CN115511100A - Air conditioner load regression prediction method based on environmental temperature related data learning - Google Patents
Air conditioner load regression prediction method based on environmental temperature related data learning Download PDFInfo
- Publication number
- CN115511100A CN115511100A CN202210966555.6A CN202210966555A CN115511100A CN 115511100 A CN115511100 A CN 115511100A CN 202210966555 A CN202210966555 A CN 202210966555A CN 115511100 A CN115511100 A CN 115511100A
- Authority
- CN
- China
- Prior art keywords
- historical
- conditioning load
- data
- temperature
- ambient temperature
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
- F24F11/46—Improving electric energy efficiency or saving
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention provides an air conditioner load regression prediction method based on environmental temperature related data learning, which comprises the following steps: respectively processing the historical ambient temperature and the historical air-conditioning load of the target building at each sampling time into data groups, sequencing the data groups according to the ambient temperature, and carrying out standardization processing on the data groups to form a data set; establishing a regression learning model of the ambient temperature and the air conditioner load based on a vector mechanism; performing precision analysis on the regression learning model by combining the sequencing condition of the data group and the training set and the testing set in the data set to obtain an air conditioner load prediction model meeting the precision requirement; and acquiring the real-time environment temperature of the target building, inputting the real-time environment temperature into the air-conditioning load prediction model, and predicting the air-conditioning load of the target building in real time. The invention realizes regression prediction analysis of the average air-conditioning load through the ambient temperature, and can obtain the optimal regression prediction curve of the outdoor ambient temperature and the air-conditioning load by using less sample data.
Description
Technical Field
The invention belongs to the field of air conditioner load energy consumption prediction, and particularly relates to an air conditioner load regression prediction method based on environmental temperature related data learning.
Background
The energy consumption of an air conditioning system of an office living building is usually more than half of the total energy consumption of the whole building, the operating energy consumption of the air conditioning system has strong correlation with the outdoor environment temperature, a load relation model of the air conditioning system at different environment temperatures is established by analyzing the operating load rule of the air conditioning system, and an air conditioning load prediction method based on the environment temperature is researched, so that the air conditioning load prediction method can be used as a decision basis for air conditioning energy-saving operation scheduling, the system operating efficiency is improved, the air conditioning energy consumption load is optimized, and the air conditioning load prediction method has important significance for reducing the total energy consumption of the whole building.
Continuous time prediction is one of the main methods for predicting the current air conditioner load, and mainly comprises an artificial neural network, a random forest method, multiple linear regression and the like. However, a large amount of characteristic data is needed for establishing a deep neural network model, and the data characteristic magnitude required by deep learning is usually difficult to achieve in actual engineering; the random forest algorithm is also widely applied to building load prediction, but the credibility of the attribute weight is reduced due to the fact that larger random forests are generated by dividing data; the multiple linear regression is a classical regression prediction method in a linear statistical analysis method, but the regression prediction precision of the method on a time series with large sample number and random fluctuation is not high.
Disclosure of Invention
In order to solve the defects and shortcomings in the prior art, the invention provides an air conditioner load regression prediction method based on environmental temperature related data learning. The Support Vector Machine (SVM) can realize better prediction performance and model generalization capability through limited small sample data, so that the invention can realize prediction of air conditioning load under different outdoor relevant ambient temperatures under the small sample data by fitting the relation between the outdoor ambient temperature and the air conditioning load by SVM model regression based on the air conditioning load and ambient temperature data, and can obtain the optimal regression prediction curve of the outdoor ambient temperature and the air conditioning load through less data samples.
The air conditioner load regression prediction method provided by the invention comprises the following steps:
s1: respectively processing the historical ambient temperature and the historical air-conditioning load of the target building at each sampling time into data groups, sequencing the data groups according to the ambient temperature, and carrying out standardization processing on the data groups to form a data set;
s2: establishing a regression learning model of the ambient temperature and the air conditioner load based on a vector mechanism;
s3: performing precision analysis on the regression learning model by combining the sequencing condition of the data group and the training set and the testing set in the data set to obtain an air conditioner load prediction model meeting the precision requirement;
s4: and acquiring the real-time environment temperature of the target building, inputting the real-time environment temperature into the air conditioner load prediction model, and predicting the air conditioner load of the target building in real time.
Optionally, the processing the historical ambient temperature and the historical air conditioning load of the target building at each sampling time into a data set respectively includes:
s11: and performing data correlation calculation on the historical ambient temperature and the historical air conditioner load to obtain a correlation coefficient r as follows:
in the formula, x i Is the historical ambient temperature at the sampling instant i, y i To sample the historical air conditioning load at time i,
in order to be the historical ambient average temperature,
the average load of the historical air conditioner is shown, and n is the total number of sampling moments;
s12: if the correlation coefficient meets the preset condition, taking the historical ambient temperature and the historical air-conditioning load as a data set { (x) i ，y i ) If not, reselecting the sampling moment to obtain the historical ambient temperature and the historical air-conditioning load of the target building, and returning to execute S11.
Optionally, the normalizing the data group to form a data set includes:
normalizing the data set based on a calculation formula { (x) i ，y i ) I =1,2,.., n } is processed as a normalized value, which is calculated as:
wherein mu x 、μ y Mean values, σ, of historical ambient temperature and historical air conditioning load, respectively x 、σ y Is the variance of the historical ambient temperature and the historical air conditioning load,
the normalized historical ambient temperature and historical air conditioning load.
Optionally, the S2 includes:
s21: standardizing the processed data set
Dividing the training set into a training set and a test set according to a preset proportion;
s22: using non-linear mapping on training set
Sample space R d Mapping to a high-dimensional feature space R m (m ≧ d), constructing a hyperplane function using the weight w and the offset b in the feature space
S23: and constructing a regression learning model by combining an optimal quadratic convex programming problem according to the hyperplane function.
Optionally, the S23 includes:
placing a band of precision intervals between the hyperplane function f (x) and the data set, the band of precision intervals being expressed as:
|y-f(x)|≤ε；
wherein epsilon represents the width of the set precision isolation zone, and y represents the historical air conditioning load in the data set;
introducing a relaxation variable, and generating an optimal quadratic convex programming problem based on an expression of the precision isolation zone as follows:
s.t.|y-f(x)|≤ε+ξ i ，i＝1，2，…，n；
wherein C is a penalty factor, ξ i Is a relaxation variable;
converting the optimal quadratic convex programming problem into a dual problem through a Lagrange optimization function to obtain a Lagrange multiplier;
based on the Lagrange multiplier and the hyperplane function, a regression learning model is expressed as:
wherein alpha is i And
is Lagrange multiplier, K (x, x) i ) Is a kernel function introduced according to a constraint s.t.
Optionally, the S3 includes:
s31: training the regression learning model according to a training set;
s32: determining a temperature interval according to the sorting condition of the data group, and calculating the average value of the historical air-conditioning load corresponding to the temperature interval in the test set;
s33: inputting the historical environment temperature corresponding to the temperature interval in the test set into a regression learning model, and calculating an accuracy decision coefficient according to the output result of the regression learning model and the average value calculated in S32;
s34: and when the precision decision coefficient is smaller than or equal to a preset value, returning to S31 for retraining, when the decision coefficient is larger than the preset value, judging that the precision requirement is met, stopping training, and taking the regression learning model obtained by current training as an air conditioner load prediction model.
Optionally, the determining the temperature interval according to the sorting condition of the data group includes:
and respectively calculating the temperature difference values of the historical environmental temperatures in the two adjacent data sets, changing the size delta T of the temperature interval according to the average value of all the temperature difference values, and setting the value delta T to be larger when the average value of all the temperature difference values is larger.
Optionally, the calculating an average value of historical air conditioning loads corresponding to the temperature intervals in the test set includes: the calculation formula of the average value is as follows:
wherein, y T，T+ΔT Corresponding temperature intervals [ T, T + Delta T ] are concentrated for testing]Average value of historical air conditioning load of (y) i Is corresponding to the temperature interval [ T, T + Delta T]N is the corresponding temperature interval [ T, T + Delta T ]]The data quantity of the historical air-conditioning load collected in the air conditioner.
Optionally, the expression of the precision determination coefficient is:
wherein R is 2 Indicating the precision determining coefficient, y i And f (x) represents an air conditioning load prediction result output by the air conditioning load prediction model.
The technical scheme provided by the invention has the following beneficial effects:
(1) According to the method, a data sample set is constructed for summer environment temperature and air conditioner load, regression prediction analysis of the environment temperature on average air conditioner load is realized through construction and training of an SVM regression learning model, and an optimal regression prediction curve of outdoor environment temperature and air conditioner load can be obtained through less sample data.
(2) The method is characterized in that the environmental temperatures with different degrees are divided, the average air-conditioning load of each interval is calculated, the optimal regression curve, the environmental temperature and the average air-conditioning load data can be evaluated, and the regression model with higher accuracy of the environmental temperature and the air-conditioning load is obtained.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on the drawings without creative efforts.
Fig. 1 is a schematic flowchart of an air conditioning load regression prediction method based on ambient temperature related data learning according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein.
It should be understood that, in various embodiments of the present invention, the sequence numbers of the processes do not mean the execution sequence, and the execution sequence of the processes should be determined by the functions and the internal logic of the processes, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
It should be understood that in the present application, "comprising" and "having" and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be understood that, in the present invention, "a plurality" means two or more. "and/or" is merely an association relationship describing an associated object, meaning that there may be three relationships, for example, and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "comprises A, B and C" and "comprises A, B, C" means that all three of A, B, C comprise, "comprises A, B or C" means that one of three of A, B, C is comprised, "comprises A, B and/or C" means that any 1 or any 2 or 3 of the three of A, B, C is comprised.
It should be understood that in the present invention, "B corresponding to a", "a corresponds to B", or "B corresponds to a" means that B is associated with a, and B can be determined from a. Determining B from a does not mean determining B from a alone, but may be determined from a and/or other information. And the matching of A and B means that the similarity of A and B is greater than or equal to a preset threshold value.
As used herein, "if" can be interpreted as "at … …" or "at … …" or "in response to a determination" or "in response to a detection", depending on context.
The technical solution of the present invention will be described in detail below with specific examples. These several specific embodiments may be combined with each other below, and details of the same or similar concepts or processes may not be repeated in some embodiments.
The embodiment is as follows:
as shown in fig. 1, the present embodiment proposes an air conditioning load regression prediction method based on ambient temperature related data learning, including:
s1: respectively processing the historical ambient temperature and the historical air-conditioning load of the target building at each sampling time into data groups, sequencing the data groups according to the ambient temperature, and carrying out standardization processing on the data groups to form a data set;
s2: establishing a regression learning model of the ambient temperature and the air conditioner load based on a vector mechanism;
s3: performing precision analysis on the regression learning model by combining the sequencing condition of the data group and the training set and the testing set in the data set to obtain an air conditioner load prediction model meeting the precision requirement;
s4: and acquiring the real-time environment temperature of the target building, inputting the real-time environment temperature into the air conditioner load prediction model, and predicting the air conditioner load of the target building in real time.
The method comprises the steps of firstly standardizing small sample data related to outdoor environment temperature and air conditioner load, then constructing a regression prediction model by using an SVM (support vector machine), obtaining an optimal regression curve of the environment temperature and average air conditioner load, and evaluating the regression curve by using test data. The method does not depend on a large amount of data, and the optimal regression prediction curve of the outdoor environment temperature and the air conditioning load can be obtained through fewer data samples.
The processing of the historical ambient temperature and the historical air conditioning load of the target building at each sampling time into data sets respectively comprises:
s11: and performing data correlation calculation on the historical ambient temperature and the historical air conditioner load to obtain a correlation coefficient r as follows:
in the formula, x i Is the historical ambient temperature at the sampling instant i, y i To sample the historical air conditioning load at time i,
in order to be the average temperature of the historical environment,
the average load of the historical air conditioners is obtained, and n is the total number of sampling moments;
s12: if the correlation coefficient meets the preset condition, taking the historical ambient temperature and the historical air-conditioning load as a data set { (x) i ，y i ) And I =1,2 and … and n }, otherwise, reselecting the sampling moment to obtain the historical ambient temperature and the historical air-conditioning load of the target building, and returning to execute S11.
In general, the ambient temperature and the air conditioning load should have a correlation, and the embodiment analyzes the ambient temperature and the air conditioning load by using pearson correlation, and the larger the correlation coefficient is, the more the ambient temperature and the air conditioning load are correlated, the more the correlation coefficient is consistent with the practical common sense, and the more the sampled data set can accurately reflect the practical regression condition. In this embodiment, the preset condition is that the correlation coefficient is greater than a certain value.
Specifically, in this embodiment, the historical ambient temperature x of 1561 hours (group) is collected for a certain building at hourly time intervals for a total of 2 months i And historical air conditioning load y i Data as input data set for learning { (x) i ，y i ) I =1,2, …, n }. The 1248 groups of data were randomly extracted as training set, and the remaining 313 groups of data were extracted as test set.
And then, sequencing the training set and the testing machine according to the temperature respectively to obtain a training set and a data set of the temperature sequence respectively. The purpose of ordering the training set and the test set in this embodiment is to facilitate subsequent determination of an indicator for assessing model accuracy from the difference in sampled historical ambient temperatures.
After the data set is formed, in order to unify the data dimension and facilitate the training of the subsequent regression learning model, the normalizing the data set to form the data set includes:
normalizing the data set based on a calculation formula { (x) i ，y i ) I =1,2,.., n } is processed as a normalized value, which is calculated as:
wherein mu x 、μ y Mean values, σ, of historical ambient temperature and historical air conditioning load, respectively x 、σ y Is the variance of the historical ambient temperature and the historical air conditioning load,
the normalized historical ambient temperature and historical air conditioning load.
In this embodiment, the S2 includes:
s21: standardizing the processed data set
Dividing the training set into a training set and a test set according to a preset proportion;
s22: using non-linear mapping on training set
Sample space R d Mapping to a high-dimensional feature space R m (m ≧ d), constructing a hyperplane function using the weight w and the offset b in the feature space
S23: and constructing a regression learning model by combining an optimal quadratic convex programming problem according to the hyperplane function.
Wherein the S23 includes:
placing a band of precision spacing between the hyperplane function f (x) and the data set, the band of precision spacing being expressed as:
|y-f(x)|≤ε；
where ε represents the width of the precision isolation zone set, and y represents the historical air conditioning load in the data set.
A spacing zone exists between the sample point and the hyperplane function, when the sample point falls inside the spacing zone, no deviation loss can be considered, and when the sample point falls outside the spacing zone, deviation can be considered.
Introducing a relaxation variable xi i The constraint cost is less when the sample point is close to the interval zone; when the sample point is far away from the spacing zone, the cost of satisfying the constraint is high, and therefore the optimal quadratic convex programming problem generated based on the expression of the precision isolation zone is as follows:
s.t.|y-f(x)|≤ε+ξ i ，i＝1，2，…，n；
wherein C is a penalty factor, ξ i Is the relaxation variable.
Since the lagrangian optimization function and the constraint condition function are continuous and differentiable convex functions and satisfy the KKT condition, the solution of the original problem is equivalent to the solution of the dual problem, and the original problem can be converted into the dual problem, specifically including: converting the optimal quadratic convex programming problem into a dual problem through a Lagrange optimization function to obtain a Lagrange multiplier;
the expression for establishing the lagrangian optimization function is as follows:
wherein alpha is i And
is a lagrange multiplier.
Defining an optimization objective:
if the parameter w is not satisfied (y) i -f(x i )-ε-ξ i ) Less than or equal to 0 and
then theta p (w) taken to infinity; if the parameter w satisfies the constraint condition, then
The optimization problem from finding the minima can be translated into:
the original problem can be further converted into a dual problem, and the expression is as follows:
based on the Lagrange multiplier and the hyperplane function, a regression learning model is expressed as:
wherein alpha is i And
is Lagrange multiplier, K (x, x) i ) Is a kernel function introduced according to a constraint s.t.
In the training phase, the S3 includes:
s31: and training the regression learning model according to the training set. In this embodiment, 80% of the sample data set and 20% of the sample data set are divided into a training set and a training set.
S32: determining a temperature interval according to the sorting condition of the data group, and calculating the average value of the historical air-conditioning loads corresponding to the temperature interval in the test set, wherein the method specifically comprises the following steps:
and respectively calculating the temperature difference values of the historical environmental temperatures in the two adjacent data sets, changing the size delta T of the temperature interval according to the average value of all the temperature difference values, and setting the value delta T to be larger when the average value of all the temperature difference values is larger.
According to the temperature interval of the historical ambient temperature in the sampled data set, the temperature interval of the historical air conditioner load average value is adaptively adjusted, and precision evaluation is facilitated.
The calculation formula of the average value is as follows:
wherein, y T，T+ΔT To testCorresponding temperature interval [ T, T + delta T ] is concentrated]Average value of historical air conditioning load of (1), y i Is corresponding to the temperature interval [ T, T + Delta T]N is the corresponding temperature interval [ T, T + Delta T ]]The data quantity of the historical air conditioner load collected in the air conditioner.
S33: inputting the historical environment temperature corresponding to the temperature interval in the test set into a regression learning model, and calculating an accuracy decision coefficient according to the output result of the regression learning model and the average value calculated in S32;
s34: and when the precision decision coefficient is smaller than or equal to a preset value, returning to S31 for retraining, when the decision coefficient is larger than the preset value, judging that the precision requirement is met, stopping training, and taking the regression learning model obtained by current training as an air conditioner load prediction model.
In order to evaluate the accuracy of the regression learning model, the present embodiment calculates the accuracy decision coefficient by the following expression:
wherein R is 2 Indicating the precision determining coefficient, y i And f (x) represents an air conditioning load prediction result output by the air conditioning load prediction model.
In this embodiment, the decision coefficient of the air-conditioning load observation data and the regression data of the ambient temperature obtained by calculation reaches 0.94, which indicates that the regression prediction model of the air-conditioning load and the ambient temperature has higher accuracy.
The sequence numbers in the above embodiments are merely for description, and do not represent the sequence of the assembly or the use of the components.
The above description is only exemplary of the present invention and should not be taken as limiting the invention, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. An air conditioning load regression prediction method based on environmental temperature related data learning is characterized by comprising the following steps:
s1: respectively processing the historical ambient temperature and the historical air-conditioning load of the target building at each sampling time into data groups, sequencing the data groups according to the ambient temperature, and carrying out standardization processing on the data groups to form a data set;
s2: establishing a regression learning model of the ambient temperature and the air conditioner load based on a vector mechanism;
s3: performing precision analysis on the regression learning model by combining the sequencing condition of the data group and the training set and the testing set in the data set to obtain an air conditioner load prediction model meeting the precision requirement;
s4: and acquiring the real-time environment temperature of the target building, inputting the real-time environment temperature into the air-conditioning load prediction model, and predicting the air-conditioning load of the target building in real time.
2. The regression prediction method for air conditioning load based on ambient temperature related data learning of claim 1, wherein the processing historical ambient temperature and historical air conditioning load of the target building at each sampling time into data sets respectively comprises:
s11: and performing data correlation calculation on the historical ambient temperature and the historical air conditioner load to obtain a correlation coefficient r as follows:
in the formula, x i Is the historical ambient temperature at the sampling instant i, y i To sample the historical air conditioning load at time i,
in order to be the historical ambient average temperature,
the average load of the historical air conditioner is shown, and n is the total number of sampling moments;
s12: if the correlation coefficient meets the preset condition, taking the historical environmental temperature and the historical air-conditioning load as a data set { (x) i ，y i ) If not, reselecting the sampling moment to obtain the historical ambient temperature and the historical air-conditioning load of the target building, and returning to execute S11.
3. The regression prediction method for air conditioning load based on learning of environmental temperature related data as claimed in claim 2, wherein the normalizing the data set to form a data set comprises:
normalizing the data set based on a calculation formula { (x) i ，y i ) I =1,2,.., n } is processed as a normalized value, which is calculated as:
wherein mu x 、μ y Mean values, σ, of historical ambient temperature and historical air conditioning load, respectively x 、σ y Is the variance of the historical ambient temperature and the historical air conditioning load,
the normalized historical ambient temperature and historical air conditioning load.
4. The regression prediction method for air conditioning load based on learning of relevant data of ambient temperature as claimed in claim 1, wherein the S2 comprises:
s21: standardizing the processed data set
Divided into predetermined proportionsTraining and testing sets;
s22: using non-linear mapping on training set
Sample space R d Mapping to a high-dimensional feature space R m (m ≧ d), constructing a hyperplane function in the feature space using the weight w and the offset v
S23: and constructing a regression learning model by combining an optimal quadratic convex programming problem according to the hyperplane function.
5. The regression prediction method for air conditioning load based on learning of relevant data of ambient temperature as claimed in claim 4, wherein the S23 comprises:
placing a band of precision spacing between the hyperplane function f (x) and the data set, the band of precision spacing being expressed as:
|y-f(x)|≤ε；
wherein epsilon represents the width of the set precision isolation zone, and y represents the historical air conditioning load in the data set;
introducing a relaxation variable, and generating an optimal quadratic convex programming problem based on an expression of a precision isolation zone as follows:
s.t.|y-f(x)|≤ε+ξ i ，i＝1，2，…，n；
wherein C is a penalty factor, ξ i Is a relaxation variable;
converting the optimal quadratic convex programming problem into a dual problem through a Lagrange optimization function to obtain a Lagrange multiplier;
based on the Lagrange multiplier and the hyperplane function, a regression learning model is expressed as:
wherein alpha is i And
is Lagrange multiplier, K (x, x) i ) Is a kernel function introduced according to a constraint s.t.
6. The regression prediction method for air conditioning load based on the learning of the environmental temperature related data as claimed in claim 1, wherein the S3 comprises:
s31: training the regression learning model according to a training set;
s32: determining a temperature interval according to the sorting condition of the data group, and calculating the average value of the historical air-conditioning loads corresponding to the temperature interval in the test set;
s33: inputting the historical environment temperature corresponding to the temperature interval in the test set into a regression learning model, and calculating an accuracy decision coefficient according to the output result of the regression learning model and the average value calculated in S32;
s34: and when the precision decision coefficient is smaller than or equal to a preset value, returning to S31 for retraining, when the decision coefficient is larger than the preset value, judging that the precision requirement is met, stopping training, and taking the regression learning model obtained by current training as an air conditioner load prediction model.
7. The regression prediction method for air conditioning load based on learning of relevant data of ambient temperature as claimed in claim 6, wherein the determining the temperature interval according to the sorting condition of the data group comprises:
and respectively calculating the temperature difference values of the historical environmental temperatures in the two adjacent data sets, changing the size delta T of the temperature interval according to the average value of all the temperature difference values, and setting the value delta T to be larger when the average value of all the temperature difference values is larger.
8. The regression prediction method for air conditioning load based on ambient temperature related data learning of claim 7, wherein the calculating the average value of the historical air conditioning loads corresponding to the temperature intervals in the test set comprises:
the calculation formula of the average value is as follows:
wherein, y T，T+ΔT Corresponding temperature intervals [ T, T + Delta T ] are concentrated for testing]Average value of historical air conditioning load of (1), y i Is corresponding to the temperature interval [ T, T + Delta T]N is the corresponding temperature interval [ T, T + Delta T ]]The data quantity of the historical air-conditioning load collected in the air conditioner.
9. The regression prediction method for air conditioning load based on learning of relevant data of ambient temperature as claimed in claim 6, wherein the expression of the accuracy decision coefficient is:
wherein R is 2 Indicating the precision determining coefficient, y i And f (x) represents an air conditioning load prediction result output by the air conditioning load prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210966555.6A CN115511100A (en) | 2022-08-12 | 2022-08-12 | Air conditioner load regression prediction method based on environmental temperature related data learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210966555.6A CN115511100A (en) | 2022-08-12 | 2022-08-12 | Air conditioner load regression prediction method based on environmental temperature related data learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115511100A true CN115511100A (en) | 2022-12-23 |
Family
ID=84501031
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210966555.6A Pending CN115511100A (en) | 2022-08-12 | 2022-08-12 | Air conditioner load regression prediction method based on environmental temperature related data learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115511100A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117034786A (en) * | 2023-10-09 | 2023-11-10 | 山东芯赛思电子科技有限公司 | IGBT junction temperature prediction method |
| CN117272845A (en) * | 2023-11-22 | 2023-12-22 | 广东蘑菇物联科技有限公司 | Method, device and equipment for evaluating energy consumption level of air compression station |
-
2022
- 2022-08-12 CN CN202210966555.6A patent/CN115511100A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117034786A (en) * | 2023-10-09 | 2023-11-10 | 山东芯赛思电子科技有限公司 | IGBT junction temperature prediction method |
| CN117034786B (en) * | 2023-10-09 | 2024-01-05 | 山东芯赛思电子科技有限公司 | IGBT junction temperature prediction method |
| CN117272845A (en) * | 2023-11-22 | 2023-12-22 | 广东蘑菇物联科技有限公司 | Method, device and equipment for evaluating energy consumption level of air compression station |
| CN117272845B (en) * | 2023-11-22 | 2024-03-08 | 广东蘑菇物联科技有限公司 | Method, device and equipment for evaluating energy consumption level of air compression station |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107392368B (en) | Meteorological forecast-based office building dynamic heat load combined prediction method | |
| Zhao et al. | A review on the prediction of building energy consumption | |
| CN115511100A (en) | Air conditioner load regression prediction method based on environmental temperature related data learning | |
| CN111199016A (en) | DTW-based improved K-means daily load curve clustering method | |
| Wang et al. | GLCM based extraction of flame image texture features and KPCA-GLVQ recognition method for rotary kiln combustion working conditions | |
| CN114119273B (en) | Non-invasive load decomposition method and system for park comprehensive energy system | |
| CN108805743A (en) | A kind of power grid enterprises' sale of electricity company operation Benefit Evaluation Method | |
| Gao et al. | Using data mining in optimisation of building energy consumption and thermal comfort management | |
| CN113269365B (en) | Short-term air conditioner load prediction method and system based on sparrow optimization algorithm | |
| CN111028100A (en) | Refined short-term load prediction method, device and medium considering meteorological factors | |
| CN104899425A (en) | Variable selection and forecast method of silicon content in molten iron of blast furnace | |
| CN115545333A (en) | Method for predicting load curve of multi-load daily-type power distribution network | |
| Baesmat et al. | A new hybrid method for electrical load forecasting based on deviation correction and MRMRMS | |
| CN111652422A (en) | Heat supply system load prediction method, device and system based on building classification | |
| CN116930042A (en) | Building waterproof material performance detection equipment and method | |
| Wang et al. | Data-driven assessment of room air conditioner efficiency for saving energy | |
| CN108830405B (en) | Real-time power load prediction system and method based on multi-index dynamic matching | |
| Wang et al. | Cloud computing and extreme learning machine for a distributed energy consumption forecasting in equipment-manufacturing enterprises | |
| CN114626655A (en) | Multi-standard comprehensive evaluation method for regional comprehensive energy system | |
| Mo et al. | A model of oil price forecasting based on autoregressive and moving average | |
| Azadeh et al. | An integrated fuzzy regression–analysis of variance algorithm for improvement of electricity consumption estimation in uncertain environments | |
| CN107992454B (en) | Air quality grade prediction method based on online sequential regression | |
| CN114234392B (en) | Air conditioner load fine prediction method based on improved PSO-LSTM | |
| Kaligambe et al. | Indoor Room Temperature and Relative Humidity Estimation in a Commercial Building Using the XGBoost Machine Learning Algorithm | |
| CN115600498A (en) | Wind speed forecast correction method based on artificial neural network |
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 |