includes decomposing the third historical data in step 403 to obtain a plurality of signal components, where different signal components represent different scale data features. Specifically, for time-series data x (t) in the acquired data, the maximum value and minimum value points of x (t) are found to form an upper envelope line U max (t) and under bagWinding U min And (t) obtaining a mean value m (t) according to the upper envelope curve and the lower envelope curve, and then calculating an intermediate signal h (t) =x (t) -m (t), wherein one time sequence comprises a plurality of maximum value points and a plurality of minimum value points, the maximum value points form the upper envelope curve, and the minimum value points form the lower envelope curve. The average value is obtained by dividing the sum of the upper envelope curve and the corresponding lower envelope curve by 2. And judging whether h (t) meets the requirement of the intrinsic mode function, if not, repeatedly calculating an intermediate signal, replacing x (t) by using h (t), and recalculating according to h (t) =x (t) -m (t).
If satisfied, h (t) is an inherent mode function and let c 1 =h (t), then c 1 Is the first intrinsic mode function IMF1. Calculating the residual quantity, r 1 =x(t)-c 1 Repeating the steps by using r as a new time sequence to obtain a second intrinsic mode function IMF2, and performing iteration to obtain r 2 =r 1 -c 2 ,……,r n =r n-1 -c n . When the set constraint is satisfied, the iteration is terminated. Through an iterative process, the time series data x (t) is decomposed into n IMFs and a residual r n I.e.Wherein x (t) is original time series data, and the time series data is calendar history data consisting of influence factors such as cold storage device outlet pressure, cold storage device outlet temperature, pre-throttling pressure and pre-throttling air flow, namely liquefaction rate influence factors; t is the t-th influencing factor; c i (t) is the ith natural mode function IMFi of the t-th influencing factor; i is an index value; n is the number of the intrinsic mode functions; r is (r) n (t) is the nth residual amount of the nth influencing factor.
Specifically, the intrinsic mode function is required to be that the number of extreme points and zero crossing points of IMF sequence data is equal to or at most one difference, or the average value of upper and lower envelopes of the maximum value and the minimum value is zero; the constraint condition is the generated residual quantity 1 r i Is a monotonic function or less than a predetermined value. After the third historical data under different working conditions are decomposed through empirical mode, random processing is carried outDividing to obtain training data sets and test data sets under different working conditions; the training data set is used for training the liquefaction rate characterization model, and the test data set is mainly used for testing the accuracy of the liquefaction rate characterization model.
Specifically, the liquefaction rate characterization model is built mainly according to a training data set, and the adopted method is a width learning method. The width learning method is an algorithm based on a random vector function, maps input data into a characteristic node matrix, converts the characteristic node matrix into an enhanced node matrix through enhanced transformation, combines the characteristic node matrix and the enhanced node matrix into new input, and further constructs a new relation between the input and the output. Specifically, assume that the training dataset is X p×q P is the number of training samples, and q is the number of training sample characteristic parameters. (1) X is X p×q Mapping into a characteristic node matrix: mapping a bias matrix for the ith set of features; />Mapping a weight matrix for the ith group of features; />Mapping functions from the input nodes to the characteristic node layers; (2) the feature node matrix is mapped into an enhanced node matrix: /> Mapping a bias matrix for the j-th set of features; />Mapping a weight matrix for the j-th group of features; zeta type toy j Mapping functions from the feature node layer to the enhancement nodes;(3) feature nodes and add nodes to the output layer: y=w t [Z 1 ,Z 2 ,…|H 1 ,H 2 ,…],W t The liquefaction rate influencing factors are the outlet pressure of the cold accumulation device, the outlet temperature of the cold accumulation device, the temperature before throttling, the pressure before throttling and the air flow before throttling, and 5 factors are taken into account. X is X p×q It is these influencing factors, i.e. p is the number of training samples, assuming 10000 training samples, p=10000, q is the number of influencing factors, q=5. The input layer of the liquefaction rate characterization model is a liquefaction rate influence factor, the output layer is the liquefaction rate, the characteristic node matrix is random, and real-time data are input to the model at the later stage, so that the liquefaction rate value at the current moment can be obtained.
Specifically, the test data set is subjected to working condition division by adopting the same method to obtain test data sets under T working conditions, the test data sets are decomposed by adopting the empirical mode decomposition method, and the test data sets are input into the liquefaction rate characterization model under the corresponding working conditions to evaluate the liquefaction rate characterization model. The evaluation method of the characterization model of the liquid collecting rate of the test data adopts a decision coefficient, the decision coefficient is more than or equal to 0.9, which indicates that the characterization model has high precision, can be used for subsequent evaluation, and is lower than 0.9, and the characterization model of the liquid collecting rate needs to be retrained.
The characterization method can monitor the change process of the liquefaction rate in real time and discover the problem of abnormal liquefaction rate in time; the invention fully considers the problem of data characteristic difference under different working conditions, establishes a model which is more in line with the actual working conditions and has significance for characterization of the liquefaction rate, classifies the historical data by adopting different loads in front of the invention to obtain different training data sets and test data sets, the different loads represent different working conditions, and different liquefaction rate evaluation models are obtained after the different training data sets are trained, namely, a plurality of liquefaction rate evaluation models are obtained according to the different loads.
As a disclosed embodiment, the present invention also provides a liquefied compressed air energy storage liquefaction rate characterization system for implementing the method of the present invention, comprising:
the determining module is used for determining influencing factors related to the liquefaction rate according to the liquefied compressed air energy storage operation mechanism;
the preprocessing module is used for collecting the load of the power station, the liquefaction rate and the influence factors as first historical data, and preprocessing the first historical data to obtain second historical data;
the classification module is used for classifying the second historical data according to the load of the power station, determining different operation conditions and obtaining third historical data under different conditions;
the decomposition module is used for decomposing the third historical data and randomly dividing the decomposed data to obtain training data sets and test data sets under different working conditions;
the building module is used for building liquefaction rate characterization models under different working conditions according to the training data set;
the evaluation module is used for dividing the working condition of the test data set according to the load of the power station, decomposing the test data set, inputting the test data set into the liquefaction rate characterization model corresponding to different working conditions, and evaluating the accuracy of the liquefaction rate characterization model.
As disclosed embodiments, the present invention also provides an electronic apparatus including:
a memory storing executable instructions;
and a processor executing the executable instructions in the memory to implement the method.
While the foregoing description illustrates and describes the preferred embodiments of the present invention, it is to be understood that the invention is not limited to the forms disclosed herein, but is not to be construed as limited to other embodiments, and is capable of numerous other combinations, modifications and environments and is capable of changes or modifications within the scope of the inventive concept as expressed herein, either as a result of the foregoing teachings or as a result of the knowledge or technology of the relevant art. And that modifications and variations which do not depart from the spirit and scope of the invention are intended to be within the scope of the appended claims.
Claims (10)
1. A method for characterizing the stored energy liquefaction rate of liquefied compressed air, comprising the steps of:
s1, determining influence factors related to the liquefaction rate according to a liquefied compressed air energy storage operation mechanism;
s2, collecting the load, the liquefaction rate and the influence factors of the power station as first historical data, and preprocessing the first historical data to obtain second historical data;
s3, classifying the second historical data, determining different operation conditions, and obtaining third historical data under different conditions;
s4, decomposing the third historical data, and randomly dividing the decomposed data to obtain training data sets and test data sets under different working conditions;
s5, constructing liquefaction rate characterization models under different working conditions according to the training data set;
s6, dividing the working condition of the test data set, decomposing the test data set, inputting the test data set into the liquefaction rate characterization model corresponding to different working conditions for evaluating the liquefaction rate characterization model,
wherein the order of S2 and S3 can be exchanged.
2. The method of claim 1, wherein the factors related to the rate of liquefaction comprise regenerator exit pressure, regenerator exit temperature, pre-throttle pressure, and pre-throttle air flow.
3. The method for characterizing the energy storage and liquefaction rate of liquefied compressed air according to claim 1, wherein the different working conditions are divided according to the collected power station load into T working conditions, and T is a positive integer.
4. The method for characterizing the energy storage and liquefaction rate of the liquefied compressed air according to claim 1, wherein the decomposition in the step S4 adopts an empirical mode decomposition method, and the training data set and the test data set under different working conditions are obtained by performing random division after decomposing the third historical data.
5. The liquefied compressed air energy storage liquefaction rate characterization method according to claim 4, wherein the liquefaction rate characterization model is built according to the training data set and is realized by a width learning method, an input layer of the liquefaction rate characterization model is a liquefaction rate influence factor, and an output layer of the liquefaction rate characterization model is a liquefaction rate.
6. The method for characterizing the liquefied compressed air storage liquefaction rate according to claim 5, wherein the width learning method comprises the steps of: mapping the data in the input training data set into a characteristic node matrix, and converting the characteristic node matrix into an enhanced node matrix through enhanced transformation; and then combining the characteristic node matrix and the enhanced node matrix into a new input, and further constructing a new relation between the input and the output.
7. The method for characterizing the energy storage and liquefaction rate of liquefied compressed air according to claim 5, wherein:
and dividing the test data set into working conditions to obtain test data sets under T working conditions, decomposing the test data sets, and inputting the decomposed data into the liquefaction rate characterization model under the corresponding working conditions, thereby evaluating the accuracy of the liquefaction rate characterization model under different working conditions.
8. The method for characterizing the liquefied compressed air energy storage liquefaction rate according to claim 7, wherein the evaluation is expressed by a decision coefficient, and if the decision coefficient is greater than or equal to 0.9, the method indicates that the characterization model has high precision and can be used for subsequent evaluation; if the decision coefficient is less than 0.9, the characterization model needs to be retrained.
9. A liquefied compressed air stored energy liquefaction rate characterization system for implementing the method of any one of claims 1-8, comprising:
the determining module is used for determining influencing factors related to the liquefaction rate according to the liquefied compressed air energy storage operation mechanism;
the preprocessing module is used for collecting the load of the power station, the liquefaction rate and the influence factors as first historical data, and preprocessing the first historical data to obtain second historical data;
the classification module is used for classifying the second historical data, determining different operation conditions and obtaining third historical data under different conditions;
the decomposition module is used for decomposing the third historical data and randomly dividing the decomposed data to obtain training data sets and test data sets under different working conditions;
the building module is used for building liquefaction rate characterization models under different working conditions according to the training data set;
the evaluation module is used for dividing the working conditions of the test data set, decomposing the test data set, inputting the test data set into the liquefaction rate characterization model corresponding to different working conditions, and evaluating the accuracy of the liquefaction rate characterization model.
10. An electronic device, the electronic device comprising:
a memory storing executable instructions;
a processor executing the executable instructions in the memory to implement the method of any of claims 1-8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311548257.6A CN117592804A (en) | 2023-11-20 | 2023-11-20 | Liquefied compressed air energy storage liquefaction rate characterization method, system and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311548257.6A CN117592804A (en) | 2023-11-20 | 2023-11-20 | Liquefied compressed air energy storage liquefaction rate characterization method, system and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117592804A true CN117592804A (en) | 2024-02-23 |
Family
ID=89909311
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311548257.6A Pending CN117592804A (en) | 2023-11-20 | 2023-11-20 | Liquefied compressed air energy storage liquefaction rate characterization method, system and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117592804A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110222453A (en) * | 2019-06-14 | 2019-09-10 | 中国矿业大学 | A kind of compressor outlet parameter prediction modeling method based on width learning system |
CN112906974A (en) * | 2021-03-11 | 2021-06-04 | 东南大学 | Load electric quantity and carbon emission prediction and calibration method thereof |
CN115201408A (en) * | 2022-06-13 | 2022-10-18 | 东南大学 | Method for predicting concentration of sulfur dioxide at desulfurization outlet under all working conditions |
US20230145258A1 (en) * | 2021-11-11 | 2023-05-11 | Laboratory Corporation Of America Holdings | Predicting a Diagnostic Test Result From Patient Laboratory Testing History |
CN116204779A (en) * | 2023-04-19 | 2023-06-02 | 中能建数字科技集团有限公司 | Method, system and readable storage medium for judging operation state of energy storage salt cavern |
-
2023
- 2023-11-20 CN CN202311548257.6A patent/CN117592804A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110222453A (en) * | 2019-06-14 | 2019-09-10 | 中国矿业大学 | A kind of compressor outlet parameter prediction modeling method based on width learning system |
CN112906974A (en) * | 2021-03-11 | 2021-06-04 | 东南大学 | Load electric quantity and carbon emission prediction and calibration method thereof |
US20230145258A1 (en) * | 2021-11-11 | 2023-05-11 | Laboratory Corporation Of America Holdings | Predicting a Diagnostic Test Result From Patient Laboratory Testing History |
CN115201408A (en) * | 2022-06-13 | 2022-10-18 | 东南大学 | Method for predicting concentration of sulfur dioxide at desulfurization outlet under all working conditions |
CN116204779A (en) * | 2023-04-19 | 2023-06-02 | 中能建数字科技集团有限公司 | Method, system and readable storage medium for judging operation state of energy storage salt cavern |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20220214322A1 (en) | Air pollutants concentration forecasting method and apparatus and storage medium | |
US11874429B2 (en) | High-temperature disaster forecast method based on directed graph neural network | |
CN109142171B (en) | Urban PM10 concentration prediction method based on feature expansion and fusing with neural network | |
CN111860982A (en) | Wind power plant short-term wind power prediction method based on VMD-FCM-GRU | |
Zhou et al. | Holographic ensemble forecasting method for short-term power load | |
Yang et al. | Hybrid prediction method for wind speed combining ensemble empirical mode decomposition and Bayesian ridge regression | |
CN106251027B (en) | Electric load probability density Forecasting Methodology based on fuzzy support vector quantile estimate | |
CN107133695A (en) | A kind of wind power forecasting method and system | |
CN113657662B (en) | Downscaling wind power prediction method based on data fusion | |
CN113361761A (en) | Short-term wind power integration prediction method and system based on error correction | |
CN114119273A (en) | Park comprehensive energy system non-invasive load decomposition method and system | |
CN114462718A (en) | CNN-GRU wind power prediction method based on time sliding window | |
CN109412161A (en) | A kind of Probabilistic Load calculation method and system | |
CN113516271A (en) | Wind power cluster power day-ahead prediction method based on space-time neural network | |
CN114021483A (en) | Ultra-short-term wind power prediction method based on time domain characteristics and XGboost | |
CN116204779A (en) | Method, system and readable storage medium for judging operation state of energy storage salt cavern | |
CN112836876B (en) | Power distribution network line load prediction method based on deep learning | |
Ajagunsegun et al. | Machine learning-based system for managing energy efficiency of public buildings: An approach towards smart cities | |
CN113762591A (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM counterstudy | |
CN113111592A (en) | Short-term wind power prediction method based on EMD-LSTM | |
CN110276478B (en) | Short-term wind power prediction method based on segmented ant colony algorithm optimization SVM | |
CN116542385A (en) | Integrated learning method and system for wind and light power prediction | |
CN117592804A (en) | Liquefied compressed air energy storage liquefaction rate characterization method, system and electronic equipment | |
CN114234392B (en) | Air conditioner load fine prediction method based on improved PSO-LSTM | |
CN115796327A (en) | Wind power interval prediction method based on VMD (vertical vector decomposition) and IWOA-F-GRU (empirical mode decomposition) -based models |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |