Air conditioner flexible adjustment reference temperature calculation method and system, electronic device and medium

Abstract

The application discloses an air conditioner flexible regulation reference temperature calculation method, system, electronic equipment and medium, and belongs to the technical field of air conditioner load regulation and control. The air conditioner flexible regulation reference temperature calculation method provided by the application obtains a calculation result about an air conditioner reference temperature by constructing a data processing model, a feature processing model and an air conditioner set temperature relationship model, and can accurately realize the calculation of the air conditioner flexible regulation reference temperature, so that the application can provide a data basis for regulating air conditioner load. The application uses a normality analysis method to analyze and process original data, improves the accuracy of the features, and is beneficial to simplifying calculation. Meanwhile, based on the physical characteristics of the air conditioner, the air conditioner regulation reference temperature setting is calculated according to the original air conditioner data, so that the physical actuality is considered and the method is practical and convenient to implement. Therefore, the application provides a basis for further tapping the adjustable potential of the air conditioner and realizing more reasonable air conditioner regulation and control in the power industry.

Description

Technical Field

[0001] This invention relates to a method, system, electronic device, and medium for calculating the reference temperature for flexible air conditioning control, and belongs to the field of air conditioning load control technology. Background Technology

[0002] In recent years, global climate change has led to frequent natural disasters, particularly the frequent occurrence of high temperatures in summer, resulting in a significant increase in air conditioning load and placing a heavy burden on the power system. Under this severe supply and demand situation, the risk of power rationing is gradually increasing. With policies such as "prioritizing restrictions on electricity consumption by high-energy-consuming and high-polluting industries" and "power rationing without blackouts, rationing without restrictions on residential use," ensuring stable power supply and prices faces significant challenges. Cooling load, primarily from air conditioning, accounts for 36% of peak power load, a major reason for the widening peak-valley difference and increasingly pronounced peak load characteristics. Therefore, regulating air conditioning load has significant potential in alleviating the tight power supply and demand situation. As a flexible and adjustable load, controlling the temperature of public air conditioning systems can alleviate this pressure to some extent.

[0003] The premise of reasonably regulating the air conditioning load is to understand the user's set air conditioning reference temperature. However, in reality, due to the highly dispersed nature of the relevant setting data, it is difficult to obtain the user's set air conditioning reference temperature data. Furthermore, the air conditioning reference temperature setting of a particular user in a particular scenario also has a certain degree of randomness. Therefore, it is not suitable to directly use the query method to collect data, which in turn affects the exploration of the air conditioning's adjustable potential and makes it impossible to achieve more reasonable air conditioning regulation.

[0004] The information disclosed in this background section is only for understanding the background of the inventive concept, and therefore may include information that does not constitute prior art. Summary of the Invention

[0005] To address the aforementioned problems, or one of them, the present invention aims to provide a method for calculating the air conditioner's power and temperature by constructing a data processing model, a feature processing model, and an air conditioner set temperature relationship model. This method analyzes the original air conditioner power and temperature data, identifies outliers and missing data, processes the outliers, and uses regression equations to interpolate and fill in the missing data for missing data, thus obtaining initial air conditioner power and temperature data. Then, feature engineering is performed on the initial air conditioner power and temperature data, and the significance of the air conditioner power and temperature data features is compared using Pearson coefficient analysis. Data sets with high significance are selected to form the feature data of air conditioner power and temperature. Based on the feature data, a power characteristic curve of the air conditioner is created to obtain the coupling relationship between the set temperature difference and power change. Based on this coupling relationship, the air conditioner set temperature is predicted to obtain a calculated result regarding the air conditioner's reference temperature. This method can accurately calculate the reference temperature for flexible air conditioner regulation, providing a data foundation for adjusting air conditioner load. The solution is scientific, reasonable, and feasible for calculating the reference temperature for flexible air conditioner regulation.

[0006] To address the aforementioned problems or one of them, the second objective of this invention is to provide an air conditioning flexible adjustment reference temperature calculation system that, by setting up a data processing module, a feature processing module, and a reference temperature calculation module, obtains the calculation result of the air conditioning reference temperature, can accurately realize the calculation of the air conditioning flexible adjustment reference temperature, provides a data basis for adjusting the air conditioning load, and is a scientific, reasonable, and feasible air conditioning flexible adjustment reference temperature calculation system.

[0007] To address the aforementioned problems or one of the aforementioned problems, the third objective of this invention is to provide an air conditioning flexible adjustment reference temperature calculation system, electronic device, and medium that, by setting up a data processing module, a feature processing module, and a reference temperature calculation module, obtains the calculation result of the air conditioning reference temperature, can accurately realize the calculation of the air conditioning flexible adjustment reference temperature, and can provide a data basis for adjusting the air conditioning load. The solution is scientific, reasonable, and feasible.

[0008] To address the aforementioned problems or one of them, the fourth objective of this invention is to provide a method, system, electronic device, and medium for calculating the reference temperature for flexible air conditioning regulation. This method uses a normality analysis method to analyze and process raw data, improving the accuracy of the proposed features and simplifying calculations. Simultaneously, based on the physical characteristics of air conditioning, it calculates the reference temperature setting for air conditioning regulation based on the raw air conditioning data. This approach considers physical realities and is practical, easy to implement, and can further explore the adjustable potential of air conditioning, providing a data foundation for achieving more rational air conditioning control.

[0009] To achieve one of the above objectives, the first technical solution of the present invention is as follows:

[0010] A method for calculating the reference temperature for flexible air conditioning control includes the following steps:

[0011] The first step is to obtain the raw air conditioner power and temperature data;

[0012] The second step involves using a pre-built data processing model to analyze the original air conditioning power and temperature data, identify outliers and missing data, process the outliers, and use the regression equation method to interpolate the missing data to obtain the initial air conditioning power and temperature data.

[0013] The third step involves using a pre-built feature processing model to perform feature engineering on the initial air conditioning power and temperature data. The Pearson coefficient analysis method is then used to compare the significance of the features of the air conditioning power and temperature data. The air conditioning dataset with high significance is selected to form the feature data of air conditioning power and temperature.

[0014] The fourth step involves using the characteristic data of air conditioner power and temperature, and employing a pre-built air conditioner set temperature relationship model, to create the air conditioner power characteristic curve and obtain the coupling relationship between the set temperature difference and power change. Based on this coupling relationship, the air conditioner set temperature is predicted to obtain the calculation result of the air conditioner reference temperature, thus completing the calculation of the air conditioner flexible adjustment reference temperature.

[0015] This invention, through continuous exploration and experimentation, constructs a data processing model, a feature processing model, and an air conditioning set temperature relationship model. It analyzes the original air conditioning power and temperature data, identifies outliers and missing data, processes the outliers, and uses regression equations to interpolate and fill in the missing data for the missing data, obtaining initial air conditioning power and temperature data. Then, it performs feature engineering on the initial air conditioning power and temperature data and uses Pearson coefficient analysis to compare the significance of the air conditioning power and temperature data features, selecting the air conditioning dataset with high significance to form the feature data of air conditioning power and temperature. Based on the feature data of air conditioning power and temperature, it creates the power characteristic curve of the air conditioner, obtaining the coupling relationship between the set temperature difference and power change. Based on this coupling relationship, it predicts the air conditioning set temperature, obtaining a calculated result about the air conditioning reference temperature. This can accurately calculate the reference temperature for flexible air conditioning adjustment. Therefore, this invention can provide a data foundation for government regulatory departments to regulate air conditioning load. The solution is scientific, reasonable, and feasible.

[0016] Furthermore, this invention uses normality analysis to process the raw data, improving the accuracy of the proposed features and simplifying calculations. Simultaneously, based on the physical characteristics of air conditioning, it calculates the reference temperature setting for air conditioning based on the raw air conditioning data, taking into account both physical realities and practicality, making it easy to implement. Therefore, the air conditioning temperature calculation method of this invention provides a foundation for the power industry to further explore the adjustable potential of air conditioning and achieve more rational air conditioning control.

[0017] As a preferred technical measure:

[0018] In the second step, the data processing model obtains the initial air conditioning power and temperature data as follows:

[0019] Step 21: Obtain k sets of raw air conditioner power and temperature data;

[0020] Step 22: Perform normal distribution analysis on each set of original air conditioning power and temperature data, which includes the following:

[0021] For the i-th set of original air conditioning power and temperature data, calculate its mean, standard deviation, skewness, and kurtosis; and plot PP and QQ plots to identify outliers and missing values ​​in the original air conditioning power and temperature data.

[0022] Outliers are data points that deviate 20% from the diagonal in the QQ graph;

[0023] Step 23 involves processing outliers and missing values ​​in the original air conditioning power and temperature data, including the following:

[0024] Missing values ​​are handled using a regression equation method, and the calculation formula is as follows:

[0025]

[0026]

[0027] y = bx + a

[0028] The missing values ​​are calculated using the obtained regression equation coefficients, and the calculation formula is as follows:

[0029] Y = bX + a;

[0030] Where x i ,y i The horizontal and vertical axes correspond to the five data points, specifically the air conditioning time and temperature. , a and b are the average values ​​of the x and y coordinates of the five points, respectively; a and b are the coefficients of the interpolation function, respectively; and X is the missing value point X. i The x-coordinate of the given value is a known quantity, and the missing value point is X. i The ordinate Y represents the missing values; y = bx + a is the interpolation function; through interpolation, Y represents the missing value point X. i The interpolated ordinate supplementary value;

[0031] For abnormal points with short-term temperature anomalies, interpolation is used for replacement; for data with high sampling frequency and small proportion, the entire data record is deleted directly; for abnormal data with periodic jumps, the entire row is deleted.

[0032] As a preferred technical measure:

[0033] In the third step, the feature processing model obtains the feature data of air conditioner power and temperature as follows:

[0034] Step 31: Obtain initial air conditioner power and temperature data;

[0035] Step 32: Based on the initial air conditioning power and temperature data, use the Pearson coefficient analysis method to calculate the Pearson coefficient between the air conditioning power and the set temperature data;

[0036] Step 33: Based on the Pearson coefficient, compare the significance of the air conditioner power and set temperature data to determine the strength of the correlation.

[0037] When the absolute value of the Pearson coefficient between the air conditioner power and the set temperature data is greater than 0.4, it indicates that the correlation between the two is strong, that is, the significance is high.

[0038] Step 34: Select the air conditioner dataset with high significance to form the feature data of air conditioner power and temperature.

[0039] As a preferred technical measure:

[0040] The Pearson coefficient between air conditioner power and set temperature data is the sample Pearson coefficient, and its calculation formula is as follows:

[0041]

[0042] Among them, Y 1,i ,Y 2,i These are the sample values ​​of the air conditioning characteristic variables Y1 and Y2, respectively. These are the average values ​​of the characteristic variables Y1 and Y2, respectively.

[0043] As a preferred technical measure:

[0044] In the fourth step, the method for obtaining the calculation result of the air conditioner reference temperature from the air conditioner set temperature relationship model is as follows:

[0045] Step 41: Obtain characteristic data of air conditioner power and temperature, and calculate the reduction rate α1 of air conditioner load power due to the air conditioner cooling set temperature. This rate characterizes the percentage reduction in air conditioner load power that occurs when the air conditioner cooling set temperature increases by 1℃. The calculation formula is as follows:

[0046]

[0047] Where λ1 is the set temperature t set The air conditioning load power at time t, λ2 is the set temperature t set Air conditioning load power at +1;

[0048] Step 42, according to t set ,t set The power change rate α1 of the air conditioner set temperature within 1℃ range is used to calculate the power change rate α for any temperature difference. n Power change rate α n The formula used to characterize the coupling relationship between the set temperature difference and power change is as follows:

[0049]

[0050] Where P1 and P2 are the power of the air conditioner at two adjacent time points, T1 and T2 are the temperatures at two adjacent time points, and α... n It is the power change rate of the air conditioner, that is, its adjustment capability at this moment;

[0051] Step 43, based on the power change rate α n Calculate the air conditioner set temperature T set This refers to the reference temperature for flexible air conditioning adjustment, and its calculation formula is as follows:

[0052] T set =T1-ΔT

[0053]

[0054] Where ΔT is the theoretical temperature adjustment amount, T set This is the predicted air conditioning set temperature at this time.

[0055] As a preferred technical measure:

[0056] It also includes a predictive evaluation model for evaluating the predicted air conditioning set temperature;

[0057] The predictive evaluation model includes the following:

[0058] The predicted air conditioning set temperature T was calculated using the standard mean absolute error (MAPE). set The evaluation is performed using the following formula:

[0059]

[0060] In the formula: y i Set the air conditioner to the actual temperature value; Set the predicted temperature value for the air conditioner; n is the sample size; for the standard mean absolute error (MAPE), a lower value is considered more accurate.

[0061] The calculated air conditioning set temperature is evaluated based on the abnormal data in the predicted values. The calculation formula is as follows:

[0062]

[0063] Where: n strange Abnormal data in the predicted values, those exceeding the normal temperature adjustment range of air conditioning [16,30], are identified as abnormal; N is the total number of samples; η is the ratio of abnormal data, the lower the value, the more accurate the prediction.

[0064] For outlier data in the predicted values, they are first deleted and then replaced using interpolation. The selected calculation time point is expanded to include data from one time point in between. The calculation formula is as follows:

[0065]

[0066] Where P1 and P3 are the power of the air conditioner at one time interval, T1 and T3 are the temperature at one time interval, and α nη This is the power change rate of the air conditioner at this time.

[0067] To achieve one of the above objectives, the second technical solution of the present invention is as follows:

[0068] A method for calculating the reference temperature for flexible air conditioning control includes the following:

[0069] Obtain raw air conditioner power and temperature data;

[0070] Using a pre-built data processing model, the original air conditioning power and temperature data are analyzed to identify outliers and missing data. Outliers are processed, and missing data are filled in by interpolation using regression equations to obtain the initial air conditioning power and temperature data.

[0071] By using a pre-built feature processing model, feature engineering is performed on the initial air conditioning power and temperature data. The significance of the features of the air conditioning power and temperature data is compared using the Pearson coefficient analysis method. The air conditioning dataset with high significance is selected to form the feature data of air conditioning power and temperature.

[0072] Based on the characteristic data of air conditioner power and temperature, a power characteristic curve of the air conditioner is generated using a pre-built air conditioner set temperature relationship model to obtain the coupling relationship between set temperature difference and power change; and the air conditioner set temperature is predicted based on the coupling relationship to obtain the calculation result of the air conditioner reference temperature.

[0073] Using a pre-built prediction and evaluation model, the predicted air conditioning set temperature is evaluated, and the reference temperature for flexible air conditioning adjustment is calculated.

[0074] To achieve one of the above objectives, the third technical solution of the present invention is as follows:

[0075] An air conditioning flexible adjustment reference temperature estimation system includes a data processing module, a feature processing module, and a reference temperature estimation module;

[0076] The data processing module is used to analyze the original air conditioner power and temperature data, identify outliers and missing data, process outliers, and use regression equations to interpolate missing data to obtain the initial air conditioner power and temperature data.

[0077] The feature processing module is used to perform feature engineering on the initial air conditioning power and temperature data, and to compare the significance of the features of the air conditioning power and temperature data using the Pearson coefficient analysis method. The air conditioning dataset with high significance is selected to form the feature data of air conditioning power and temperature.

[0078] The reference temperature calculation module is used to generate the power characteristic curve of the air conditioner based on the characteristic data of the air conditioner's power and temperature, obtain the coupling relationship between the set temperature difference and the power change, and predict the set temperature of the air conditioner based on the coupling relationship to obtain the calculation result of the reference temperature of the air conditioner, thus completing the calculation of the reference temperature for flexible adjustment of the air conditioner.

[0079] Through continuous exploration and experimentation, this invention, by setting up a data processing module, a feature processing module, and a reference temperature calculation module, obtains the calculation results of the air conditioning reference temperature, which can accurately realize the calculation of the reference temperature for flexible air conditioning adjustment. Therefore, this invention can provide a data foundation for government management departments to adjust air conditioning load. The solution is scientific, reasonable, and feasible.

[0080] Furthermore, this invention uses normality analysis to process the raw data, improving the accuracy of the proposed features and simplifying calculations. Simultaneously, based on the physical characteristics of air conditioning, it calculates the reference temperature setting for air conditioning based on the raw air conditioning data, taking into account both physical realities and practicality, making it easy to implement. Therefore, the air conditioning temperature calculation method of this invention provides a foundation for the power industry to further explore the adjustable potential of air conditioning and achieve more rational air conditioning control.

[0081] To achieve one of the above objectives, the fourth technical solution of the present invention is as follows:

[0082] An electronic device comprising:

[0083] One or more processors;

[0084] Storage device for storing one or more programs;

[0085] When the one or more programs are executed by the one or more processors, the one or more processors implement the above-described method for calculating the reference temperature for flexible air conditioning control.

[0086] To achieve one of the above objectives, the fifth technical solution of the present invention is as follows:

[0087] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for calculating a reference temperature for flexible air conditioning control.

[0088] Compared with existing technical solutions, the present invention has the following beneficial effects:

[0089] This invention, through continuous exploration and experimentation, constructs a data processing model, a feature processing model, and an air conditioning set temperature relationship model. It analyzes the original air conditioning power and temperature data, identifies outliers and missing data, processes the outliers, and uses regression equations to interpolate and fill in the missing data for the missing data, obtaining initial air conditioning power and temperature data. Then, it performs feature engineering on the initial air conditioning power and temperature data and uses Pearson coefficient analysis to compare the significance of the air conditioning power and temperature data features, selecting the air conditioning dataset with high significance to form the feature data of air conditioning power and temperature. Based on the feature data of air conditioning power and temperature, it creates the power characteristic curve of the air conditioner, obtaining the coupling relationship between the set temperature difference and power change. Based on this coupling relationship, it predicts the air conditioning set temperature, obtaining a calculated result about the air conditioning reference temperature. This can accurately calculate the reference temperature for flexible air conditioning adjustment. Therefore, this invention can provide a data foundation for government regulatory departments to regulate air conditioning load. The solution is scientific, reasonable, and feasible.

[0090] Furthermore, through continuous exploration and experimentation, this invention, by setting up a data processing module, a feature processing module, and a reference temperature calculation module, obtains the calculation results of the air conditioning reference temperature, which can accurately realize the calculation of the reference temperature for flexible air conditioning adjustment. Therefore, this invention can provide a data foundation for government management departments to adjust air conditioning load. The solution is scientific, reasonable, and feasible.

[0091] Furthermore, this invention uses normality analysis to process the raw data, improving the accuracy of the proposed features and simplifying calculations. Simultaneously, based on the physical characteristics of air conditioning, it calculates the reference temperature setting for air conditioning regulation based on the raw air conditioning data, taking into account both physical realities and practicality, making it easy to implement. Therefore, the air conditioning temperature calculation method of this invention provides a foundation for the power industry to further explore the adjustable potential of air conditioning and achieve more rational air conditioning control. Attached Figure Description

[0092] Figure 1 This is the first flowchart of the air conditioning flexible adjustment reference temperature calculation method of the present invention;

[0093] Figure 2 This is the second flowchart of the air conditioning flexible adjustment reference temperature calculation method of the present invention;

[0094] Figure 3This is the third flowchart of the air conditioning flexible adjustment reference temperature calculation method of the present invention. Detailed Implementation

[0095] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0096] Conversely, this invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of the invention as defined in the claims. Furthermore, to provide a better understanding of the invention, certain specific details are described in detail below. However, those skilled in the art will fully understand the invention even without these detailed descriptions.

[0097] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. The term "or / and" as used herein includes any and all combinations of one or more of the associated listed items.

[0098] like Figure 1 As shown, a specific embodiment of the air conditioning flexible adjustment reference temperature calculation method of the present invention is as follows:

[0099] A method for calculating the reference temperature for flexible air conditioning control includes the following steps:

[0100] The first step is to obtain the raw air conditioner power and temperature data;

[0101] The second step involves using a pre-built data processing model to analyze the original air conditioning power and temperature data, identify outliers and missing data, process the outliers, and use the regression equation method to interpolate the missing data to obtain the initial air conditioning power and temperature data.

[0102] The third step involves using a pre-built feature processing model to perform feature engineering on the initial air conditioning power and temperature data. The Pearson coefficient analysis method is then used to compare the significance of the features of the air conditioning power and temperature data. The air conditioning dataset with high significance is selected to form the feature data of air conditioning power and temperature.

[0103] The fourth step involves using the characteristic data of air conditioner power and temperature, and employing a pre-built air conditioner set temperature relationship model, to create the air conditioner power characteristic curve and obtain the coupling relationship between the set temperature difference and power change. Based on this coupling relationship, the air conditioner set temperature is predicted to obtain the calculation result of the air conditioner reference temperature, thus completing the calculation of the air conditioner flexible adjustment reference temperature.

[0104] like Figure 2 As shown, a specific embodiment of the air conditioning flexible adjustment reference temperature calculation method of the present invention is as follows:

[0105] A method for calculating the reference temperature for flexible air conditioning control includes the following:

[0106] First, the original air conditioner power and temperature data are imported. A brief analysis is performed on this data to identify outliers and missing data points. These outliers are then processed, and for missing data, interpolation using regression equations is used to fill in the gaps, resulting in initial air conditioner power and temperature data. Next, feature engineering is applied to this initial data, and the significance of power-temperature data features is compared using Pearson coefficient analysis. Data sets with high significance are selected to form the feature data for air conditioner power and temperature. Based on this feature data, a power characteristic curve for the air conditioner is created, and the coupling relationship between the set temperature difference and power variation is analyzed. Based on this coupling relationship, the air conditioner set temperature is predicted, yielding a calculated result. The calculated air conditioner set temperature is evaluated using absolute percentage error (MAPE) and outlier data. Therefore, this invention's air conditioner temperature calculation method can further explore the adjustable potential of air conditioners in the power industry, providing a foundation for more rational air conditioner control.

[0107] like Figure 3 As shown, a specific embodiment of the air conditioning flexible adjustment reference temperature calculation method of the present invention is as follows:

[0108] A method for calculating the reference temperature for flexible air conditioning control includes the following steps:

[0109] Step 1: Import the original air conditioner power and temperature data, perform a brief analysis of the original data to identify outliers and missing data points, process these outliers, and use the regression equation method to interpolate and fill in the missing data for the missing points.

[0110] Step 2: Perform feature engineering on the data and compare the significance of each data feature using Pearson coefficient analysis. Select the air conditioning dataset with higher significance and discard the other datasets.

[0111] Step 3: Plot the power characteristic curves of each air conditioner, analyze the relationship between the set temperature difference and power change, and predict the set temperature of the air conditioner based on this physical relationship.

[0112] Step 4: Use the absolute percentage error (MAPE) evaluation method and combine it with abnormal data to evaluate the calculated air conditioning set temperature, and make predictions again for abnormal data.

[0113] A specific embodiment of the data analysis and processing of this invention:

[0114] A brief analysis of the original data is performed to identify outliers and missing data points. Outliers are then addressed, and missing data is imputed using regression equations. The specific steps are as follows:

[0115] 1) Import k sets of original data.

[0116] 2) Perform normality analysis (SPSS) on each group of data.

[0117] For the i-th data set: calculate its mean, standard deviation, skewness, and kurtosis. Plot the PP and QQ plots to identify outliers and missing values ​​in the data set. Data points in the QQ plot that deviate 20% from the diagonal are considered outliers.

[0118] 3) Handle outlier data in the data set:

[0119] ① Handling missing values

[0120] For data showing abnormal temperatures over a short period, interpolation is used to replace the data. For other features, especially those with high sampling frequency but low proportion, the entire data record is deleted.

[0121] The interpolation method uses a regression equation:

[0122] For the missing point X i X before and after i-2 ,X i-1 ,X i+1 ,X i+2 ,X i+3 A simple linear regression equation was calculated for the five data points.

[0123]

[0124]

[0125] y = bx + a

[0126] Using the obtained regression equation coefficients, calculate the missing value Y. i =bX i +a.

[0127] Where x i ,y iThe x and y coordinates correspond to the five data points, specifically the air conditioning time and temperature. These are the average values ​​of the x and y coordinates of the five points, respectively. a and b are the coefficients of the interpolation function, and y = bx + a is the interpolation function. Through interpolation, Y... i Then it is the missing point X. i The supplementary value obtained through interpolation.

[0128] ② Handling outliers

[0129] For anomalies related to temperature, interpolation is also used for replacement. However, for periodic jump-type anomalies, it is difficult to guarantee the reliability of other features, so the entire row is deleted.

[0130] A specific embodiment of this invention using Pearson coefficient analysis to compare the significance of various data features:

[0131] Feature engineering was performed on the data, and the significance of each data feature was compared using Pearson coefficient analysis. The air conditioning dataset with higher significance was selected, while other datasets were discarded. Specifically:

[0132] Different air conditioners exhibit varying correlations between their power output and set temperature; only data combinations with strong correlations are suitable as prediction datasets. Therefore, Pearson coefficient analysis is used to compare the significance of the air conditioner power output and set temperature data characteristics to determine the strength of the correlation. The correlation is compared using Pearson coefficient analysis to calculate the Pearson coefficient between the air conditioner power output and set temperature data characteristics.

[0133] Pearson coefficient:

[0134] The Pearson correlation coefficient between two variables is defined as the product of the covariance of the two air conditioning characteristic variables and their standard deviations:

[0135]

[0136] In the formula: cov(Y1,Y2) is the covariance of the air conditioning characteristic variables Y1 and Y2; and The standard deviations of Y1 and Y2 are respectively; the correlation coefficient is 1. The range of values ​​is [-1, 1], where -1 represents a perfect negative correlation and +1 represents a perfect positive correlation.

[0137] The above formula defines the population correlation coefficient, commonly represented by the lowercase Greek letter ρ (rho). By estimating the sample covariance and standard deviation, the sample correlation coefficient (sample Pearson coefficient) can be obtained, commonly represented by the lowercase English letter r.

[0138]

[0139] Among them, Y 1,i ,Y 2,i These are the sample values ​​of the air conditioning characteristic variables Y1 and Y2, respectively. These are the average values ​​of the characteristic variables Y1 and Y2, respectively.

[0140] If the absolute value of the calculated set temperature and power Pearson coefficient is greater than 0.4, the next step can be carried out; otherwise, the prediction method cannot be used.

[0141] A specific embodiment of the present invention for predicting the set temperature of an air conditioner:

[0142] Based on the air conditioner power data in step 1, the power characteristic curves of each air conditioner are plotted, the relationship between the set temperature difference and power change is analyzed, and the set temperature of the air conditioner is predicted based on this physical relationship.

[0143] 1) For the air conditioner power characteristic curve, perform characteristic analysis on the load curve of each load unit.

[0144] First, calculate the air conditioning load reduction rate α1 caused by the air conditioning cooling set temperature. This means that the air conditioning load power can be reduced by increasing the set temperature by 1℃.

[0145] For T set ,T set The calculation method for the power change rate α1 of a 1℃ air conditioner set temperature difference between +1℃ is as follows:

[0146]

[0147] Where λ1 is the set temperature T set The air conditioning load power at that time, λ2 is the set temperature T set Air conditioning load power at +1.

[0148] 2) Based on the above principle, the power change rate α for any temperature difference can be calculated. n And calculate the air conditioner set temperature T set This refers to the reference temperature for flexible air conditioning adjustment.

[0149]

[0150]

[0151] T set =T1-ΔT

[0152] Where P1 and P2 are the power of the air conditioner at two adjacent time points, T1 and T2 are the temperatures at two adjacent time points, and α... n It is the power change rate of the air conditioner, that is, its regulating capacity at this moment. ΔT is the theoretical temperature regulation amount, T setThis is the predicted temperature set by the air conditioner at this time.

[0153] A specific embodiment of the present invention for evaluating the predicted value of air conditioner set temperature:

[0154] The predicted value T of the air conditioning set temperature was calculated using the standard mean absolute error (MAPE). set An evaluation is conducted, and outliers in the predicted values ​​are calculated to assess the calculated air conditioning setpoint temperature. Outliers are then re-predicted.

[0155] The specific evaluation method and calculation method for the standard mean absolute error are as follows:

[0156]

[0157] In the formula: y i Set the air conditioner to the actual temperature value; Set a predicted temperature value for the air conditioner; n is the sample size. For MAPE, a lower value is considered more accurate.

[0158] The outlier ratio is calculated using the following formula:

[0159]

[0160] Where: n strange Abnormal data in the predicted values ​​are considered to be those exceeding the normal air conditioning range [16,30]; N is the total number of samples. The lower the value of η, the more accurate the prediction.

[0161] For any outliers predicted, they are first removed. The method for supplementing the values ​​is similar to the power change rate calculation method in step 3. The selected calculation time points are expanded to include data from every other time point, specifically:

[0162]

[0163] Where P1 and P3 are the power of the air conditioner at one time interval, T1 and T3 are the temperature at one time interval, and α nη This is the rate of change of the air conditioner's power at this time. The method for predicting the set temperature afterwards remains the same as before.

[0164] Other specific embodiments of the present invention for evaluating the predicted value of air conditioner set temperature:

[0165] The predicted value T of the air conditioning set temperature is evaluated using standard root mean square error (NRMSE), standard mean absolute error (NMAE), and accuracy (AR). set The evaluation is conducted using the following formula:

[0166]

[0167]

[0168]

[0169] In the formula: y i Set the air conditioner to the actual temperature value; Set a predicted temperature value for the air conditioner; n is the sample size. We consider a difference of less than 1 degree Celsius between the predicted and actual values ​​to be TP (Accurate), and a difference exceeding 1 degree Celsius to be TN (Inaccurate). For NRMSE, the NMAE evaluation method considers a lower value to be more accurate, and a higher AR value to be more accurate.

[0170] A system embodiment applying the method of the present invention:

[0171] An air conditioning flexible adjustment reference temperature estimation system includes a data processing module, a feature processing module, and a reference temperature estimation module;

[0172] The data processing module is used to analyze the original air conditioner power and temperature data, identify outliers and missing data, process outliers, and use regression equations to interpolate missing data to obtain the initial air conditioner power and temperature data.

[0173] The feature processing module is used to perform feature engineering on the initial air conditioning power and temperature data, and to compare the significance of the features of the air conditioning power and temperature data using the Pearson coefficient analysis method. The air conditioning dataset with high significance is selected to form the feature data of air conditioning power and temperature.

[0174] The reference temperature calculation module is used to generate the power characteristic curve of the air conditioner based on the characteristic data of the air conditioner's power and temperature, obtain the coupling relationship between the set temperature difference and the power change, and predict the set temperature of the air conditioner based on the coupling relationship to obtain the calculation result of the reference temperature of the air conditioner, thus completing the calculation of the reference temperature for flexible adjustment of the air conditioner.

[0175] This invention uses normality analysis to analyze and process the raw data, improving the accuracy of the extracted features and simplifying calculations. This invention utilizes the physical characteristics of air conditioners to calculate the reference temperature setting for air conditioner regulation based on the raw air conditioner data, which is both realistic and convenient.

[0176] An embodiment of a device applying the method of the present invention:

[0177] An electronic device comprising:

[0178] One or more processors;

[0179] Storage device for storing one or more programs;

[0180] When the one or more programs are executed by the one or more processors, the one or more processors implement the above-described method for calculating the reference temperature for flexible air conditioning control.

[0181] An embodiment of a computer medium applying the method of the present invention:

[0182] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for calculating a reference temperature for flexible air conditioning control.

[0183] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, and computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0184] This invention is described in terms of flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowcharts and / or block diagrams, as well as combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowcharts and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0185] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0186] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The steps of the function specified in one or more boxes.

[0187] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for calculating the reference temperature for flexible air conditioning control, characterized in that: Includes the following steps: The first step is to obtain the raw air conditioner power and temperature data; The second step involves using a pre-built data processing model to analyze the original air conditioning power and temperature data, identify outliers and missing data, process the outliers, and use the regression equation method to interpolate the missing data to obtain the initial air conditioning power and temperature data. The third step involves using a pre-built feature processing model to perform feature engineering on the initial air conditioning power and temperature data. The Pearson coefficient analysis method is then used to compare the significance of the features of the air conditioning power and temperature data. The air conditioning dataset with high significance is selected to form the feature data of air conditioning power and temperature. The fourth step involves using the characteristic data of air conditioner power and temperature, and employing a pre-built air conditioner set temperature relationship model, to create the air conditioner power characteristic curve and obtain the coupling relationship between the set temperature difference and power change. Based on this coupling relationship, the air conditioner set temperature is predicted to obtain the calculation result of the air conditioner reference temperature, thus completing the calculation of the air conditioner flexible adjustment reference temperature. In the fourth step, the method for obtaining the calculation result of the air conditioner reference temperature from the air conditioner set temperature relationship model is as follows: Step 41: Obtain characteristic data of air conditioner power and temperature, and calculate the reduction rate of air conditioner load power due to the air conditioner cooling set temperature. It is used to characterize the rise in the air conditioning cooling set temperature. The percentage reduction in air conditioning load power is calculated using the following formula: in The set temperature is air conditioning load power at that time The set temperature is Air conditioning load power at that time; Step 42, according to Between Power variation rate of air conditioner set temperature The power change rate at any temperature difference was calculated. Power change rate The formula used to characterize the coupling relationship between the set temperature difference and power change is as follows: in It is the power of the air conditioner at two consecutive moments. It refers to the temperature at two consecutive moments. It is the power change rate of the air conditioner, that is, its adjustment capability at this moment; Step 43, based on the power change rate Calculate the air conditioner set temperature This refers to the reference temperature for flexible air conditioning adjustment, and its calculation formula is as follows: in, It is the theoretical temperature regulation amount. This is the predicted air conditioning set temperature at this time.

2. The method for calculating the reference temperature for flexible air conditioning control as described in claim 1, characterized in that: In the second step, the data processing model obtains the initial air conditioning power and temperature data as follows: Step 21: Obtain k sets of raw air conditioner power and temperature data; Step 22: Perform normal distribution analysis on each set of original air conditioning power and temperature data, which includes the following: For the i-th set of original air conditioning power and temperature data, calculate its mean, standard deviation, skewness, and kurtosis; and plot PP and QQ plots to identify outliers and missing values ​​in the original air conditioning power and temperature data. Outliers are data points that deviate 20% from the diagonal in the QQ graph; Step 23 involves processing outliers and missing values ​​in the original air conditioning power and temperature data, including the following: Missing values ​​are handled using a regression equation method, and the calculation formula is as follows: The missing values ​​are calculated using the obtained regression equation coefficients, and the calculation formula is as follows: ； in The horizontal and vertical axes correspond to the five data points, specifically the air conditioning time and temperature. These are the average values ​​of the x and y coordinates of the five points, respectively. These are the coefficients of the interpolation function; X represents the missing value point. The x-coordinate; Missing value point The supplementary value obtained through interpolation; For abnormal points with short-term temperature anomalies, interpolation is used for replacement; for data with high sampling frequency and small proportion, the entire data record is deleted directly; for abnormal data with periodic jumps, the entire row is deleted.

3. The method for calculating the reference temperature for flexible air conditioning control as described in claim 1, characterized in that: In the third step, the feature processing model obtains the feature data of air conditioner power and temperature as follows: Step 31: Obtain initial air conditioner power and temperature data; Step 32: Based on the initial air conditioning power and temperature data, use the Pearson coefficient analysis method to calculate the Pearson coefficient between the air conditioning power and the set temperature data; Step 33: Based on the Pearson coefficient, compare the significance of the air conditioner power and set temperature data to determine the strength of the correlation. When the absolute value of the Pearson coefficient between the air conditioner power and the set temperature data is greater than 0.4, it indicates that the correlation between the two is strong, that is, the significance is high. Step 34: Select the air conditioner dataset with high significance to form the feature data of air conditioner power and temperature.

4. The method for calculating the reference temperature for flexible air conditioning control as described in claim 3, characterized in that: The Pearson coefficient between air conditioner power and set temperature data is the sample Pearson coefficient, and its calculation formula is as follows: in, These are the characteristic variables of air conditioning. The sample values, and They are the characteristic variables. The average value.

5. The method for calculating the reference temperature for flexible air conditioning control as described in claim 1, characterized in that: It also includes a predictive evaluation model for evaluating the predicted air conditioning set temperature; The predictive evaluation model includes the following: The predicted air conditioning set temperature was calculated using the standard mean absolute error (MAPE). The evaluation is performed using the following formula: In the formula: Set the air conditioner to the actual temperature value; Set the predicted temperature value for the air conditioner; n is the sample size; for the standard mean absolute error (MAPE), a lower value is considered more accurate. The calculated air conditioning set temperature is evaluated based on the abnormal data in the predicted values. The calculation formula is as follows: In the formula: Abnormal data in the predicted values, those exceeding the normal temperature adjustment range of air conditioning [16,30], are considered abnormal; N is the total number of samples; This represents the ratio of outliers; the lower the value, the more accurate the result. For outlier data in the predicted values, they are first deleted and then replaced using interpolation. The selected calculation time point is expanded to include data from one time point in between. The calculation formula is as follows: in It is the power of the air conditioner at a certain time interval. It is the temperature at a time interval. This is the power change rate of the air conditioner at this time.

6. A method for calculating the reference temperature for flexible air conditioning control, characterized in that: Includes the following: Obtain raw air conditioner power and temperature data; Using a pre-built data processing model, the original air conditioning power and temperature data are analyzed to identify outliers and missing data. Outliers are processed, and missing data are filled in by interpolation using regression equations to obtain the initial air conditioning power and temperature data. By using a pre-built feature processing model, feature engineering is performed on the initial air conditioning power and temperature data. The significance of the features of the air conditioning power and temperature data is compared using the Pearson coefficient analysis method. The air conditioning dataset with high significance is selected to form the feature data of air conditioning power and temperature. Based on the characteristic data of air conditioner power and temperature, a power characteristic curve of the air conditioner is generated using a pre-constructed air conditioner set temperature relationship model, obtaining the coupling relationship between set temperature difference and power change; and the air conditioner set temperature is predicted based on the coupling relationship to obtain the estimated result of the air conditioner reference temperature; the method for obtaining the estimated result of the air conditioner reference temperature using the air conditioner set temperature relationship model is as follows: Step 41: Obtain characteristic data of air conditioner power and temperature, and calculate the reduction rate of air conditioner load power due to the air conditioner cooling set temperature. It is used to characterize the rise in the air conditioning cooling set temperature. The percentage reduction in air conditioning load power is calculated using the following formula: in The set temperature is air conditioning load power at that time The set temperature is Air conditioning load power at that time; Step 42, according to Between Power variation rate of air conditioner set temperature The power change rate at any temperature difference was calculated. Power change rate The formula used to characterize the coupling relationship between the set temperature difference and power change is as follows: in It is the power of the air conditioner at two consecutive moments. It refers to the temperature at two consecutive moments. It is the power change rate of the air conditioner, that is, its adjustment capability at this moment; Step 43, based on the power change rate Calculate the air conditioner set temperature This refers to the reference temperature for flexible air conditioning adjustment, and its calculation formula is as follows: in, It is the theoretical temperature regulation amount. This is the predicted air conditioner set temperature at this time; Using a pre-built prediction and evaluation model, the predicted air conditioning set temperature is evaluated, and the reference temperature for flexible air conditioning adjustment is calculated.

7. An air conditioning flexible regulation reference temperature calculation system, employing the air conditioning flexible regulation reference temperature calculation method according to any one of claims 1-6, characterized in that: It includes a data processing module, a feature processing module, a baseline temperature calculation module, and a prediction and evaluation module; The data processing module is used to analyze the original air conditioner power and temperature data, identify outliers and missing data, process outliers, and use regression equations to interpolate missing data to obtain the initial air conditioner power and temperature data. The feature processing module is used to perform feature engineering on the initial air conditioning power and temperature data, and to compare the significance of the features of the air conditioning power and temperature data using the Pearson coefficient analysis method. The air conditioning dataset with high significance is selected to form the feature data of air conditioning power and temperature. The reference temperature estimation module is used to generate a power characteristic curve of the air conditioner based on the characteristic data of the air conditioner's power and temperature, obtain the coupling relationship between the set temperature difference and power change, and predict the set temperature of the air conditioner based on the coupling relationship to obtain the estimation result of the air conditioner's reference temperature. The prediction and evaluation module is used to evaluate the predicted air conditioning set temperature using a pre-built prediction and evaluation model, and to calculate the reference temperature for flexible air conditioning adjustment.

8. An electronic device, characterized in that: It includes: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the air conditioning flexible adjustment reference temperature calculation method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: It stores a computer program that, when executed by a processor, implements a method for calculating the reference temperature for flexible air conditioning as described in any one of claims 1-6.

Images