CN115169707A - Equipment energy consumption prediction method and device based on multiple linear regression - Google Patents
Equipment energy consumption prediction method and device based on multiple linear regression Download PDFInfo
- Publication number
- CN115169707A CN115169707A CN202210827474.8A CN202210827474A CN115169707A CN 115169707 A CN115169707 A CN 115169707A CN 202210827474 A CN202210827474 A CN 202210827474A CN 115169707 A CN115169707 A CN 115169707A
- Authority
- CN
- China
- Prior art keywords
- energy consumption
- linear regression
- value
- principal component
- multiple linear
- 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
- 238000005265 energy consumption Methods 0.000 title claims abstract description 85
- 238000012417 linear regression Methods 0.000 title claims abstract description 65
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000004364 calculation method Methods 0.000 claims abstract description 10
- 238000011156 evaluation Methods 0.000 claims description 22
- 238000012360 testing method Methods 0.000 claims description 14
- 239000013598 vector Substances 0.000 claims description 13
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000000513 principal component analysis Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 6
- 239000002131 composite material Substances 0.000 claims description 3
- 238000012847 principal component analysis method Methods 0.000 abstract description 7
- 230000009466 transformation Effects 0.000 abstract description 3
- 238000013178 mathematical model Methods 0.000 description 8
- 230000008859 change Effects 0.000 description 3
- 230000001186 cumulative effect Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000004378 air conditioning Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 229910052731 fluorine Inorganic materials 0.000 description 1
- 125000001153 fluoro group Chemical group F* 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses an equipment energy consumption prediction method and device based on multiple linear regression, wherein influence factors influencing energy consumption of electric equipment are used as independent variables through a principal component analysis method, all related variables are linearly transformed through continuous transformation of coordinates and are converted into another group of irrelevant variables, and the former variables with higher contribution rate are selected as principal components through calculation, so that the purposes of reflecting the energy consumption of the electric equipment by using fewer independent variables and avoiding the problem of multiple collinearity among the independent variables can be achieved. And establishing a multiple linear regression equation according to the principal component variables obtained by the principal component analysis method, and predicting the energy consumption of each electric device by using the multiple linear regression method. According to the method and the device, the energy consumption of the electric equipment is predicted, and the predicted energy consumption data is analyzed, so that more reasonable and effective measures for reducing the energy consumption are obtained.
Description
Technical Field
The application relates to the technical field of hotel energy consumption monitoring, in particular to a method and a device for predicting equipment energy consumption based on multiple linear regression.
Background
Energy consumption of hotel rooms is mainly used for air conditioners, lighting, televisions, small household appliances and the like, and compared with other types of energy consumption, the energy consumption of the hotel rooms has the characteristics that: if the types of electric equipment of different hotels are different, the number of rooms of a high-end hotel is large, the area of the rooms is large, a central air conditioner is adopted, and the central air conditioner is provided with an intelligent temperature controller, so that the temperature of the rooms and the wind speed of a fan can be monitored frequently; the middle and low-end chain hotels have fewer rooms and small area, and adopt split air conditioners. The running time of the rooms is different, the change of the annual average check-in rate causes the change of the load of the personnel in the hotel, for example, the check-in rate of holidays, golden weeks in travel and the like is higher; the selection of indoor environment and air conditioning parameters is adjusted according to the requirements of the living people, so the randomness is high, and the like. The above factors all result in different operating periods for each service facility of a hotel room and thus in the particularities of its energy consumption.
Therefore, how to predict the energy consumption of different electric equipment to obtain a more accurate energy consumption data, and then carry out data analysis on such data to finally obtain the most effective energy consumption saving measure becomes the problem that needs to be solved urgently in the field.
Disclosure of Invention
Therefore, the method and the device for predicting the energy consumption of the equipment based on the multiple linear regression are provided, and the problem that how to predict the energy consumption of different electric equipment in the prior art so as to obtain the most effective energy consumption saving measure is solved.
In order to achieve the above object, the present application provides the following technical solutions:
in a first aspect, a method for predicting energy consumption of a device based on multiple linear regression includes:
obtaining influence factors influencing energy consumption of the electric equipment;
carrying out standardization processing on the influence factors according to a first formula;
the first formula is:
wherein,
is the sample mean of the j index, s j Is the sample standard deviation of the jth index, m is the number of energy consumption influence factors, namely, the index variables for performing principal component analysis are m, n is the number of electric equipment, namely, n evaluation objects are total, and the value of the jth index variable of the ith evaluation object is x ij ；
Calculating a correlation coefficient matrix according to a second formula;
the second formula is: r = (R) ij ) m×m Wherein
r ii ＝1，r ii ＝r ij ，r ij Is the correlation coefficient of the ith index and the jth index;
calculating a characteristic value and a characteristic vector, and obtaining m second index variables according to the characteristic vector;
the second index variable is:
selecting proper principal components, calculating a comprehensive score, and selecting a variable with a higher comprehensive score as a principal component variable;
establishing a multiple linear regression equation according to the selected principal component variable with higher comprehensive score;
checking the significance of the multiple linear regression equation;
taking the maximum probability value P corresponding to the t value max ；
Judging the maximum probability value P max Whether the value is less than or equal to a critical value;
and if the energy consumption is less than or equal to the critical value, outputting a multiple linear regression equation between the energy consumption of the electric equipment and each influence factor.
Further, selecting a proper principal component, calculating a comprehensive score, and selecting a variable with a higher comprehensive score as a principal component variable specifically comprises:
calculating the information contribution rate and the accumulated contribution rate of the characteristic value according to a third formula and a fourth formula;
the third formula is:
the fourth formula is:
when alpha is p When the index value is close to 1, the first p index variables y are selected 1 ，y 2 ，…，y p As p main components, replacing the original m index variables;
calculating a composite score of each principal component according to a fifth formula;
the fifth formula is:
and taking the variable with higher comprehensive score as a principal component variable.
Further, said α is p 0.85,0.90 or 0.95.
Further, calculating a constant term and a coefficient value of the multiple linear regression equation according to a fifth formula and a sixth formula;
the fifth formula is:
wherein,
the sixth formula is:
further, before checking the significance of the regression equation, the method further comprises: the sum of the squares of the total deviations is decomposed into a sum of the squares of the regression and the residual.
Further, checking the significance of the regression equation includes checking the significance of the regression equation and checking the significance of partial regression coefficients.
Further, the significance of the regression equation is checked by using an F distribution.
Furthermore, F distribution and t distribution are adopted for testing the partial regression coefficient.
Further, the critical value is 0.05.
In a second aspect, an apparatus for predicting energy consumption of a device based on multiple linear regression includes:
the influence factor acquisition module is used for acquiring influence factors influencing the energy consumption of the electric equipment;
the standardization module is used for carrying out standardization processing on the influence factors according to a first formula;
the first formula is:
wherein,
is the sample mean of the j index, s j Is the sample standard deviation of the jth index, and m is the number of energy consumption influencing factorsThat is, the index variables for performing the principal component analysis are m, n is the number of the electric devices, that is, n evaluation objects are shared, and the value of the jth index variable of the ith evaluation object is x ij ；
The correlation coefficient matrix calculation module is used for calculating a correlation coefficient matrix according to a second formula;
the second formula is: r = (R) ij ) m×m In which
r ii ＝1，r ii ＝r ij ，r ij Is the correlation coefficient of the ith index and the jth index;
the characteristic value and characteristic vector calculation module is used for calculating a characteristic value and a characteristic vector and obtaining m second index variables according to the characteristic vector;
the second index variable is:
the principal component variable selection module is used for selecting proper principal components, calculating a comprehensive evaluation value and selecting a variable with a higher comprehensive evaluation value as a principal component variable;
the multivariate linear regression equation establishing module is used for establishing a multivariate linear regression equation according to the selected higher principal component variable of the comprehensive evaluation value;
the significance testing module is used for testing the significance of the multiple linear regression equation;
a probability value selection module for selecting the maximum probability value P corresponding to the t value max ；
A judging module for judging the maximum probability value P max Whether the value is less than or equal to a critical value;
and if the energy consumption is less than or equal to the critical value, outputting a multiple linear regression equation between the energy consumption of the electric equipment and each influence factor.
Compared with the prior art, the method has the following beneficial effects that:
the application provides a method and a device for predicting equipment energy consumption based on multiple linear regression, wherein influence factors influencing the energy consumption of electric equipment are used as independent variables through a principal component analysis method, all related variables are linearly transformed into another group of irrelevant variables through continuous transformation of coordinates, and the former variables with higher contribution rate are selected as principal components through calculation, so that the purposes of reflecting the energy consumption of the electric equipment by using fewer independent variables and avoiding the problem of multiple collinearity among the independent variables can be achieved. And establishing a multiple linear regression equation according to the principal component variables obtained by the principal component analysis method, and predicting the energy consumption of each electric device by using the multiple linear regression method. According to the method and the device, the energy consumption of the electric equipment is predicted, and the predicted energy consumption data are analyzed, so that more reasonable and effective energy consumption reduction measures are obtained.
Drawings
To more intuitively illustrate the prior art and the present application, several exemplary drawings are given below. It should be understood that the specific shapes, configurations and illustrations in the drawings are not to be construed as limiting, in general, the practice of the present application; for example, it is within the ability of those skilled in the art to make routine adjustments or further optimizations based on the technical concepts disclosed in the present application and the exemplary drawings, for the increase/decrease/attribution of certain units (components), specific shapes, positional relationships, connection manners, dimensional ratios, and the like.
Fig. 1 is an overall flowchart of an energy consumption prediction method based on multiple linear regression according to an embodiment of the present disclosure;
FIG. 2 is a flowchart of principal component analysis provided in an embodiment of the present application;
FIG. 3 is a flowchart of a multiple linear regression equation establishment process provided in an embodiment of the present application;
FIG. 4 is a flowchart of regression equation verification provided in accordance with an embodiment of the present application;
FIG. 5 is a flowchart illustrating simulation operation according to an embodiment of the present application;
fig. 6 is a flowchart of Python operation according to an embodiment of the present application.
Detailed Description
The present application will be described in further detail below with reference to specific embodiments thereof, with reference to the accompanying drawings.
In the description of the present application: "plurality" means two or more unless otherwise specified. The terms "first", "second", "third", and the like in this application are intended to distinguish one referenced item from another without having a special meaning in technical connotation (e.g., should not be construed as emphasizing a degree or order of importance, etc.). The terms "comprising," "including," "having," and the like, are intended to be inclusive and mean "not limited to" (some elements, components, materials, steps, etc.).
In the present application, terms such as "upper", "lower", "left", "right", "middle", and the like are usually used for the purpose of visual understanding with reference to the drawings, and are not intended to be an absolute limitation of the positional relationship in an actual product. Changes in these relative positional relationships are also considered to be within the scope of the present disclosure without departing from the technical concepts disclosed in the present disclosure.
The first embodiment is as follows:
referring to fig. 1, the present embodiment provides an energy consumption prediction method based on multiple linear regression, which mainly includes two major modules, that is, establishing a mathematical model for principal component analysis, obtaining principal components using SPSS, and establishing a mathematical model for multiple linear regression to obtain a prediction result. The method comprises the following steps of calculating a multiple linear regression mathematical model to obtain a specific energy consumption result of the electric equipment, and comparing the specific energy consumption result with actual energy consumption data to prove the reliability of the multiple linear regression algorithm, wherein the specific energy consumption result comprises the following steps:
s1: establishing a mathematical model of principal component analysis
Referring to fig. 2, the mathematical model of the principal component analysis calculates the contribution rate of the factors affecting the energy consumption, such as the outdoor air temperature, the room air humidity, the heat preservation performance of the room, the room area, the room orientation, the air flow velocity of the room, the heat loss through the gaps between the doors and windows, etc., to obtain the main influencing factors.
S11: carrying out standardization processing on the original data;
specifically, assuming that there are m influencing factors on energy consumption, the index variables for principal component analysis are m: x is a radical of a fluorine atom 1 ,x 2 ,...,x m . The electric equipment is n, namely n evaluation objects are arranged in total, and the value of the jth index variable of the ith evaluation object is x ij . Each index value x ij Conversion into a standardized index
Wherein,
is the sample mean of the j index, s j Is the sample standard deviation of the j-th index.
S12: calculating a correlation coefficient matrix R;
specifically, the correlation coefficient matrix R = (R) ij ) m×m
Wherein r is ii ＝1，r ii ＝r ij ，r ij Is the correlation coefficient of the ith index and the jth index.
S13: calculating a characteristic value and a characteristic vector;
specifically, the eigenvalues of the index variables need to be sorted from large to small, so that the eigenvalue λ of the correlation coefficient matrix R is calculated 1 ≥λ 2 ≥...≥λ m Not less than 0, and corresponding feature vector u 1 ,u 2 ,...,u m Wherein u is j ＝(u 1j ,u 2j ,…,u nj ) T The feature vectors form m new index variables, as shown in the following formula:
wherein, y 1 Is the first principal component, y 2 Is the second principal component, y 3 Is the third main component, 8230;, y m Is the m-th main component.
S14: and selecting proper main components and calculating a comprehensive evaluation value.
S141: calculating the information contribution rate and the accumulated contribution rate of the characteristic value;
specifically, the characteristic value λ is calculated j (j =1,2,..., m) and the cumulative contribution rate, i.e. principal component y j The information contribution rate of (1);
wherein alpha is p As a principal component y 1 ，y 2 ，…，y p When the cumulative contribution rate of p Is close to 1 (alpha) p =0.85,0.90, 0.95), the first p index variables y are selected 1 ，y 2 ，…，y p As p principal components, the original m index variables are replaced, so that p principal components can be comprehensively analyzed.
S142: calculating a comprehensive score;
specifically, the composite score:
wherein, b j And evaluating the information contribution rate of the jth principal component according to the comprehensive score value.
S2: mathematical model for establishing multiple linear regression
Referring to fig. 3, in the present application, multiple linear regression is selected to predict energy consumption, and since the linear regression model determines that the variables are related, under observation of a large amount of data, the variables will show a certain regularity, which can be expressed by means of a functional relation. Therefore, when a multiple linear regression model is established, it is necessary to standardize input data of the mathematical model, and to specify input variables in the mathematical model and variables to be measured. And then calculating according to the relation between the independent variable and the dependent variable to obtain an expression of a multiple linear regression equation, and then carrying out significance test on the multiple linear regression model. The method comprises the following steps:
s21: analyzing the correlation;
specifically, the correlation analysis is used to measure the degree of linear correlation between the dependent variable and the independent variables, i.e. to represent the observed value Y and the estimated value
The degree of correlation between them. The correlation is generally represented by a correlation coefficient r:
wherein R is a coefficient of determination.
S22: establishing a regression equation;
s221: the form of the linear regression equation is assumed to be as follows:
Y＝β 0 +β 1 X 1 +β 2 X 2 +…+β m X m +e
the above formula showsThe amount of strain Y in the data can be approximately represented as the independent variable X 1 ,X 2 ,...,X m Is a linear function of (a). Wherein beta is 0 Is a constant term, β 1 ,β 2 ,…,β m Is a partial regression coefficient, which means that X is constant for other independent variables j The average change of Y is increased or decreased by one unit, and e is the random error after the influence of m variables on Y is removed.
S222: a least squares method;
to make it possible to
Is closest to the sample data, i.e. is
To be closest to the value of sample Y:
if a linear regression equation exists, the coefficient b before each term X is required i And constant term b 0 . If the obtained regression equation is to have a better effect, the estimated value is closest to the sample value Y. The sum of squared errors Q is minimized, and the least square method is used, so that the problem of positive and negative offset is avoided. Then the most efficient problem of the regression model is converted into the minimum problem of Q
Calculating the minimum value of Q, substituting the estimated values, and respectively comparing X with X 1 ...X m Obtaining a linear equation set by solving the partial derivative, and obtaining b by solving through a matrix method 0 ...b m I.e., constant terms and coefficient values.
S23: significance testing of regression equations
Referring to fig. 4, 5 and 6, the regression equation is obtained based on the sample data, so it is necessary to verify whether the regression equation has reached the significance level for the whole, and whether the whole can be measured, specifically:
s231: and decomposing the sum of squares of the total deviation into a regression sum of squares and a residual sum of squares. For the regression equation, the smaller the sum of squared residuals is the better, since it corresponds to the difference between the predicted and sample values, which needs to be checked with the F-distribution. The purpose of the test is to determine whether the regression sum of squares is greater than the residual sum of squares, and if so, then the equation can be said to be statistically significant.
SS General assembly ＝SS Go back to +SS Disabled person
S232: f, checking the significance of the regression equation by distribution;
SS general assembly ＝SS Go back to +SS Disabled person
F～F(m,n-m-1)
α＝0.05
Where m is the number of variables and n is the sample capacity, the F-profile can be obtained. The critical value of the curve at 0.01 or 0.05 can be compared with the obtained F value, so as to judge whether the regression equation is significant.
S233: calculating the determination coefficient R 2
Wherein R is more than or equal to 0 2 1, the closer the coefficient is to 1, the better the fitting degree of the regression equation to the data is.
S234: significance of partial regression coefficients for F distribution test
df 1 ＝1
df 2 ＝n-m-1
Wherein SS Go back to (X j ) Represents partial regression sum of squares, the larger the value of which indicates that the corresponding independent variable is more important, and only in the case that m independent variables are completely independent is: sigma SS Chinese character hui (X j )＝SS Go back to 。
S235: t-distribution test for significance of partial regression coefficients
df＝n-m-1
Wherein, b j Is an estimate of the partial regression coefficients,
is b j Standard deviation of (d).
S3: get theMaximum probability value P corresponding to t value max ；
S4: judging whether P is present max ≤0.05；
If not, receiving H0: re-screening variables if the linear relation is not obvious;
if so, rejecting H0: the linear relation is not obvious, and a linear regression equation is output.
S5: the result is characterized;
specifically, the main influence factors of the central air conditioner are subjected to multiple linear regression prediction, and after the program operation is finished, an equation of each influence factor and the specific energy consumption of the central air conditioner, a percentage point diagram of central air conditioner energy consumption prediction, an energy consumption prediction fitting diagram and the like can be generated to represent the energy consumption condition of the central air conditioner.
According to the method, influencing factors influencing the energy consumption of the electric equipment are used as independent variables through a principal component analysis method, all relevant variables are linearly transformed into another group of irrelevant variables through continuous transformation of coordinates, and the first several variables with higher contribution rates are selected as principal components through calculation, so that the purposes of reflecting the energy consumption of the electric equipment by using fewer independent variables and avoiding the problem of multiple collinearity among the independent variables can be achieved. And establishing a multiple linear regression equation according to the principal component variables obtained by the principal component analysis method, and predicting the energy consumption of each electric device by using the multiple linear regression method. Through predicting the energy consumption of the electric equipment, the predicted energy consumption data is analyzed, and more reasonable and effective measures for reducing the energy consumption are obtained.
Example two:
the embodiment provides an apparatus for predicting energy consumption of a device based on multiple linear regression, which includes:
the influence factor acquisition module is used for acquiring influence factors influencing the energy consumption of the electric equipment;
the standardization module is used for carrying out standardization processing on the influence factors according to a first formula;
the first formula is:
wherein,
is the sample mean of the j index, s j Is the sample standard deviation of the jth index, m is the number of energy consumption influence factors, namely, the index variables for performing principal component analysis are m, n is the number of electric equipment, namely, n evaluation objects are total, and the value of the jth index variable of the ith evaluation object is x ij ；
The correlation coefficient matrix calculation module is used for calculating a correlation coefficient matrix according to a second formula;
the second formula is: r = (R) ij ) m×m Wherein
r ii ＝1，r ii ＝r ij ，r ij Is the correlation coefficient of the ith index and the jth index;
the eigenvalue and eigenvector calculation module is used for calculating an eigenvalue and an eigenvector and obtaining m second index variables according to the eigenvector;
the second index variable is:
the principal component variable selection module is used for selecting proper principal components, calculating a comprehensive evaluation value and selecting a variable with a higher comprehensive evaluation value as a principal component variable;
the multiple linear regression equation establishing module is used for establishing a multiple linear regression equation according to the selected higher principal component variable of the comprehensive evaluation value;
the significance testing module is used for testing the significance of the multiple linear regression equation;
a probability value selection module for selecting the maximum probability value P corresponding to the t value max ；
A judging module for judging the maximum probability value P max Whether the value is less than or equal to a critical value;
and if the energy consumption is less than or equal to the critical value, outputting a multiple linear regression equation between the energy consumption of the electric equipment and each influence factor.
For specific limitations of the device energy consumption prediction apparatus based on multiple linear regression, reference may be made to the above limitations of the device energy consumption prediction method based on multiple linear regression, and details are not repeated here.
In summary, the present application has the following advantages:
(1) The used principal component analysis method reveals the relation among a plurality of variables through a few principal components, expresses the energy consumption result of the central air conditioner through a plurality of main factors, and reduces the calculation amount of energy consumption under the condition of ensuring the result.
(2) The electric equipment which has no specific running time length, power and the like and can not directly calculate the energy consumption by a formula, such as a central air conditioner and the like, is calculated by a multiple linear regression method.
(3) A plurality of influence factors of the electric equipment are found, and the optimal combination of the plurality of influence factors is used for forecasting together, so that the forecasting is more effective and more practical than that of one influence factor.
(4) The linear relation between the power utilization equipment and each influence factor can be analyzed visually and rapidly through the multivariate linear regression, the correlation degree and the fitting degree between each factor can be obtained accurately, and the prediction accuracy is improved.
All the technical features of the above embodiments can be arbitrarily combined (as long as there is no contradiction between the combinations of the technical features), and for brevity of description, all the possible combinations of the technical features in the above embodiments are not described; such non-explicitly written embodiments should be considered as being within the scope of the present description.
The present application has been described in considerable detail with reference to the foregoing general description and specific examples. It should be understood that several conventional adaptations or further innovations of these specific embodiments may also be made based on the technical idea of the present application; however, such conventional modifications and further innovations can also fall into the scope of the claims of the present application as long as they do not depart from the technical idea of the present application.
Claims (10)
1. A method for predicting equipment energy consumption based on multiple linear regression is characterized by comprising the following steps:
obtaining influence factors influencing the energy consumption of the electric equipment;
carrying out standardization processing on the influence factors according to a first formula;
the first formula is:
wherein,
is the sample mean of the j index, s j Is the sample standard deviation of the jth index, m is the number of energy consumption influencing factors, namely m index variables for performing principal component analysis, n is the number of electric equipment, namely n evaluation objects are total, and the value of the jth index variable of the ith evaluation object is x ij ；
Calculating a correlation coefficient matrix according to a second formula;
the second formula is: r = (R) ij ) m×m In which
r ii ＝1，r ii ＝r ij ，r ij Is the correlation coefficient of the ith index and the jth index;
calculating a characteristic value and a characteristic vector, and obtaining m second index variables according to the characteristic vector;
the second index variable is:
selecting proper principal components, calculating a comprehensive score, and selecting a variable with a higher comprehensive score as a principal component variable;
establishing a multiple linear regression equation according to the selected principal component variable with higher comprehensive score;
checking the significance of the multiple linear regression equation;
taking the maximum probability value P corresponding to the t value max ；
Judging the maximum probability value P max Whether the value is less than or equal to a critical value;
and if the energy consumption is less than or equal to the critical value, outputting a multiple linear regression equation between the energy consumption of the electric equipment and each influence factor.
2. The method for predicting the energy consumption of equipment based on the multiple linear regression as claimed in claim 1, wherein the method comprises the steps of selecting appropriate principal components, calculating a comprehensive score, selecting a variable with a higher comprehensive score as a principal component variable, and specifically comprises the following steps:
calculating the information contribution rate and the accumulated contribution rate of the characteristic value according to a third formula and a fourth formula;
the third formula is:
the fourth formula is:
when alpha is p When the index value is close to 1, the first p index variables y are selected 1 ，y 2 ，…，y p As p main components, replacing the original m index variables;
calculating a composite score of each principal component according to a fifth formula;
the fifth formula is:
and taking the variable with higher comprehensive score as a principal component variable.
3. The method for predicting the energy consumption of equipment based on the multiple linear regression as claimed in claim 2, wherein the alpha is p 0.85,0.90 or 0.95.
4. The multiple linear regression-based equipment energy consumption prediction method according to claim 1, characterized in that constant terms and coefficient values of the multiple linear regression equation are calculated according to a fifth formula and a sixth formula;
the fifth formula is:
wherein,
the sixth formula is:
5. the method for predicting the energy consumption of the equipment based on the multiple linear regression is characterized in that the step of verifying the significance of the regression equation comprises the following steps: the sum of the squares of the total deviations is decomposed into a sum of the squares of the regression and the residual.
6. The multiple linear regression-based equipment energy consumption prediction method of claim 1, wherein verifying the significance of the regression equation comprises verifying the significance of a regression equation and verifying the significance of partial regression coefficients.
7. The method for predicting the energy consumption of the equipment based on the multiple linear regression is characterized in that F distribution is adopted for checking the significance of the regression equation.
8. The method for predicting the energy consumption of the equipment based on the multiple linear regression is characterized in that F distribution and t distribution are adopted for testing when testing the partial regression coefficients.
9. The multiple linear regression-based plant energy consumption prediction method of claim 1, wherein the threshold value is 0.05.
10. An apparatus for predicting energy consumption of a device based on multiple linear regression, comprising:
the influence factor acquisition module is used for acquiring influence factors influencing the energy consumption of the electric equipment;
the standardization module is used for carrying out standardization processing on the influence factors according to a first formula;
the first formula is:
wherein,
is the sample mean of the j index, s j Is the sample standard deviation of the jth index, m is the number of energy consumption influencing factors, namely m index variables for performing principal component analysis, n is the number of electric equipment, namely n evaluation objects are total, and the value of the jth index variable of the ith evaluation object is x ij ；
The correlation coefficient matrix calculation module is used for calculating a correlation coefficient matrix according to a second formula;
the second formula is: r = (R) ij ) m×m Wherein
r ii ＝1，r ii ＝r ij ，r ij Is the correlation coefficient of the ith index and the jth index;
the characteristic value and characteristic vector calculation module is used for calculating a characteristic value and a characteristic vector and obtaining m second index variables according to the characteristic vector;
the second index variable is:
the principal component variable selection module is used for selecting proper principal components, calculating a comprehensive evaluation value and selecting a variable with a higher comprehensive evaluation value as a principal component variable;
the multivariate linear regression equation establishing module is used for establishing a multivariate linear regression equation according to the selected higher principal component variable of the comprehensive evaluation value;
the significance testing module is used for testing the significance of the multiple linear regression equation;
a probability value selection module for selecting the maximum probability value P corresponding to the t value max ；
A judging module for judging the maximum probability value P max Whether the value is less than or equal to a critical value;
and if the energy consumption is less than or equal to the critical value, outputting a multiple linear regression equation between the energy consumption of the electric equipment and each influence factor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210827474.8A CN115169707A (en) | 2022-07-14 | 2022-07-14 | Equipment energy consumption prediction method and device based on multiple linear regression |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210827474.8A CN115169707A (en) | 2022-07-14 | 2022-07-14 | Equipment energy consumption prediction method and device based on multiple linear regression |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115169707A true CN115169707A (en) | 2022-10-11 |
Family
ID=83495463
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210827474.8A Pending CN115169707A (en) | 2022-07-14 | 2022-07-14 | Equipment energy consumption prediction method and device based on multiple linear regression |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115169707A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115809730A (en) * | 2022-11-28 | 2023-03-17 | 东北石油大学 | Large crude oil storage tank heat loss prediction method |
-
2022
- 2022-07-14 CN CN202210827474.8A patent/CN115169707A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115809730A (en) * | 2022-11-28 | 2023-03-17 | 东北石油大学 | Large crude oil storage tank heat loss prediction method |
| CN115809730B (en) * | 2022-11-28 | 2023-05-09 | 东北石油大学 | Large crude oil storage tank heat loss prediction method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Multi-criteria building energy performance benchmarking through variable clustering based compromise TOPSIS with objective entropy weighting | |
| Wang et al. | Online model-based fault detection and diagnosis strategy for VAV air handling units | |
| Jiang et al. | Applying grey forecasting to predicting the operating energy performance of air cooled water chillers | |
| Yalcintas | An energy benchmarking model based on artificial neural network method with a case example for tropical climates | |
| CN113036913B (en) | Method and device for monitoring state of comprehensive energy equipment | |
| CN113240184B (en) | Building space unit cold load prediction method and system based on federal learning | |
| CN110333962B (en) | Electronic component fault diagnosis model based on data analysis and prediction | |
| Gao et al. | Fault detection and diagnosis method for cooling dehumidifier based on LS-SVM NARX model | |
| Kang et al. | Bayesian-Emulator based parameter identification for calibrating energy models for existing buildings | |
| Qian et al. | Power consumption and energy efficiency of VRF system based on large scale monitoring virtual sensors | |
| Noh et al. | Data-driven forecasting algorithms for building energy consumption | |
| CN106295858A (en) | A kind of electric energy meter non-health degree Forecasting Methodology | |
| CN115827411B (en) | On-line monitoring and operation and maintenance assessment system and method for automation equipment | |
| CN111161095A (en) | Method for detecting abnormal consumption of building energy | |
| CN114169254A (en) | Abnormal energy consumption diagnosis method and system based on short-term building energy consumption prediction model | |
| CN115169707A (en) | Equipment energy consumption prediction method and device based on multiple linear regression | |
| CN113988655A (en) | Power transmission line running state evaluation method considering multiple meteorological factors | |
| CN114200839A (en) | Office building energy consumption intelligent control model for dynamic monitoring of coupled environmental behaviors | |
| Sözer et al. | Predicting the indoor thermal data for heating season based on short-term measurements to calibrate the simulation set-points | |
| CN115511100A (en) | Air conditioner load regression prediction method based on environmental temperature related data learning | |
| JP2004234302A (en) | Process management device | |
| CN112580254B (en) | Multi-dimensional analysis and early warning method for converter transformer based on multi-level decision fusion | |
| CN113591402A (en) | Digital power transformer health state fuzzy comprehensive evaluation method | |
| CN116738856B (en) | Building energy consumption prediction method and system | |
| Su et al. | Operating optimization of air-conditioning water system in a subway station using data mining and dynamic system models |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |