CN116821736A - Modeling method and system for data-driven model of water chilling unit - Google Patents
Modeling method and system for data-driven model of water chilling unit Download PDFInfo
- Publication number
- CN116821736A CN116821736A CN202310672702.3A CN202310672702A CN116821736A CN 116821736 A CN116821736 A CN 116821736A CN 202310672702 A CN202310672702 A CN 202310672702A CN 116821736 A CN116821736 A CN 116821736A
- Authority
- CN
- China
- Prior art keywords
- data
- chiller
- training sample
- modeling
- operation index
- 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
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 128
- 238000000034 method Methods 0.000 title claims abstract description 72
- 238000012549 training Methods 0.000 claims abstract description 74
- 238000001914 filtration Methods 0.000 claims abstract description 6
- 238000004590 computer program Methods 0.000 claims description 13
- 238000012216 screening Methods 0.000 claims description 11
- 230000007812 deficiency Effects 0.000 claims description 9
- 230000009191 jumping Effects 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 3
- 230000002708 enhancing effect Effects 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 14
- 238000005457 optimization Methods 0.000 abstract description 7
- 238000013473 artificial intelligence Methods 0.000 abstract description 5
- 238000002360 preparation method Methods 0.000 abstract description 3
- 239000000498 cooling water Substances 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000010438 heat treatment Methods 0.000 description 3
- 238000009423 ventilation Methods 0.000 description 3
- 238000004378 air conditioning Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 235000002918 Fraxinus excelsior Nutrition 0.000 description 1
- 244000181980 Fraxinus excelsior Species 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Landscapes
- Feedback Control In General (AREA)
Abstract
The application discloses a modeling method and a system for a data driving model of a water chiller, wherein the method comprises the steps of obtaining data of operation indexes of the water chiller, and constructing a training sample database through clustering filtering and enhancement; judging whether the training sample database meets the modeling requirement of the data driving model according to the operation index threshold value determined based on the running condition of the water chiller, and if so, training the data driving model of the water chiller by adopting the training sample database to predict the power of the water chiller. The application utilizes the combination of a plurality of methods of 'rules of cold machine modeling' + 'clustering' + 'enhancement', can enhance the minimum cost of a sample to realize the data preparation and training process of data driving model modeling, realize the automatic model creation under a small sample, quickly realize the participation of the data driving model in the optimization of an energy system, and increase the landing speed of artificial intelligence in the optimization of the energy system.
Description
Technical Field
The application relates to a water chilling unit data driving technology in the field of heating ventilation and air conditioning, in particular to a modeling method and a system for a data driving model of a water chilling unit.
Background
The water chiller is the main energy-consuming equipment of the heating ventilation air conditioning system, the highest proportion can reach 60% -70%, and the energy saving and fault diagnosis and careful maintenance of the soil-necessary water chiller become more and more important. Meanwhile, as more new energy sources are connected into the power grid along with the construction of a novel power system, the unstable characteristics of the new energy sources such as photovoltaic and wind energy cause great challenges to the stable operation of the power grid, and the adjustable load performance of the user side energy source system, especially the heating ventilation air conditioner, can be fully exerted, so that the unstable operation of the power grid can be reduced to the greatest extent. Therefore, the energy-saving characteristic and the adjustable load characteristic of the water chilling unit are fully exerted, and the method has great significance for energy conservation, carbon reduction and stable power grid operation.
The artificial intelligent model driving method is a main method for modeling the water chilling unit, the water chilling unit is used as an energy-saving and adjustable load resource, the description is clear about the performance characteristics of the water chilling unit, namely, the change between the energy consumption and the operation parameters, and the data driving model is generally adopted to create the performance description of the water chilling unit. The data driving model describing the water chilling unit is divided into a black box model and an ash box model. The black box model is a pure data driving model, and along with the development of artificial intelligence, the black box model is widely applied in actual engineering, but has poor interpretation. The ash box model establishes a physical model of the water chilling unit, factors in the physical model are determined in a data statistics mode, and the more common ash box model comprises SL, BQ, MP, GNU, GNS and LS models. However, in an actual project, a water chiller model creation process is complex, multiplexing rates of different water chiller model creation methods are low, and mainly because of uneven quality of running data of different water chillers in the actual project, the quality of the data is poor, and if the method is suitable for different projects, the data difference of different water chillers is large, and the water chiller model can be quickly created, the method is a long-standing technical challenge.
Disclosure of Invention
The application aims to solve the technical problems: aiming at the problems in the prior art, the application provides a modeling method and a modeling system for a data driving model of a water chilling unit, which can effectively improve the accuracy of power prediction of the data driving model through clustering filtering and enhancement, can flexibly adapt to water chilling units of different types and scales, and has stronger universality.
In order to solve the technical problems, the application adopts the following technical scheme:
a data-driven model modeling method for a chiller, comprising:
s101, acquiring data of operation indexes of a water chilling unit;
s102, constructing a training sample database aiming at the data of the operation index through cluster filtering and enhancement;
s103, judging whether the training sample database meets the modeling requirement of the data driving model according to an operation index threshold value determined based on the operation condition of the water chilling unit, and if not, jumping to the step S101; otherwise, jump to step S104;
s104, training a data driving model of the water chiller by using the training sample database to predict the power of the water chiller.
Optionally, the data of the operation index of the chiller in step S101 includes a power of the chiller, a chiller water outlet temperature, a chiller water inlet temperature, a chiller water outlet temperature, and a load factor, and the power is used as a tag of a data driving model of the chiller, and the chiller water outlet temperature, the chiller water inlet temperature, the chiller water outlet temperature, and the load factor are used as inputs of the data driving model of the chiller.
Optionally, step S102 includes:
s201, multi-dimensional clustering is carried out on the data of the operation index of the water chilling unit;
s202, screening a cluster obtained by multi-dimensional clustering;
s203, aiming at the cluster clusters remained after screening, calculating the variance of each operation index in the cluster clusters;
s204, comparing the variance of each operation index with a set variance threshold to judge whether each operation index meets the condition, and deleting the data of each operation index if the operation index does not meet the condition;
s205, constructing a training sample database by using the data of the remaining operation indexes;
s206, judging whether the number of the training sample databases is larger than or equal to a set sample number threshold value or not, if so, step S103; otherwise, the sample deficiency rate is calculated by subtracting the training sample number of the training sample database from the set sample number threshold, dividing the training sample number by the set sample number threshold, if the sample deficiency rate is lower than the preset threshold, performing sample enhancement on the training sample database to enable the training sample number in the training sample database to be not less than the set sample number threshold, and if the sample deficiency rate is greater than or equal to the preset threshold, jumping to the step S101 to continuously acquire the data of the operation index.
Optionally, in step S201, the clustering method used in the multi-dimensional clustering is kmeans multi-dimensional clustering.
Optionally, when screening the cluster clusters obtained by multi-dimensional clustering in step S202, the method includes calculating covariance of each cluster obtained by multi-dimensional clustering, and deleting a cluster if the covariance of a cluster is greater than a set value, so that the screened data of the operation index only remains the cluster with the covariance less than the set value.
Optionally, when the training sample database is subjected to sample enhancement in step S206, the method includes sorting the data of each running index in the training sample database, and then generating new running index data between the data of two adjacent running indexes in the sorting result by interpolation or smoothing until the number of training samples in the training sample database is not less than the set sample number threshold.
Optionally, determining in step S103 whether the training sample database meets the modeling requirement of the data-driven model includes: comparing all data of the operation index with a corresponding operation index threshold value determined based on the running condition of the water chilling unit aiming at each operation index in the training sample database, judging that the operation index meets the modeling requirement of the data driving model if all the data of the operation index are within the operation index threshold value, judging that the training sample database meets the modeling requirement of the data driving model if all the operation indexes meet the modeling requirement of the data driving model, and otherwise judging that the training sample database does not meet the modeling requirement of the data driving model.
Optionally, after step S104, the method further includes:
s301, acquiring data of operation indexes of a water chilling unit;
s302, carrying out power prediction on the data of the running index of the water chilling unit by using a trained data driving model of the water chilling unit, thereby obtaining the power of the water chilling unit.
In addition, the application also provides a data driving model modeling system for the water chiller, which comprises a microprocessor and a memory which are connected with each other, wherein the microprocessor is programmed or configured to execute the data driving model modeling method for the water chiller.
Furthermore, the application provides a computer readable storage medium having a computer program stored therein, the computer program being for being programmed or configured by a microprocessor to perform the data driven model modeling method for a chiller.
Compared with the prior art, the application has the following advantages:
1. aiming at the problems that the running index number of the water chilling unit is large and the unsteady state data is complex, the training sample database is constructed by clustering filtering and enhancing the running index data, so that the interference of the complex unsteady state data on the data driving model of the water chilling unit can be effectively eliminated, and the data driving model modeling under the condition of small samples can be realized.
2. According to the method, whether the training sample database meets the modeling requirement of the data driving model is judged according to the operation index threshold value determined based on the running condition of the water chiller, and through the rule of the modeling process of the water chiller, on one hand, the data obtained by the modeling method of the data driving model for the water chiller in the embodiment can be more accurate, and on the other hand, the rule of the modeling process of the water chiller is determined by combining with a specific water chiller, so that the universality of the modeling method of the data driving model for the water chiller in the embodiment is better, and the modeling method of the data driving model for the water chiller can be flexibly adapted to water chillers of different types and scales.
Drawings
FIG. 1 is a flow chart of a method according to an embodiment of the application.
Detailed Description
As shown in fig. 1, the data driving model modeling method for a water chiller according to the present embodiment includes:
s101, acquiring data of operation indexes of a water chilling unit;
s102, constructing a training sample database aiming at the data of the operation index through cluster filtering and enhancement;
s103, judging whether the training sample database meets the modeling requirement of the data driving model according to an operation index threshold value determined based on the operation condition of the water chilling unit, and if not, jumping to the step S101; otherwise, jump to step S104;
s104, training a data driving model of the water chiller by using the training sample database to predict the power of the water chiller.
In general, the power prediction of the water chiller may employ two operation indexes, namely, the water outlet temperature of the water chiller and the water inlet temperature of the water chiller, because the two operation indexes are directly related physical parameters of the power of the water chiller. However, it is found through research that only two operation indexes of the water outlet temperature of the cooling machine and the water inlet temperature of the cooling machine are used, and it is often difficult to meet the power accurate prediction requirement of the data driving model. In order to solve the problems, the analysis finds that: the cooling water of the water chilling unit actually forms part of the power loss of the water chilling unit, and the cooling water can be expressed as the part of the power loss of the water chilling unit directly influenced by the inlet water temperature and the outlet water temperature of the cooling water; in addition, the load factor (load percentage, load factor refers to the percentage of the average load to the maximum load in the statistical period) of the water chiller also affects the power condition of the water chiller, because the load factor expresses the power state of the water chiller. Based on the above findings, as an alternative implementation manner, the data of the operation index of the chiller in step S101 in this embodiment includes the power of the chiller, the chiller water outlet temperature, the chiller water inlet temperature, the chiller water outlet temperature, and the load factor (load percentage), and the power is used as the tag of the data driving model of the chiller, and the chiller water outlet temperature, the chiller water inlet temperature, the chiller water outlet temperature, and the load factor are used as the input of the data driving model of the chiller. By adding the three indexes of the cooling water inlet temperature, the cooling water outlet temperature and the load rate into the running index of the water chilling unit for predicting the power, the accurate modeling of the index required by the accurate power prediction requirement of the data driving model can be realized, and the accuracy of the power prediction of the data driving model can be improved.
In this embodiment, step S102 includes:
s201, multi-dimensional clustering is carried out on the data of the operation index of the water chilling unit;
s202, screening a cluster obtained by multi-dimensional clustering;
s203, aiming at the cluster clusters remained after screening, calculating the variance of each operation index in the cluster clusters;
s204, comparing the variance of each operation index with a set variance threshold to judge whether each operation index meets the condition, and deleting the data of each operation index if the operation index does not meet the condition;
s205, constructing a training sample database by using the data of the remaining operation indexes;
s206, judging whether the number of the training sample databases is larger than or equal to a set sample number threshold value or not, if so, step S103; otherwise, the sample deficiency rate is calculated by subtracting the training sample number of the training sample database from the set sample number threshold, dividing the training sample number by the set sample number threshold, if the sample deficiency rate is lower than the preset threshold, performing sample enhancement on the training sample database to enable the training sample number in the training sample database to be not less than the set sample number threshold, and if the sample deficiency rate is greater than or equal to the preset threshold, jumping to the step S101 to continuously acquire the data of the operation index. The range of the enhancement of the sample data is limited by the sample shortage rate, so that the data distortion of the training sample database caused by excessive enhancement data can be avoided.
Referring to steps S201 to S206, the method of the embodiment rapidly screens modeling data clusters by using a clustering algorithm for the data of the operation index of the water chiller, and determines by combining the multidimensional variance to form a minimum data set and a data change range required for training the data driving model of the water chiller as references, so that the efficient screening data of the operation index of the water chiller can be realized. In addition, it is found that the variance of the data sample is smaller and the information is less due to the fact that the intelligent regulation and control measures are not adopted in the initial investment and actual operation process of the water chilling unit, step S102 of the embodiment comprises the steps of comparing the variance of each operation index with a set variance threshold to judge whether each operation index meets the condition or not, and deleting the data of each operation index if the variance of each operation index does not meet the condition, so that effective modeling information can be formed in small working conditions such as a unit switch and state adjustment, and in addition, value data fitting and equipment model establishment can be rapidly recognized by adopting a clustering comprehensive method in the modeling process.
It should be noted that, the multi-dimensional clustering performed in step S201 may employ a desired clustering algorithm as needed. For example, as an alternative implementation manner, in the multi-dimensional clustering in step S201 of this embodiment, the clustering method used is kmeans multi-dimensional clustering.
In this embodiment, when the cluster obtained by multi-dimensional clustering is screened in step S202, the method includes calculating covariance of each cluster obtained by multi-dimensional clustering, and deleting a cluster if the covariance of a cluster is greater than a set value, so that the screened data of the operation index only remains the cluster whose covariance is less than the set value.
It should be noted that, based on the teaching of step S203, the person skilled in the art may also use other statistical indicators instead of the variance as needed, which may have more or less similar effects. It should be noted that, considering the natural differences of different operation indexes, as a preferred implementation manner, in this embodiment, the variance thresholds are in one-to-one correspondence with the operation indexes, and all or part of the variance thresholds corresponding to all the operation indexes are different, so that accurate operation index screening can be implemented.
In this embodiment, when the training sample database is subjected to sample enhancement in step S206, the method includes sorting the data of each operation index in the training sample database, generating new operation index data between the data of two adjacent operation indexes in the sorting result by interpolation or smoothing, until the number of training samples in the training sample database is not less than the set sample number threshold, and further reducing the number of samples by performing sample enhancement, thereby realizing data-driven model modeling under the condition of small samples.
In this embodiment, the step S103 of determining whether the training sample database meets the modeling requirement of the data driving model includes: comparing all data of the operation index with a corresponding operation index threshold value determined based on the running condition of the water chilling unit aiming at each operation index in the training sample database, judging that the operation index meets the modeling requirement of the data driving model if all the data of the operation index are within the operation index threshold value, judging that the training sample database meets the modeling requirement of the data driving model if all the operation indexes meet the modeling requirement of the data driving model, and otherwise judging that the training sample database does not meet the modeling requirement of the data driving model. In this embodiment, based on the running condition of the water chiller, the running index threshold including the chiller outlet water temperature, the chiller inlet water temperature, the cooling water outlet water temperature and the load factor of the water chiller may be determined. The rule of the cold machine modeling process in the step S103 is to screen whether the clustered multivariable data meets modeling requirements, the rule is formulated according to the threshold value determined by the cold machine model parameters such as water outlet temperature, water inlet temperature, cooling water inlet temperature and load factor in the normal operation range of the mechanism, such as water inlet temperature (10-15), after the multivariable clustering, the threshold value determined according to the cold machine mechanism operation range is used as a basis for judging whether the clustered data meets the modeling requirements. According to the rules of the cold machine modeling process, on one hand, the data obtained by the data driving model modeling method for the cold machine set can be more accurate, and on the other hand, the rules of the cold machine modeling process are determined by combining with a specific cold machine set, so that the data driving model modeling method for the cold machine set is better in universality and can be flexibly suitable for cold machines of different types and scales.
Step S104 adopts a training sample database to train a data driving model of the water chiller, and the trained data driving model of the water chiller can be used for predicting the power (real-time power) of the water chiller. After step S104 in this embodiment, the method further includes:
s301, acquiring data of operation indexes of a water chiller, namely: data of cold machine water outlet temperature, cold machine water inlet temperature, cooling water outlet temperature and load rate;
s302, power prediction is carried out on the data of the running index of the water chilling unit by utilizing a trained data driving model of the water chilling unit, so that the power (real-time power) of the water chilling unit is obtained, and the method can be used for various running optimizations and other applications of the water chilling unit according to the predicted power and electric energy consumption.
In summary, the modeling method of the data driving model for the chiller in this embodiment combines multiple methods such as clustering, enhancement, and data driving model to solve the problem of longer creation process of the chiller model based on AI intelligent control, and can realize data preparation for quick model creation and training of the data driving model. The data driving model obtained by the method can be used for accurately predicting the power (real-time power) of the water chilling unit. According to the method, artificial intelligence and big data are comprehensively utilized, the clustering method is adopted to conduct automatic clustering analysis on main input parameters of a water chilling unit model respectively, statistical indexes of model data parameters are obtained, the difference of data created by the model is judged according to the automatically set threshold value, and the clustered data are automatically selected and input into the model to be created. The method of the embodiment has the advantages of replicability, dynamic adjustment and self-adaptive update, can accelerate the creation process and the automation process of the water chilling unit model, and lays a model foundation for energy-saving optimization and flexible regulation and control based on the water chilling unit model. In the general practice of the art for training sample databases, due to lack of optimization of data, the training data of the data-driven model requires very long cold machine running time to achieve a large amount of data accumulation, but as a result, artificial intelligence landing speed can be greatly prolonged. The application utilizes the combination of a plurality of methods of 'rules of cold machine modeling' + 'clustering' + 'enhancement', can enhance the minimum cost of samples to realize the data preparation and training process of data driving model modeling, can realize automatic model creation under small samples, can quickly realize the participation of the data driving model in the optimization of an energy system, and can increase the landing speed of artificial intelligence in the optimization of the energy system.
In addition, the application also provides a data driving model modeling system for the water chiller, which comprises a microprocessor and a memory which are connected with each other, wherein the microprocessor is programmed or configured to execute the data driving model modeling method for the water chiller.
Furthermore, the application provides a computer readable storage medium having a computer program stored therein, the computer program being for being programmed or configured by a microprocessor to perform the data driven model modeling method for a chiller.
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-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. 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.
The above description is only a preferred embodiment of the present application, and the protection scope of the present application is not limited to the above examples, and all technical solutions belonging to the concept of the present application belong to the protection scope of the present application. It should be noted that modifications and adaptations to the present application may occur to one skilled in the art without departing from the principles of the present application and are intended to be within the scope of the present application.
Claims (10)
1. A data-driven modeling method for a chiller, comprising:
s101, acquiring data of operation indexes of a water chilling unit;
s102, constructing a training sample database aiming at the data of the operation index through cluster filtering and enhancement;
s103, judging whether the training sample database meets the modeling requirement of the data driving model according to an operation index threshold value determined based on the operation condition of the water chilling unit, and if not, jumping to the step S101; otherwise, jump to step S104;
s104, training a data driving model of the water chiller by using the training sample database to predict the power of the water chiller.
2. The modeling method of a data driving model for a chiller according to claim 1, wherein the data of the operation index of the chiller in step S101 includes a power of the chiller, a chiller water outlet temperature, a chiller water inlet temperature, a chiller water outlet temperature, and a load factor, and wherein the power is used as a tag of the data driving model of the chiller, and the chiller water outlet temperature, the chiller water inlet temperature, the chiller water outlet temperature, and the load factor are used as inputs of the data driving model of the chiller.
3. The method of modeling a data driven model for a chiller according to claim 1, wherein step S102 comprises:
s201, multi-dimensional clustering is carried out on the data of the operation index of the water chilling unit;
s202, screening a cluster obtained by multi-dimensional clustering;
s203, aiming at the cluster clusters remained after screening, calculating the variance of each operation index in the cluster clusters;
s204, comparing the variance of each operation index with a set variance threshold to judge whether each operation index meets the condition, and deleting the data of each operation index if the operation index does not meet the condition;
s205, constructing a training sample database by using the data of the remaining operation indexes;
s206, judging whether the number of the training sample databases is larger than or equal to a set sample number threshold value or not, if so, step S103; otherwise, the sample deficiency rate is calculated by subtracting the training sample number of the training sample database from the set sample number threshold, dividing the training sample number by the set sample number threshold, if the sample deficiency rate is lower than the preset threshold, performing sample enhancement on the training sample database to enable the training sample number in the training sample database to be not less than the set sample number threshold, and if the sample deficiency rate is greater than or equal to the preset threshold, jumping to the step S101 to continuously acquire the data of the operation index.
4. The modeling method of a data driving model for a chiller according to claim 3, wherein the clustering method used in the step S201 is kmeans multidimensional clustering.
5. The modeling method of a data driving model for a water chiller according to claim 3, wherein when screening the clusters obtained by multi-dimensional clustering in step S202, the method comprises calculating covariance of each cluster obtained by multi-dimensional clustering, and deleting a cluster if the covariance of a cluster is greater than a set value, so that only clusters with covariance smaller than the set value are reserved in the data of the operation index after screening.
6. The modeling method of a data driven model for a chiller according to claim 3, wherein the step S206 of enhancing the training sample database includes sorting the data of each operation index in the training sample database, and generating new operation index data between the data of two adjacent operation indexes in the sorting result by interpolation or smoothing until the number of training samples in the training sample database is not less than the set sample number threshold.
7. The modeling method of a data-driven model for a chiller according to claim 1, wherein determining whether the training sample database meets the modeling requirement of the data-driven model in step S103 comprises: comparing all data of the operation index with a corresponding operation index threshold value determined based on the running condition of the water chilling unit aiming at each operation index in the training sample database, judging that the operation index meets the modeling requirement of the data driving model if all the data of the operation index are within the operation index threshold value, judging that the training sample database meets the modeling requirement of the data driving model if all the operation indexes meet the modeling requirement of the data driving model, and otherwise judging that the training sample database does not meet the modeling requirement of the data driving model.
8. The method of modeling a data driven model for a chiller according to claim 1 further comprising, after step S104:
s301, acquiring data of operation indexes of a water chilling unit;
s302, carrying out power prediction on the data of the running index of the water chilling unit by using a trained data driving model of the water chilling unit, thereby obtaining the power of the water chilling unit.
9. A data driven model modeling system for a chiller, comprising a microprocessor and a memory interconnected, wherein the microprocessor is programmed or configured to perform the data driven model modeling method for a chiller of any of claims 1-8.
10. A computer readable storage medium having a computer program stored therein, wherein the computer program is for programming or configuring by a microprocessor to perform the data driven model modeling method for a chiller as claimed in any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310672702.3A CN116821736A (en) | 2023-06-07 | 2023-06-07 | Modeling method and system for data-driven model of water chilling unit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310672702.3A CN116821736A (en) | 2023-06-07 | 2023-06-07 | Modeling method and system for data-driven model of water chilling unit |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116821736A true CN116821736A (en) | 2023-09-29 |
Family
ID=88125062
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310672702.3A Pending CN116821736A (en) | 2023-06-07 | 2023-06-07 | Modeling method and system for data-driven model of water chilling unit |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116821736A (en) |
-
2023
- 2023-06-07 CN CN202310672702.3A patent/CN116821736A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111237989B (en) | Building ventilation air conditioner control method and device based on load prediction | |
| CN108317670B (en) | Refrigeration system energy-saving control method and system based on machine learning | |
| CN109890176B (en) | Device for optimizing energy consumption efficiency of machine room based on artificial intelligence | |
| Wang et al. | A practical approach to chiller plants’ optimisation | |
| CN110260470B (en) | Central air conditioner parallel connection cold machine load optimization distribution method based on swarm intelligence architecture | |
| CN114330852B (en) | Energy-saving optimization method and device for tail end air conditioning system of integrated data center cabinet | |
| CN106873372B (en) | Reservoir regulation for flood control optimization method based on the control of Flood Control Dispatch data adaptive | |
| CN107143981B (en) | A kind of controlling system of central air conditioner and method | |
| CN113739365A (en) | Central air-conditioning cold station group control energy-saving control method, device, equipment and storage medium | |
| CN107036238A (en) | Intelligent energy-saving control method for dynamically predicting external air and load | |
| CN109507961A (en) | A kind of semiconductor production line dynamic load uniform feeding control method | |
| CN113110057A (en) | Heating power station energy-saving control method based on artificial intelligence and intelligent decision system | |
| CN112070353A (en) | Method and system for accurately detecting energy efficiency of data center | |
| CN116266253A (en) | Optimization control method, system and computer readable storage medium for air conditioner parameters | |
| CN115983438A (en) | Method and device for determining operation strategy of data center terminal air conditioning system | |
| CN106016605A (en) | Parallel cooling pump control method and device thereof based on genetic algorithm | |
| CN113790516B (en) | Global optimization energy-saving control method and system for central air-conditioning refrigeration station and electronic equipment | |
| CN113110056B (en) | Heat supply intelligent decision-making method and intelligent decision-making machine based on artificial intelligence | |
| CN113225994A (en) | Intelligent air conditioner control method facing data center | |
| CN116430748B (en) | Energy-saving control method and system based on intelligent equipment | |
| CN116028193B (en) | Big data task dynamic high-energy-efficiency scheduling method and system for mixed part cluster | |
| CN116821736A (en) | Modeling method and system for data-driven model of water chilling unit | |
| CN105423492A (en) | Machine room monitoring system and method | |
| CN104991443A (en) | Fuzzy control method based on adaptive domain partitioning | |
| CN116963461A (en) | Energy saving method and device for machine room air conditioner |
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 |