CN113701275B - Method and device for predicting cold accumulation amount of ice cold accumulation air conditioner based on machine learning - Google Patents
Method and device for predicting cold accumulation amount of ice cold accumulation air conditioner based on machine learning Download PDFInfo
- Publication number
- CN113701275B CN113701275B CN202111027278.4A CN202111027278A CN113701275B CN 113701275 B CN113701275 B CN 113701275B CN 202111027278 A CN202111027278 A CN 202111027278A CN 113701275 B CN113701275 B CN 113701275B
- Authority
- CN
- China
- Prior art keywords
- cold accumulation
- cold
- prediction model
- day
- data
- 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.)
- Active
Links
- 238000009825 accumulation Methods 0.000 title claims abstract description 228
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000010801 machine learning Methods 0.000 title claims abstract description 21
- 238000012360 testing method Methods 0.000 claims abstract description 48
- 238000011156 evaluation Methods 0.000 claims abstract description 45
- 238000012549 training Methods 0.000 claims abstract description 41
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 12
- 238000007637 random forest analysis Methods 0.000 claims abstract description 10
- 230000004044 response Effects 0.000 claims abstract description 7
- 238000012937 correction Methods 0.000 claims description 37
- 238000013210 evaluation model Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 7
- 238000004378 air conditioning Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 1
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F5/00—Air-conditioning systems or apparatus not covered by F24F1/00 or F24F3/00, e.g. using solar heat or combined with household units such as an oven or water heater
- F24F5/0007—Air-conditioning systems or apparatus not covered by F24F1/00 or F24F3/00, e.g. using solar heat or combined with household units such as an oven or water heater cooling apparatus specially adapted for use in air-conditioning
- F24F5/0017—Air-conditioning systems or apparatus not covered by F24F1/00 or F24F3/00, e.g. using solar heat or combined with household units such as an oven or water heater cooling apparatus specially adapted for use in air-conditioning using cold storage bodies, e.g. ice
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F5/00—Air-conditioning systems or apparatus not covered by F24F1/00 or F24F3/00, e.g. using solar heat or combined with household units such as an oven or water heater
- F24F5/0007—Air-conditioning systems or apparatus not covered by F24F1/00 or F24F3/00, e.g. using solar heat or combined with household units such as an oven or water heater cooling apparatus specially adapted for use in air-conditioning
- F24F5/0017—Air-conditioning systems or apparatus not covered by F24F1/00 or F24F3/00, e.g. using solar heat or combined with household units such as an oven or water heater cooling apparatus specially adapted for use in air-conditioning using cold storage bodies, e.g. ice
- F24F2005/0032—Systems storing energy during the night
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/14—Thermal energy storage
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Mechanical Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Combustion & Propulsion (AREA)
- Chemical & Material Sciences (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Tourism & Hospitality (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Operations Research (AREA)
- Computing Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Fuzzy Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Development Economics (AREA)
- Evolutionary Biology (AREA)
- Sustainable Development (AREA)
Abstract
The invention discloses a method and a device for predicting cold storage capacity of an ice storage air conditioner based on machine learning, wherein a cold storage capacity data set is constructed according to historical data, and is divided into a training set and a test set; inputting the training set as input data into a cold accumulation prediction model, and training the cold accumulation prediction model; evaluating the trained cold accumulation prediction model by taking the test set as input data to obtain an evaluation score; in response to the evaluation score being larger than or equal to the threshold value, the cold accumulation amount of the cold accumulation day is predicted by adopting a trained cold accumulation amount prediction model; according to the invention, a random forest regression algorithm is adopted as a cold accumulation prediction model, historical cold accumulation, highest temperature and lowest temperature are used as input data, so that cold accumulation prediction can be realized, the cold accumulation prediction model is evaluated, and the cold accumulation prediction accuracy can be further improved.
Description
Technical Field
The invention belongs to a method and a device for predicting cold storage capacity of an ice storage air conditioner, and particularly relates to a method and a device for predicting cold storage capacity of an ice storage air conditioner based on machine learning.
Background
Due to the improvement of industrial development and the living standard of people's material culture, the popularization rate of air conditioners is increased year by year, the power consumption is increased rapidly, the peak power is short, and the off-peak power cannot be fully applied. Therefore, how to shift the peak power demand to balance the power supply and improve the effective utilization of the electric energy becomes a problem that many countries pay attention to solve at present.
The aggressiveness of using off-peak power is further driven by the policy of "time-of-use price" and certain incentives. This makes the off-peak cold-storage technology be regarded and developed. The ice storage air conditioner makes ice by utilizing the night low-valley load electric power and stores the ice in an ice storage device, and the ice is melted in the daytime to release the stored cold so as to reduce the electric load of the air conditioner and the installed capacity of an air conditioning system during the peak time of a power grid. Since the ice storage air conditioner is used after ice making, the cold storage amount needs to be predicted in the process of ice making.
In the aspect of cold accumulation prediction, the cold accumulation amount of the day after the previous day is generally predicted, and a worker of the ice cold accumulation air conditioner subjectively predicts the cold accumulation amount of the day after the previous day according to specific use requirements so as to determine the cold accumulation amount of the day after the previous day. However, this method extremely depends on the personal ability of experienced workers, once the workers leave the post, the empirical cold accumulation prediction method is difficult to be completely handed over during work handover, the workers need to adjust every day, time and labor are consumed, errors are large, energy is wasted, and even the effect of 'shifting peaks and filling valleys' is difficult to achieve.
Disclosure of Invention
The invention aims to provide a method and a device for predicting cold accumulation of an ice cold accumulation air conditioner based on machine learning.
The invention adopts the following technical scheme: a method for predicting cold storage capacity of an ice storage air conditioner based on machine learning comprises the following steps:
according to historical data, a cold accumulation data set is constructed, and the cold accumulation data set is divided into a training set and a testing set; the training set and the test set respectively comprise a plurality of groups of cold accumulation data groups, and each group of cold accumulation data group respectively comprises historical cold accumulation amount before a cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day and the cold accumulation amount of the cold accumulation day;
inputting the training set as input data into a cold accumulation prediction model, and training the cold accumulation prediction model; the cold accumulation prediction model is selected as a random forest regression algorithm;
evaluating the trained cold accumulation prediction model by taking the test set as input data to obtain an evaluation score;
and in response to the evaluation score being larger than or equal to the threshold, predicting the cold accumulation amount of the cold accumulation day by adopting a trained cold accumulation amount prediction model by taking the historical cold accumulation amount before the cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day as input data.
And further, responding to the evaluation score smaller than the threshold value, adjusting the hyper-parameter of the cold accumulation prediction model, and continuing to train the cold accumulation prediction model by using the training set until the evaluation score is larger than or equal to the threshold value.
Further, the evaluation of the trained cold accumulation prediction model comprises:
respectively correcting the actual cold storage value in the cold storage array and the predicted cold storage value of the cold storage prediction model to obtain a real correction value and a prediction correction value;
calculating the ratio of the real correction value to the predicted correction value;
summing the ratio sum corresponding to each cold accumulation data group in the test set;
an evaluation score is determined from the ratio sum.
Further, evaluating the trained cold accumulation prediction model through an evaluation model; the evaluation model specifically comprises:
wherein f (x) is an evaluation score, n is the number of the cold accumulation data sets in the test set, and y i The real value of the cold accumulation amount in the ith cold accumulation array,
the historical cold accumulation amount before the cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day in the ith cold accumulation array are used as cold accumulation amount prediction values input into the cold accumulation amount prediction model, and lambda is a correction coefficient of the test set.
Further, a correction coefficient is determined according to the cold accumulation amount in the test set.
Further, λ =10 is adopted [lg(x)] Calculating to obtain a sub-correction coefficient corresponding to each nonzero cold accumulation in the test set, and forming a sub-correction coefficient set; wherein x is the cold accumulation amount in the test set, and the cold accumulation amount is not 0;
a mode is selected from the correction coefficient set, and the mode is used as a correction coefficient.
Further, the historical cold accumulation amount comprises the cold accumulation amount of m ice storage air-conditioning working days which are before and adjacent to the cold accumulation day, wherein m =7.
Further, the ratio of the cold storage data set in the training set and the test set is 4:1.
In another technical scheme of the present invention, a device for predicting cold storage capacity of an ice storage air conditioner based on machine learning comprises:
the building module is used for building a cold accumulation data set according to the historical data and dividing the cold accumulation data set into a training set and a test set; the training set and the test set respectively comprise a plurality of groups of cold accumulation data groups, and each group of cold accumulation data group respectively comprises historical cold accumulation amount before a cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day and the cold accumulation amount of the cold accumulation day;
the training module is used for inputting the training set as input data into the cold accumulation prediction model and training the cold accumulation prediction model; the cold accumulation prediction model is selected as a random forest regression algorithm;
the evaluation module is used for evaluating the trained cold accumulation prediction model by taking the test set as input data to obtain an evaluation score;
and the prediction module is used for responding to the evaluation score which is more than or equal to the threshold value, taking the historical cold accumulation amount before the cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day as input data, and predicting the cold accumulation amount of the cold accumulation day by adopting a trained cold accumulation amount prediction model.
The other technical scheme of the invention is as follows: the device comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the cold storage amount prediction method of the ice storage air conditioner based on the machine learning is realized when the processor executes the computer program.
The beneficial effects of the invention are: according to the invention, a random forest regression algorithm is adopted as a cold accumulation prediction model, historical cold accumulation, highest temperature and lowest temperature are used as input data, so that cold accumulation prediction can be realized, the cold accumulation prediction model is evaluated, and the cold accumulation prediction accuracy can be further improved.
Drawings
Fig. 1 is a flow chart of a method for predicting cold accumulation of an ice cold accumulation air conditioner based on machine learning according to the invention;
fig. 2 is a flowchart of a device for predicting cold storage capacity of an ice storage air conditioner based on machine learning according to an embodiment of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
The invention discloses a method for predicting cold accumulation of an ice cold accumulation air conditioner based on machine learning, which comprises the following steps as shown in figure 1: step S110, constructing a cold accumulation data set according to historical data, and dividing the cold accumulation data set into a training set and a testing set; the training set and the test set respectively comprise a plurality of groups of cold accumulation data groups, and each group of cold accumulation data group respectively comprises historical cold accumulation amount before a cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day and the cold accumulation amount of the cold accumulation day; step S120, inputting the training set as input data into a cold accumulation prediction model, and training the cold accumulation prediction model; the cold accumulation prediction model is selected as a random forest regression algorithm; step S130, evaluating the trained cold accumulation prediction model by taking the test set as input data to obtain an evaluation score; and S140, responding to the evaluation score being larger than or equal to the threshold, taking the historical cold accumulation amount before the cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day as input data, and predicting the cold accumulation amount of the cold accumulation day by adopting a trained cold accumulation amount prediction model.
According to the invention, a random forest regression algorithm is adopted as a cold accumulation prediction model, historical cold accumulation, highest temperature and lowest temperature are used as input data, so that cold accumulation prediction can be realized, the cold accumulation prediction model is evaluated, and the cold accumulation prediction accuracy can be further improved.
In addition, in the embodiment of the invention, in response to the evaluation score being smaller than the threshold value, the hyper-parameter of the cold accumulation prediction model is adjusted, the training set is continuously used for training the cold accumulation prediction model, and then the trained cold accumulation prediction model is evaluated until the evaluation score is larger than or equal to the threshold value. Specifically, if the evaluation score is always smaller than the threshold, the above process is repeatedly performed until the evaluation score is equal to or greater than the threshold.
The adjustment method of the hyper-parameters is various, and the adjustment method can be manually adjusted according to work experience, and grid search, random search, bayesian search and the like can be selected.
With respect to the evaluation method, the R2 coefficient is commonly used, and the disadvantage of R2 is influenced by the size of the data set. The influence of data fluctuation and the influence of data variance can generate large errors on the evaluation result, therefore, in the embodiment of the present invention, the evaluating the trained cold accumulation prediction model includes:
and respectively correcting the actual cold storage value in the cold storage array and the predicted cold storage value of the cold storage prediction model to obtain a real correction value and a prediction correction value. The error influence caused by different conversion units of refrigerating capacity can be eliminated through correction. And meanwhile, the deficiency that MAPE cannot calculate because historical data with the refrigerating capacity of 0 exist in the real data is complemented. Calculating the ratio of the real correction value to the predicted correction value; summing the ratio sum corresponding to each cold accumulation data group in the test set; an evaluation score is determined from the ratio sum.
As a specific implementation form, the evaluation model specifically includes:
whereinF (x) is the evaluation score, n is the number of cold accumulation data sets in the test set, y i The actual value of the cold accumulation amount in the ith cold accumulation array,
the historical cold accumulation amount before the cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day in the ith cold accumulation array are used as cold accumulation amount prediction values input into the cold accumulation amount prediction model, and lambda is a correction coefficient of the test set.
As for the correction coefficient, it is possible to determine the correction coefficient according to the cold storage amount in the test set. Specifically, λ =10 is adopted [lg (x)] Calculating to obtain a sub-correction coefficient corresponding to each nonzero cold accumulation in the test set, and forming a sub-correction coefficient set; wherein x is the cold accumulation amount in the test set, and the cold accumulation amount is not 0; a mode is selected from the correction coefficient set, and the mode is used as a correction coefficient.
More specifically, this embodiment provides for using λ =10 [lg(x)] Calculating a sub-correction coefficient, wherein [ lg (x)]The expression is taken to be an integer for lg (x), if the result of lg (x) is 5.88, the value is 5, and the result of lg (x) is 3.42, the value is 3. As the refrigerating capacity data is constant in unit, the number of bits is stable. The coefficient is therefore easy to determine. For example, the historical cooling capacity value is around 80000, and λ is 10000. If the historical cooling capacity data is 50, the coefficient is 10.
In the embodiment of the invention, the historical cold accumulation amount comprises the cold accumulation amount of m ice storage air-conditioning working days which are before and adjacent to the cold accumulation day, wherein m is an integer, and experiments show that when m =7 is preferred, the prediction effect can be greatly improved, the calculation complexity is not high, and the prediction result can be rapidly obtained.
In the embodiment of the invention, in order to further improve the accuracy of the cold accumulation prediction model, the proportion of the cold accumulation arrays in the training set and the test set needs to be designed, and the ratio of the cold accumulation data arrays in the training set and the test set is preferably 4:1.
Hereinafter, an ice storage air conditioner in a hotel will be described in detail as an example.
Firstly, the refrigerating capacity of the previous 7 days is selected, the maximum temperature and the minimum temperature of the current day are used as input variables, the output variable is the refrigerating capacity of the current day, the historical data of the previous half year is selected as a basis in a time window, the daily refrigerating capacity is calculated, the historical maximum and minimum temperature of the current day is obtained through a weather station, and a data set is constructed.
According to the method, the method comprises the following steps of: and 2, taking 80% of data in the data set as a training set and 20% of data in the data set as a test set, and evaluating the model effect by using the test set data.
And then, training the model through a machine learning algorithm, selecting a random forest regression algorithm to train, obtaining an optimal parameter combination through grid search, and inputting parameters. And training by using the training set data to obtain a cold accumulation prediction model.
Then, using the test set data, inputting the refrigerating capacity 7 days before the cold accumulation day, the maximum and minimum temperature of the cold accumulation day, and outputting the predicted cold accumulation amount of the cold accumulation day. The following data results are obtained, as shown in table 1.
TABLE 1
And finally, evaluating the prediction result. In the evaluation stage, firstly, the R2 coefficient is used for evaluation, and on the basis of the full score of 1, an evaluation result is obtained: 0.911318. in the present embodiment, the evaluation result threshold is set to 0.95. Therefore, it is known that the R2 coefficient has a large estimation error with respect to the cold accumulation amount model.
When the evaluation model of the invention is used for evaluation, taking the data set 1 as an example, the calculation process of the correction coefficient λ is as follows: x =85.77, lg (85.77) =1.9333, [1.9333]＝1，λ＝10 1 =10. By adopting the method, the evaluation score (namely the evaluation result) of the evaluation model is continuously calculated, and the evaluation result is 0.97087928 on the basis that the full score is 1, obviously, the method provided by the inventionThe evaluation result of the evaluation model in (1) is closer to 1, and therefore, the accuracy of the evaluation model formed by adding the correction coefficient in the present embodiment is better than that of the R2 coefficient.
Therefore, the method for predicting the cold storage capacity of the ice storage air conditioner based on machine learning can minimize the actual operation cost and optimize the control, namely, has accurate load prediction. The invention greatly improves the prediction precision of the day-by-day cold load by a simple and effective method, and keeps the prediction precision above 98 percent, thereby keeping the cold accumulation amount of the ice cold accumulation air conditioner in a better range, reducing the energy consumption and improving the efficiency.
The invention also discloses a device for predicting the cold storage capacity of the ice storage air conditioner based on machine learning, which comprises the following components as shown in figure 2: the building module 210 is configured to build a cold storage capacity data set according to the historical data, and divide the cold storage capacity data set into a training set and a test set; the training set and the test set respectively comprise a plurality of groups of cold accumulation data groups, and each group of cold accumulation data group respectively comprises historical cold accumulation amount before a cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day and the cold accumulation amount of the cold accumulation day; the training module 220 is configured to input the training set as input data to the cold accumulation prediction model, and train the cold accumulation prediction model; the cold accumulation prediction model is selected as a random forest regression algorithm; the evaluation module 230 is configured to evaluate the trained cold accumulation prediction model by using the test set as input data to obtain an evaluation score; and the prediction module 240 is used for predicting the cold storage amount of the cold storage day by using the historical cold storage amount before the cold storage day, the highest temperature and the lowest temperature of the cold storage day as input data and adopting a trained cold storage amount prediction model in response to the evaluation score being greater than or equal to the threshold value.
It should be noted that, for the information interaction, execution process, and other contents between the modules of the apparatus, the specific functions and technical effects of the embodiments of the method are based on the same concept, and thus reference may be made to the section of the embodiments of the method specifically, and details are not described here.
It will be clear to those skilled in the art that, for convenience and simplicity of description, the foregoing division of the functional modules is merely illustrated, and in practical applications, the above function distribution may be performed by different functional modules according to needs, that is, the internal structure of the apparatus is divided into different functional modules to perform all or part of the above described functions. Each functional module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional modules are only used for distinguishing one functional module from another, and are not used for limiting the protection scope of the application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The invention also discloses a device for predicting the cold storage capacity of the ice storage air conditioner based on machine learning, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the method for predicting the cold storage capacity of the ice storage air conditioner based on machine learning is realized when the processor executes the computer program.
The device can be a desktop computer, a notebook, a palm computer, a cloud server and other computing equipment. The apparatus may include, but is not limited to, a processor, a memory. Those skilled in the art will appreciate that the apparatus may include more or fewer components, or some components in combination, or different components, and may also include, for example, input-output devices, network access devices, etc.
The Processor may be a Central Processing Unit (CPU), or other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The storage may in some embodiments be an internal storage unit of the device, such as a hard disk or a memory of the device. The memory may also be an external storage device of the apparatus in other embodiments, such as a plug-in hard disk, smart Media Card (SMC), secure Digital (SD) Card, flash memory Card (Flash Card), etc. provided on the apparatus. Further, the memory may also include both an internal storage unit and an external storage device of the apparatus. The memory is used for storing an operating system, application programs, a BootLoader (BootLoader), data, and other programs, such as program codes of the computer programs. The memory may also be used to temporarily store data that has been output or is to be output.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment. Those of ordinary skill in the art will appreciate that the various illustrative modules and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
Claims (9)
1. A method for predicting cold accumulation of an ice cold accumulation air conditioner based on machine learning is characterized by comprising the following steps:
according to historical data, a cold accumulation data set is constructed and divided into a training set and a test set; the training set and the test set respectively comprise a plurality of groups of cold accumulation data groups, and each group of cold accumulation data groups respectively comprise historical cold accumulation amount before a cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day and the cold accumulation amount of the cold accumulation day;
inputting the training set as input data into a cold accumulation prediction model, and training the cold accumulation prediction model; the cold accumulation prediction model is selected as a random forest regression algorithm;
evaluating the trained cold accumulation prediction model by taking the test set as input data to obtain an evaluation score;
evaluating the trained cold accumulation prediction model through an evaluation model; the evaluation model specifically comprises:
wherein f (x) is an evaluation score, n is the number of the cold accumulation data sets in the test set, and y i The real value of the cold accumulation amount in the ith cold accumulation array,
the historical cold accumulation amount before the cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day in the ith cold accumulation array are used as cold accumulation amount prediction values input into a cold accumulation amount prediction model, and lambda is a correction coefficient of a test set;
and in response to the evaluation score being larger than or equal to the threshold value, predicting the cold accumulation amount of the cold accumulation day by adopting the trained cold accumulation amount prediction model by taking the historical cold accumulation amount before the cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day as input data.
2. The method as claimed in claim 1, wherein in response to the evaluation score being less than the threshold, adjusting the parameters of the cold accumulation prediction model, and continuing to train the cold accumulation prediction model using the training set until the evaluation score is greater than or equal to the threshold.
3. A method as claimed in claim 1 or 2, wherein the evaluation of the trained cold accumulation prediction model comprises:
respectively correcting the actual cold storage value in the cold storage array and the predicted cold storage value of the cold storage prediction model to obtain a real correction value and a prediction correction value;
calculating the ratio of the real correction value to the predicted correction value;
summing the ratio sum corresponding to each cold accumulation data group in the test set;
and determining an evaluation score according to the ratio sum.
4. A method as claimed in claim 1, wherein the correction factor is determined according to the cold accumulation in the test set.
5. A method as claimed in claim 4, wherein determining the correction factor according to the cold accumulation in the test set comprises:
with λ =10 [lg(x)] Calculating to obtain a sub-correction coefficient corresponding to each nonzero cold accumulation in the test set, and forming a sub-correction coefficient set; wherein x is the cold accumulation amount in the test set, and the cold accumulation amount is not 0;
and taking a mode from the correction coefficient set, and taking the mode as a correction coefficient.
6. The method for predicting the cold accumulation amount of the ice-storage air conditioner based on the machine learning as claimed in claim 1, 4 or 5, wherein the historical cold accumulation amount comprises the cold accumulation amount of m ice-storage air-conditioning working days which are before the cold accumulation day and adjacent to the cold accumulation day, wherein m =7.
7. The method as claimed in claim 6, wherein the ratio of the cold accumulation data set in the training set and the test set is 4:1.
8. The utility model provides an ice cold-storage air conditioner cold-storage volume prediction unit based on machine learning which characterized in that includes:
the building module is used for building a cold storage capacity data set according to historical data and dividing the cold storage capacity data set into a training set and a test set; the training set and the test set respectively comprise a plurality of groups of cold accumulation data groups, and each group of cold accumulation data groups respectively comprise historical cold accumulation amount before a cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day and the cold accumulation amount of the cold accumulation day;
the training module is used for inputting the training set as input data into a cold accumulation prediction model and training the cold accumulation prediction model; the cold accumulation prediction model is selected as a random forest regression algorithm;
and the evaluation module is used for evaluating the trained cold accumulation prediction model by taking the test set as input data to obtain an evaluation score, wherein the evaluation model specifically comprises the following steps:
wherein f (x) is an evaluation score, n is the number of the cold accumulation data sets in the test set, and y i The real value of the cold accumulation amount in the ith cold accumulation array,
the historical cold accumulation amount before the cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day in the ith cold accumulation array are used as cold accumulation amount prediction values input into a cold accumulation amount prediction model, and lambda is a correction coefficient of a test set;
and the prediction module is used for predicting the cold accumulation amount of the cold accumulation day by adopting the trained cold accumulation amount prediction model by taking the historical cold accumulation amount before the cold accumulation day, the highest temperature and the lowest temperature of the cold accumulation day as input data in response to the condition that the evaluation score is more than or equal to the threshold value.
9. A device for predicting cold storage capacity of ice storage air conditioner based on machine learning, comprising a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor implements a method for predicting cold storage capacity of ice storage air conditioner based on machine learning according to any one of claims 1 to 7 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111027278.4A CN113701275B (en) | 2021-09-02 | 2021-09-02 | Method and device for predicting cold accumulation amount of ice cold accumulation air conditioner based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111027278.4A CN113701275B (en) | 2021-09-02 | 2021-09-02 | Method and device for predicting cold accumulation amount of ice cold accumulation air conditioner based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113701275A CN113701275A (en) | 2021-11-26 |
| CN113701275B true CN113701275B (en) | 2022-12-27 |
Family
ID=78657509
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111027278.4A Active CN113701275B (en) | 2021-09-02 | 2021-09-02 | Method and device for predicting cold accumulation amount of ice cold accumulation air conditioner based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113701275B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108844190A (en) * | 2018-06-20 | 2018-11-20 | 中国科学院自动化研究所 | Air-conditioning self study optimal control system |
| CN111754049A (en) * | 2020-07-14 | 2020-10-09 | 南京优助智能科技有限公司 | Dynamic optimization and display method for cold discharge time table of ice cold storage system |
| CN112815473A (en) * | 2020-12-31 | 2021-05-18 | 珠海横琴能源发展有限公司 | Optimal control device and control method for cold accumulation air conditioning system |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10600005B2 (en) * | 2018-06-01 | 2020-03-24 | Sas Institute Inc. | System for automatic, simultaneous feature selection and hyperparameter tuning for a machine learning model |
| CN110225075A (en) * | 2019-03-25 | 2019-09-10 | 北京快电科技有限公司 | A kind of building energy internet wisdom operation cloud operating system |
| CN112696750A (en) * | 2021-01-22 | 2021-04-23 | 德莱创节能科技（苏州）有限公司 | Cold accumulation control method and device for air conditioning device with cold accumulation function |
-
2021
- 2021-09-02 CN CN202111027278.4A patent/CN113701275B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108844190A (en) * | 2018-06-20 | 2018-11-20 | 中国科学院自动化研究所 | Air-conditioning self study optimal control system |
| CN111754049A (en) * | 2020-07-14 | 2020-10-09 | 南京优助智能科技有限公司 | Dynamic optimization and display method for cold discharge time table of ice cold storage system |
| CN112815473A (en) * | 2020-12-31 | 2021-05-18 | 珠海横琴能源发展有限公司 | Optimal control device and control method for cold accumulation air conditioning system |
Non-Patent Citations (2)
| Title |
|---|
| 冰蓄冷空调负荷预测的预估校正模型;李壮举等;《计算机工程与设计》;20200229;574-579 * |
| 基于运行模式划分的复杂集中式空调控制系统逐时能耗预测方法;闫军威;《建筑科学》;20200229;177-190 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113701275A (en) | 2021-11-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11281173B2 (en) | Systems and methods for maintaining occupant comfort for various environmental conditions | |
| US12130598B2 (en) | Building control system with features for operating under intermittent connectivity to a cloud computation system | |
| CN102855343B (en) | Load amount prediction device, load amount prediction method, and load amount prediction program | |
| Park et al. | Predictive model for PV power generation using RNN (LSTM) | |
| Olivieri et al. | Evaluation of commercial building demand response potential using optimal short-term curtailment of heating, ventilation, and air-conditioning loads | |
| CN111680841B (en) | Short-term load prediction method, system and terminal equipment based on principal component analysis | |
| Zhang et al. | Development and evaluation of cooling load prediction models for a factory workshop | |
| US20200125045A1 (en) | Hvac controller with predictive cost optimization | |
| CN112950098B (en) | Comprehensive energy system-based energy planning method and device and terminal equipment | |
| CN112598195A (en) | Building type comprehensive energy system operation optimization method and device and terminal equipment | |
| CN111695742A (en) | Day-ahead scheduling interval optimization method, system and equipment for comprehensive energy system | |
| Huang et al. | Sizing heating, ventilating, and air-conditioning systems under uncertainty in both load-demand and capacity-supply side from a life-cycle aspect | |
| CN111738529A (en) | Comprehensive energy system demand response method, system and equipment based on reinforcement learning | |
| CN115829235A (en) | Power utilization scheduling method based on big data analysis | |
| Yang et al. | Developing data-driven models to predict BEMS energy consumption for demand response systems | |
| Yang et al. | Bi-level decentralized control of electric heating loads considering wind power accommodation in real-time electricity market | |
| CN111680712A (en) | Transformer oil temperature prediction method, device and system based on similar moments in the day | |
| Lee et al. | On-policy learning-based deep reinforcement learning assessment for building control efficiency and stability | |
| Zhang et al. | Stochastic unit commitment with air conditioning loads participating in reserve service | |
| Sarkar et al. | Concurrent carbon footprint reduction (c2fr) reinforcement learning approach for sustainable data center digital twin | |
| CN113701275B (en) | Method and device for predicting cold accumulation amount of ice cold accumulation air conditioner based on machine learning | |
| Naderi et al. | Clustering based analysis of residential duck curve mitigation through solar pre-cooling: A case study of Australian housing stock | |
| CN115759422A (en) | Heating heat load prediction method, system, device, and medium | |
| CN113067340B (en) | Dynamic state estimation method and system for constant temperature control load system | |
| CN115577913A (en) | Computing method, terminal and storage medium for active load schedulable potential |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |