An airport energy consumption monitoring and management system and method

Abstract

The present application discloses an airport energy consumption monitoring and management system and method, relating to the technical field of energy consumption monitoring and management. The method includes training a first energy consumption prediction model when the energy consumption data one, two, and three of the first type of medium are all the same; training a second energy consumption prediction model when the energy consumption data one, two, and three of the first type of medium are all different from each other and there are two that are the same among the energy consumption data one, two, and three of the second type of medium; training a third energy consumption prediction model when there are two that are the same among the energy consumption data one, two, and three of the first type of medium and there are at least two that are the same among the energy consumption data one, two, and three of the second type of medium; determining the relationship of the current energy consumption data of the three same media, so as to select the corresponding model from the first, second, and third energy consumption prediction models for energy consumption prediction, and finally performing distribution adjustment control of airport energy. The airport energy consumption monitoring and management system and method provided by the present invention can improve the efficiency of energy consumption monitoring.

Description

Technical Field

[0001] This application relates to the technical field of energy consumption monitoring and management, and particularly to an airport energy consumption monitoring and management system and method. Background Art

[0002] Airport energy consumption monitoring and management is of great significance for ensuring energy use safety, improving energy utilization efficiency, reducing operating costs, enhancing management levels, and contributing to sustainable development.

[0003] In the prior art, the detection and analysis of airport energy consumption only consider a single angle and do not conduct detailed analysis under various conditions, resulting in large errors in the detection results of energy consumption and reducing the work efficiency of the entire energy consumption monitoring and management. Summary of the Invention

[0004] To overcome the deficiencies of the above prior art, this application provides an airport energy consumption monitoring and management system and method.

[0005] In a first aspect, an airport energy consumption monitoring and management method provided by this application includes:

[0006] Obtain three target detection areas that require energy consumption detection, and respectively conduct energy consumption data statistics on the same media and different media in the energy-consuming media in the three target detection areas during the most recent historical period to obtain a first type of energy consumption dataset and a second type of energy consumption dataset;

[0007] If the first type of media energy consumption data one, the first type of media energy consumption data two, and the first type of media energy consumption data three in the first type of energy consumption dataset are all the same, then perform corresponding correlation extraction on the first type of energy consumption dataset and the second type of energy consumption dataset to train a first energy consumption prediction model;

[0008] If the first type of media energy consumption data one, the first type of media energy consumption data two, and the first type of media energy consumption data three in the first type of energy consumption dataset are all different from each other, and there are two that are the same among the second type of media energy consumption data one, the second type of media energy consumption data two, and the second type of media energy consumption data three in the second type of energy consumption dataset, then combine the corresponding correlation extraction and difference comparison of the first type of energy consumption dataset and the second type of energy consumption dataset to train a second energy consumption prediction model;

[0009] If there are two that are the same among the first type of media energy consumption data one, the first type of media energy consumption data two, and the first type of media energy consumption data three in the first type of energy consumption dataset, and there are at least two that are the same among the second type of media energy consumption data one, the second type of media energy consumption data two, and the second type of media energy consumption data three in the second type of energy consumption dataset, then count a second difference feature set;

[0010] If two of the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption data set are the same, and the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three in the second type of energy consumption data set are all the same, then the third difference feature set is counted, and according to the second difference feature set or the third difference feature set, the third energy consumption prediction model is trained;

[0011] Obtain the energy consumption data of the same medium at the current detection moment to obtain three current same-medium energy consumption data, and perform a relationship judgment on the three current same-medium energy consumption data, so as to select one corresponding model from the first energy consumption prediction model, the second energy consumption prediction model, and the third energy consumption prediction model for energy consumption prediction, and finally perform the distribution adjustment control of airport energy and output the energy distribution adjustment control result.

[0012] Preferably, obtain three target detection areas that need to be detected for energy consumption, and respectively count all the media that need to consume energy in the three target detection areas in the recent historical period to obtain three types of energy-consuming media sets;

[0013] Extract the common part of the three media from the three types of energy-consuming media sets, output the same media, and respectively count the energy consumption data of the same media in the recent historical period to obtain the same-medium energy consumption data, and the same-medium energy consumption data includes the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the third type of medium energy consumption data three;

[0014] The first type of medium energy consumption data one, the first type of medium energy consumption data two, and the third type of medium energy consumption data three are combined into the first type of energy consumption data set;

[0015] Extract the different part of the remaining three media from the three types of energy-consuming media sets, output the different media, and respectively count the energy consumption data of the same media in the recent historical period to obtain the different-medium energy consumption data, and the different-medium energy consumption data includes the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three;

[0016] The second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three are combined into the second type of energy consumption data set.

[0017] Preferably, if the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three are all the same, then extract the correlation degree between the second type of medium energy consumption data one and the first type of medium energy consumption data one to obtain the correlation degree one;

[0018] Extract the correlation between the second type of medium energy consumption data two and the first type of medium energy consumption data two to obtain correlation two;

[0019] Extract the correlation between the second type of medium energy consumption data three and the first type of medium energy consumption data three to obtain correlation three;

[0020] The correlation one, correlation two, and correlation three are combined into the first correlation feature set;

[0021] Train the first energy consumption prediction model according to the first correlation feature set, the first type of energy consumption data set, and the second type of energy consumption data set.

[0022] Preferably, if the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three are all different from each other, then judge the relationship between the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three, and output the judgment information;

[0023] According to the judgment information, if two of the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three are the same, output one of the judgment results;

[0024] Perform a difference comparison between the two same medium energy consumption data in the one judgment result and the different one medium energy consumption data in the one judgment result to obtain one difference feature;

[0025] According to the one judgment result, perform a difference comparison between the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three to obtain the second difference feature;

[0026] The one difference feature and the second difference feature are combined into the first difference feature set;

[0027] Perform correlation extraction between the two same medium energy consumption data in the one judgment result and the corresponding two medium energy consumption data in the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three respectively to obtain correlation four and correlation five;

[0028] The correlation four and correlation five are combined into the second correlation feature set;

[0029] Train the second energy consumption prediction model according to the first difference feature set, the second correlation feature set, the first type of energy consumption data set, and the second type of energy consumption data set.

[0030] Preferably, if two of the first type of medium energy consumption data 1, the first type of medium energy consumption data 2, and the first type of medium energy consumption data 3 are the same as each other, output the second judgment result, and if two of the second type of medium energy consumption data 1, the second type of medium energy consumption data 2, and the second type of medium energy consumption data 3 are the same as each other, output the third judgment result;

[0031] Perform a difference comparison between the two same medium energy consumption data in the third judgment result to obtain the third difference feature;

[0032] Based on the two same medium energy consumption data in the third judgment result, perform a difference comparison between the corresponding two medium energy consumption data in the second judgment result to obtain the fourth difference feature;

[0033] The third difference feature and the fourth difference feature are combined into the second difference feature set.

[0034] Preferably, if two of the first type of medium energy consumption data 1, the first type of medium energy consumption data 2, and the first type of medium energy consumption data 3 are the same as each other, output the second judgment result, and if the second type of medium energy consumption data 1, the second type of medium energy consumption data 2, and the second type of medium energy consumption data 3 are all the same as each other, output the fourth judgment result;

[0035] Based on the two same medium energy consumption data in the second judgment result, perform a difference comparison between the corresponding two medium energy consumption data in the fourth judgment result and another medium energy consumption data respectively to obtain the fifth difference feature and the sixth difference feature;

[0036] The fifth difference feature and the sixth difference feature are combined into the third difference feature set;

[0037] Train the third energy consumption prediction model using the second difference feature set or the third difference feature set, the first type of energy consumption data set, and the second type of energy consumption data set.

[0038] Preferably, according to the same medium, obtain the energy consumption data of the same medium at the current detection moment to obtain three current energy consumption data of the same medium;

[0039] Judge the relationship between the three current energy consumption data of the same medium. If the three current energy consumption data of the same medium are all the same, input the three current energy consumption data of the same medium and the different medium into the first energy consumption prediction model for testing, and output the energy consumption prediction result;

[0040] If the three current energy consumption data of the same medium are all different, input the three current energy consumption data of the same medium and the different medium into the second energy consumption prediction model for testing, and output the energy consumption prediction result;

[0041] If two of the three current identical medium energy consumption data are the same, then the three current identical medium energy consumption data and the different medium are input into the third energy consumption prediction model for testing, and the energy consumption prediction result is output;

[0042] According to the energy consumption prediction result and the three current identical medium energy consumption data, the allocation adjustment control of airport energy is carried out, and the energy allocation adjustment control result is output.

[0043] In a second aspect, an airport energy consumption monitoring and management system includes:

[0044] A historical energy consumption statistics unit for obtaining three target detection areas that need to be detected for energy consumption, and respectively statistically obtaining a first type of energy consumption data set and a second type of energy consumption data set for the same medium and different media in the media that need to consume energy in the three target detection areas in the most recent historical period;

[0045] A first model training unit for determining that if the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption data set are all the same, then the first type of energy consumption data set and the second type of energy consumption data set are subjected to corresponding correlation extraction to train a first energy consumption prediction model;

[0046] A second model training unit for determining that if the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption data set are all different from each other, and two of the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three in the second type of energy consumption data set are the same, then the first type of energy consumption data set and the second type of energy consumption data set are combined with corresponding correlation extraction and difference comparison to train a second energy consumption prediction model;

[0047] A data feature statistics unit for determining that if two of the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption data set are the same, and at least two of the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three in the second type of energy consumption data set are the same, then a second difference feature set is statistically obtained;

[0048] A third model training unit, configured to determine that if two of the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption data set are the same, and all of the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three in the second type of energy consumption data set are the same, then count the third difference feature set, and train the third energy consumption prediction model according to the second difference feature set or the third difference feature set;

[0049] A current energy consumption data prediction unit, configured to obtain three current same medium energy consumption data by getting the energy consumption data of the same medium at the current detection moment, make a relationship judgment on the three current same medium energy consumption data, so as to select one corresponding model from the first energy consumption prediction model, the second energy consumption prediction model, and the third energy consumption prediction model for energy consumption prediction, and finally perform allocation adjustment control of airport energy and output an energy allocation adjustment control result.

[0050] Compared with the prior art, the present invention has the following features and beneficial effects:

[0051] By performing differential analysis on all energy-consuming media in different areas of the airport, that is, judging and analyzing the relationship between energy consumption data among the same media and different media respectively, and analyzing various different processing methods. That is, if the relationship among the first energy consumption data of the first type of media, the second energy consumption data of the first type of media, and the first energy consumption data of the first type of media in the same media statistics is the same, the correlation degree can be directly extracted between the first energy consumption data of the second type of media, the second energy consumption data of the second type of media, and the second energy consumption data corresponding to the different media, so as to obtain the degree of influence correlation between the energy consumption of the media, which is convenient for subsequent regular change analysis of the energy consumption status of all media in different areas of the airport at the current moment. According to the above characteristic statistics, the first energy consumption prediction model is trained. In order to reduce the large error in the energy consumption prediction results caused by the characteristics of the energy consumption change of the media in various situations, it is necessary to perform diverse and relatively comprehensive analysis. That is, when the first energy consumption data of the first type of media, the second energy consumption data of the first type of media, and the first energy consumption data of the first type of media are all different, judge the relationship among the first energy consumption data of the second type of media, the second energy consumption data of the second type of media, and the second energy consumption data of the second type of media. If there are two media energy consumption data that are the same among them, perform an analysis process different from training the first energy consumption prediction model to train the second energy consumption prediction model. And when the first energy consumption data of the first type of media, the second energy consumption data of the first type of media, and the first energy consumption data of the first type of media are all different, and there are at least two media energy consumption data that are the same among the first energy consumption data of the second type of media, the second energy consumption data of the second type of media, and the second energy consumption data of the second type of media, perform an analysis process different from the above training of the first energy consumption prediction model and the second energy consumption prediction model to train the third energy consumption prediction model. Through the training of the three models, it is possible to avoid misjudgment caused by using a single prediction model in traditional technologies to analyze and process only one energy consumption situation, thereby improving the utilization rate of resources and the efficiency of energy consumption monitoring work. Finally, according to the energy consumption prediction results, the corresponding energy distribution adjustment control of the airport energy is carried out. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a block diagram of the steps of a method for monitoring and managing airport energy consumption mainly embodied in this embodiment.

[0053] Figure 2 is a block diagram of the structure of a system for monitoring and managing airport energy consumption mainly embodied in this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] The present invention will be further described in detail below with reference to the following embodiments.

[0055] Refer to Figure 1 , a method for monitoring and managing airport energy consumption, the method includes the following steps:

[0056] S1. Obtain three target detection areas that need to be detected for energy consumption. For the same medium and different media in the energy-consuming media within the three target detection areas in the most recent historical period, respectively, conduct energy consumption data statistics to obtain the first type of energy consumption dataset and the second type of energy consumption dataset.

[0057] S2. If the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption dataset are all the same, then perform corresponding correlation degree extraction on the first type of energy consumption dataset and the second type of energy consumption dataset to train the first energy consumption prediction model.

[0058] S3. If the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption dataset are all different from each other, and among the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three in the second type of energy consumption dataset, two of them are the same, then perform a combination of corresponding correlation degree extraction and difference comparison on the first type of energy consumption dataset and the second type of energy consumption dataset to train the second energy consumption prediction model.

[0059] S4. If two of the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption dataset are the same as each other, and there are at least two of the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three in the second type of energy consumption dataset that are the same as each other, then count the second difference feature set.

[0060] S5. If two of the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption dataset are the same as each other, and the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three in the second type of energy consumption dataset are all the same, then count the third difference feature set. According to the second difference feature set or the third difference feature set, train the third energy consumption prediction model.

[0061] S6. Obtain the energy consumption data of the same medium at the current detection moment to get three current energy consumption data of the same medium. Perform a relationship judgment on the three current energy consumption data of the same medium to select one corresponding model from the first energy consumption prediction model, the second energy consumption prediction model, and the third energy consumption prediction model for energy consumption prediction, and finally perform allocation adjustment control of airport energy and output the energy allocation adjustment control result.

[0062] Specifically, by differentiating and analyzing all energy-consuming media in different areas of the airport, that is, judging and analyzing the relationship between energy consumption data among the same media and different media respectively, and analyzing various different processing methods. That is, if the relationship among the first type of energy consumption data one, the first type of energy consumption data two, and the first type of energy consumption data of the same media statistics is the same, the correlation degree extraction of the second type of energy consumption data one, the second type of energy consumption data two, and the second type of energy consumption data corresponding to the different media can be directly carried out to obtain the influence correlation degree among the energy consumption of the media, so as to facilitate the subsequent regular change analysis of the energy consumption status of all media in different areas of the airport at the current moment. According to the above characteristic statistics, the first energy consumption prediction model is trained. In order to reduce the large error in the energy consumption prediction results caused by the characteristics of the energy consumption change of the media in various situations, a diverse and more comprehensive analysis is required. That is, when the first type of energy consumption data one, the first type of energy consumption data two, and the first type of energy consumption data are all different, judge the relationship among the second type of energy consumption data one, the second type of energy consumption data two, and the second type of energy consumption data. If there are two types of energy consumption data that are the same among them, an analysis process different from training the first energy consumption prediction model is carried out to train the second energy consumption prediction model. And when the first type of energy consumption data one, the first type of energy consumption data two, and the first type of energy consumption data are all different, and there are at least two types of energy consumption data that are the same among the second type of energy consumption data one, the second type of energy consumption data two, and the second type of energy consumption data, an analysis process different from the above training of the first energy consumption prediction model and the second energy consumption prediction model is carried out to train the third energy consumption prediction model. Through the training of the three models, it is possible to avoid misjudgment caused by using a single prediction model in traditional technology to analyze and process only one type of energy consumption situation, thereby improving the utilization rate of resources and the efficiency of energy consumption monitoring work. Finally, according to the energy consumption prediction results, the corresponding energy distribution adjustment control of the airport energy is carried out.

[0063] Specifically, step S1 includes the following sub-steps:

[0064] Obtain three target detection areas that need to be detected for energy consumption, and respectively count all media that need to consume energy in the three target detection areas in the recent historical period to obtain three types of energy-consuming media sets.

[0065] Extract the common part of the three types of energy-consuming media sets, output the same media, and respectively count the energy consumption data of the same media in the recent historical period to obtain the same media energy consumption data. The same media energy consumption data includes the first type of energy consumption data one, the first type of energy consumption data two, and the third type of energy consumption data three.

[0066] The energy consumption data of the first type of medium, the energy consumption data of the second type of medium, and the energy consumption data of the third type of medium are combined into the first energy consumption dataset.

[0067] Extract the part of the medium that is different from the remaining three from the three energy-consuming medium sets, output the different medium, and respectively count the energy consumption data of the same medium in the recent historical period to obtain the energy consumption data of the different medium. The energy consumption data of the different medium includes the energy consumption data of the second type of medium one, the energy consumption data of the second type of medium two, and the energy consumption data of the second type of medium three.

[0068] The energy consumption data of the second type of medium one, the energy consumption data of the second type of medium two, and the energy consumption data of the second type of medium three are combined into the second energy consumption dataset.

[0069] Specifically, for example, there are three target detection areas (if they are area A, area B, and area C respectively, such as the terminal building, the runway and apron, the maintenance area, the freight area, etc.), and three energy-consuming medium sets (for example, all the energy-consuming media in area A are a1 and a2; all the energy-consuming media in area B are b1 and b2; all the energy-consuming media in area C are c1 and c2). It should be noted that the energy-consuming media include electricity, natural gas or steam, fuel oil, water, and renewable energy. The same medium (if it is a1, b1, c3, for example, all are electricity), the energy consumption data of the first type of medium one, the energy consumption data of the first type of medium two, and the energy consumption data of the third type of medium three (that is, the degree of energy consumption, for example, 60%, 40%, 30% respectively, or the values among them), the different medium (that is, a2; b2; c2 are different categories from each other), the energy consumption data of the second type of medium one, the energy consumption data of the second type of medium two, and the energy consumption data of the second type of medium three (if they are 60%, 40%, 30% respectively, or the values among them).

[0070] Specifically, step S2 includes the following sub-steps:

[0071] If the energy consumption data of the first type of medium one, the energy consumption data of the first type of medium two, and the energy consumption data of the first type of medium three are all the same, then extract the correlation degree between the energy consumption data of the second type of medium one and the energy consumption data of the first type of medium one to obtain the first correlation degree;

[0072] Extract the correlation degree between the energy consumption data of the second type of medium two and the energy consumption data of the first type of medium two to obtain the second correlation degree;

[0073] Extract the correlation degree between the energy consumption data of the second type of medium three and the energy consumption data of the first type of medium three to obtain the third correlation degree;

[0074] The first correlation degree, the second correlation degree, and the third correlation degree are combined into the first correlation degree feature set.

[0075] Train the first energy consumption prediction model according to the first correlation degree feature set, the first energy consumption dataset, and the second energy consumption dataset.

[0076] Specifically, if the energy consumption data of the first type of medium 1, the energy consumption data of the first type of medium 2, and the energy consumption data of the first type of medium 3 are all the same (if all are 60% of power consumption), the correlation degree 1 (for example, a2 is fuel and the energy consumption is 40%, then calculate the ratio between the two, and calculate the ratios between the two in other historical periods, and perform an averaging process on the multiple ratio data obtained by statistics to obtain an average ratio, which is the correlation degree 1, if it is 1 / 3), the correlation degree 2 and the correlation degree 3 (the same as the explanation of the correlation degree 1, not elaborated here, and they are processed for b2 and c2 respectively), the first energy consumption prediction model (that is, according to the energy consumption data records statistically in historical periods, and the statistics of the correlation degree of energy consumption data between each medium, to train a machine learning model. For example, intuitively explained: x, y, k, where x is the energy consumption data, k is the correlation degree feature, and y is the first energy consumption prediction model. According to the multiple energy consumption data and multiple correlation degree features obtained by statistics in multiple historical periods, a linear function equation about x, y, and k can be trained).

[0077] Specifically, step S3 includes the following sub-steps:

[0078] If the energy consumption data of the first type of medium 1, the energy consumption data of the first type of medium 2, and the energy consumption data of the first type of medium 3 are all different from each other, then judge the relationship between the energy consumption data of the second type of medium 1, the energy consumption data of the second type of medium 2, and the energy consumption data of the second type of medium 3, and output the judgment information.

[0079] According to the judgment information, if two of the energy consumption data of the second type of medium 1, the energy consumption data of the second type of medium 2, and the energy consumption data of the second type of medium 3 are the same, output one of the judgment results.

[0080] Perform a difference comparison between the two same medium energy consumption data in one of the judgment results and the different one medium energy consumption data in one of the judgment results to obtain one difference feature.

[0081] According to one of the judgment results, perform a difference comparison between the energy consumption data of the first type of medium 1, the energy consumption data of the first type of medium 2, and the energy consumption data of the first type of medium 3 to obtain the second difference feature.

[0082] One difference feature and the second difference feature are combined into the first difference feature set.

[0083] Perform a correlation degree extraction between the two same medium energy consumption data in one of the judgment results and the corresponding two medium energy consumption data in the energy consumption data of the first type of medium 1, the energy consumption data of the first type of medium 2, and the energy consumption data of the first type of medium 3 respectively to obtain the correlation degree 4 and the correlation degree 5.

[0084] The correlation degree 4 and the correlation degree 5 are combined into the second correlation degree feature set.

[0085] Train a second energy consumption prediction model based on the first difference feature set, the second correlation feature set, the first type of energy consumption data set, and the second type of energy consumption data set.

[0086] Specifically, if the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three are all different from each other (if they are 60%, 40%, and 30% respectively), and if two of the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three are the same (if they are 20%, 20%, and 40% respectively), one difference feature (20%, 20% respectively), the other difference feature (60% - 30% = 30%, 40% - 30% = 10% respectively), correlation degree four (that is, if the a2 and b2 energy consumption data are the same, then the corresponding a1 and b1 are each processed in the steps of the correlation degree, and the processing steps are the same as the explanation of the above correlation degree one, and will not be elaborated here. If they are 1 / 4 and 2 / 3 respectively), train a second energy consumption prediction model based on the first difference feature set, the second correlation feature set, the first type of energy consumption data set, and the second type of energy consumption data set (the explanation is the same as the above first energy consumption prediction model, and will not be elaborated here).

[0087] Specifically, step S4 includes the following sub-steps:

[0088] If two of the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three are the same, output the second judgment result, and if two of the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three are the same, output the third judgment result.

[0089] Perform a difference comparison between the two same medium energy consumption data in the third judgment result to obtain the third difference feature.

[0090] Based on the two same medium energy consumption data in the third judgment result, perform a difference comparison between the corresponding two medium energy consumption data in the second judgment result to obtain the fourth difference feature.

[0091] The third difference feature and the fourth difference feature are combined into the second difference feature set.

[0092] Specifically, if two of the first - type medium energy - consumption data 1, first - type medium energy - consumption data 2, and first - type medium energy - consumption data 3 are the same as each other (if they are 50%, 30%, and 50% respectively), and if two of the second - type medium energy - consumption data 1, second - type medium energy - consumption data 2, and second - type medium energy - consumption data 3 are the same as each other (if they are 30%, 30%, and 40% respectively), their three - difference features (i.e., 10%, 10% respectively, that is, the difference comparison between a2 and c2, and the difference comparison between b2 and c2), and their four - difference features (according to the above - mentioned difference comparison between a2 and c2, and the difference comparison between b2 and c2, then correspondingly perform the difference comparison between a1 and c1, and correspondingly perform the difference comparison between b1 and c1, that is, 0% and 20% respectively).

[0093] Specifically, step S5 includes the following sub - steps:

[0094] If two of the first - type medium energy - consumption data 1, first - type medium energy - consumption data 2, and first - type medium energy - consumption data 3 are the same as each other, output its second judgment result, and if the second - type medium energy - consumption data 1, second - type medium energy - consumption data 2, and second - type medium energy - consumption data 3 are all the same, output its fourth judgment result.

[0095] According to the two same medium energy - consumption data in its second judgment result, perform the difference comparison between the corresponding two medium energy - consumption data in its fourth judgment result and another medium energy - consumption data respectively to obtain its fifth - difference feature and its sixth - difference feature.

[0096] Its fifth - difference feature and its sixth - difference feature are combined into the third - difference feature set.

[0097] Train the third energy - consumption prediction model with the second - difference feature set or the third - difference feature set, the first - type energy - consumption data set, and the second - type energy - consumption data set.

[0098] Specifically, if two of the first - type medium energy - consumption data 1, first - type medium energy - consumption data 2, and first - type medium energy - consumption data 3 are the same as each other (if they are 50%, 30%, and 50% respectively), and if the second - type medium energy - consumption data 1, second - type medium energy - consumption data 2, and second - type medium energy - consumption data 3 are all the same (all 40%), its fifth - difference feature (that is, correspondingly perform the difference comparison between a2 and b2, which is 0%) and its sixth - difference feature (that is, correspondingly perform the difference comparison between c2 and b2, which is 10%), train the third energy - consumption prediction model with the second - difference feature set or the third - difference feature set, the first - type energy - consumption data set, and the second - type energy - consumption data set (the explanation is the same as that of the above - mentioned first energy - consumption prediction model, and no more explanation is given here).

[0099] Specifically, step S6 includes the following sub - steps:

[0100] Based on the same medium, obtain the energy consumption data of each of the same media at the current detection moment to obtain the energy consumption data of the same media, and obtain three pieces of current energy consumption data of the same media.

[0101] Judge the relationship among the three pieces of current energy consumption data of the same media. If the three pieces of current energy consumption data of the same media are all the same, then input the three pieces of current energy consumption data of the same media and the different media into the first energy consumption prediction model for testing, and output the energy consumption prediction result.

[0102] If the three pieces of current energy consumption data of the same media are all different, then input the three pieces of current energy consumption data of the same media and the different media into the second energy consumption prediction model for testing, and output the energy consumption prediction result.

[0103] If there are two pieces of current energy consumption data of the same media among the three pieces of current energy consumption data of the same media that are the same, then input the three pieces of current energy consumption data of the same media and the different media into the third energy consumption prediction model for testing, and output the energy consumption prediction result.

[0104] According to the energy consumption prediction result and the three pieces of current energy consumption data of the same media, perform allocation adjustment control of airport energy, and output the energy allocation adjustment control result.

[0105] Specifically, for example, there are three current energy consumption data of the same medium (the energy consumption data of a1, b1, and c1 respectively, that is, the initially preliminarily monitored energy consumption data, so as to conduct more accurate energy consumption prediction based on the energy consumption data of the same medium monitored, in order to further adjust the distribution amount of the originally preset energy consumption in the remaining medium, improve the energy utilization rate and energy conservation. If they are 50%, 50%, and 50% respectively), if the three current energy consumption data of the same medium are not all the same, then the three current energy consumption data of the same medium and the different medium are input into the second energy consumption prediction model for testing (then the energy consumption of the remaining medium is predicted through the third energy consumption prediction model. If the energy consumption prediction results are 30%, 20%, and 35%), if the three current energy consumption data of the same medium are not all the same (if they are 50%, 40%, and 60% respectively, then the energy consumption of the remaining medium is predicted through the second energy consumption prediction model), if there are two of the three current energy consumption data of the same medium that are the same (if they are 40%, 40%, and 60% respectively, then the energy consumption of the remaining medium is predicted through the third energy consumption prediction model). Taking the case where if the three current energy consumption data of the same medium are not all the same, then the three current energy consumption data of the same medium and the different medium are input into the second energy consumption prediction model for testing (then the energy consumption of the remaining medium is predicted through the third energy consumption prediction model. If the energy consumption prediction results are 30%, 20%, and 35%) as an example, if the originally preset power energy distribution amounts are 50%, 50%, and 50% respectively, and the fuel energy distribution amounts are 50%, 30%, and 20% respectively, then according to the energy consumption prediction results: 30%, 20%, and 35%, the fuel energy distribution amount in area A is reduced to 30%, the fuel energy distribution amount in area B is reduced to 20%, and the fuel energy distribution amount in area C is increased to 35%, so as to adjust and control the reasonable distribution of airport energy and improve energy utilization.

[0106] An airport energy consumption monitoring and management system, by applying an airport energy consumption monitoring and management method as described above, includes a historical energy consumption statistics unit, a first model training unit, a second model training unit, a third model training unit, and a current energy consumption data prediction unit, referring to Figure 2, obtain three target detection areas that require energy consumption detection through the historical energy consumption statistics unit, and respectively conduct energy consumption data statistics on the same medium and different media in the required energy-consuming media in the three target detection areas during the most recent historical period to obtain a first type of energy consumption dataset and a second type of energy consumption dataset; judge through the first model training unit that if the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption dataset are all the same, then extract the corresponding correlation degree of the first type of energy consumption dataset and the second type of energy consumption dataset to train the first energy consumption prediction model; judge through the second model training unit that if the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption dataset are all different from each other, and there are two of the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three in the second type of energy consumption dataset that are the same, then combine the corresponding correlation degree extraction and difference comparison of the first type of energy consumption dataset and the second type of energy consumption dataset to train the second energy consumption prediction model; judge through the data feature statistics unit that if there are two of the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption dataset that are the same, and there are at least two of the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three in the second type of energy consumption dataset that are the same, then count the second difference feature set; judge through the third model training unit that if there are two of the first type of medium energy consumption data one, the first type of medium energy consumption data two, and the first type of medium energy consumption data three in the first type of energy consumption dataset that are the same, and the second type of medium energy consumption data one, the second type of medium energy consumption data two, and the second type of medium energy consumption data three in the second type of energy consumption dataset are all the same, then count the third difference feature set, and train the third energy consumption prediction model according to the second difference feature set or the third difference feature set; obtain the energy consumption data of the same medium at the current detection moment through the current energy consumption data prediction unit to obtain three current energy consumption data of the same medium, judge the relationship of the three current energy consumption data of the same medium, so as to select one corresponding model from the first energy consumption prediction model, the second energy consumption prediction model, and the third energy consumption prediction model for energy consumption prediction, and finally conduct distribution adjustment control of airport energy and output the energy distribution adjustment control result.

[0107] The above are all the preferred embodiments of this application, and the protection scope of this application is not limited by this. Therefore, all equivalent changes made according to the structure, shape, and principle of this application should be covered by the protection scope of this application.

Claims

1. A method for monitoring and managing airport energy consumption, characterized in that: include: Obtain three target detection areas for energy consumption detection, and perform energy consumption statistics on the same media and different media in the three target detection areas in the recent historical period to obtain the first type of energy consumption data set and the second type of energy consumption data set; If the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 in the first type of energy consumption data set are the same, extracting the corresponding correlation between the first type of energy consumption data set and the second type of energy consumption data set to train a first energy consumption prediction model; If the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 in the first type of energy consumption data set are different from each other, and two of the second type of medium energy consumption data 1, the second type of medium energy consumption data 2 and the second type of medium energy consumption data 3 in the second type of energy consumption data set are the same, then the first type of energy consumption data set and the second type of energy consumption data set are combined with features of corresponding correlation extraction and difference comparison to train a second energy consumption prediction model; If two of the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 in the first type of energy consumption data set are identical to each other, and at least two of the second type of medium energy consumption data 1, the second type of medium energy consumption data 2 and the second type of medium energy consumption data 3 in the second type of energy consumption data set are identical to each other, then a second difference feature set is calculated; If two of the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 in the first type of energy consumption data set are the same, and the second type of medium energy consumption data 1, the second type of medium energy consumption data 2 and the second type of medium energy consumption data 3 in the second type of energy consumption data set are the same, then a third difference feature set is statistically calculated, and a third energy consumption prediction model is trained according to the second difference feature set or the third difference feature set; Obtain the energy consumption data of the same medium at the current detection moment to obtain three current energy consumption data of the same medium, make a relationship judgment on the three current energy consumption data of the same medium, and select one of the corresponding models from the first energy consumption prediction model, the second energy consumption prediction model and the third energy consumption prediction model to perform energy consumption prediction, and finally perform airport energy distribution adjustment control, and output the energy distribution adjustment control result.

2. The airport energy consumption monitoring and management method according to claim 1 is characterized in that: The steps of obtaining three target detection areas for energy consumption detection, and performing energy consumption data statistics on the same media and the different media in the three target detection areas in the recent historical period to obtain the first type of energy consumption data set and the second type of energy consumption data set are specifically as follows: Obtain three target detection areas for energy consumption detection, and count all media that need to consume energy in the three target detection areas in the recent historical period to obtain three types of energy-consuming medium sets; Extracting the same part of the three types of energy-consuming media from the three types of energy-consuming media, outputting the same media, and respectively counting the energy consumption data of the same media in the recent historical period to obtain the same medium energy consumption data, the same medium energy consumption data including the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the third type of medium energy consumption data 3; The first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the third type of medium energy consumption data 3 are combined into a first type of energy consumption data set; Extract the remaining three different media from the three types of energy consumption media, output the difference media, and respectively count the energy consumption data of the same media in the recent historical period to obtain the difference medium energy consumption data, the difference medium energy consumption data includes the second type medium energy consumption data 1, the second type medium energy consumption data 2, and the second type medium energy consumption data 3; The second-category medium energy consumption data one, the second-category medium energy consumption data two and the second-category medium energy consumption data three are combined into a second-category energy consumption data set.

3. The airport energy consumption monitoring and management method according to claim 2 is characterized in that: If the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 in the first type of energy consumption data set are the same, the first type of energy consumption data set and the second type of energy consumption data set are subjected to corresponding correlation extraction to train a first energy consumption prediction model, specifically comprising: If the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 are all the same, extracting the correlation between the second type of medium energy consumption data 1 and the first type of medium energy consumption data 1 to obtain correlation degree 1; Extracting the correlation between the second type of medium energy consumption data 2 and the first type of medium energy consumption data 2 to obtain correlation degree 2; Extracting the correlation between the second-category medium energy consumption data three and the first-category medium energy consumption data three to obtain correlation degree three; The first correlation degree, the second correlation degree and the third correlation degree are combined into a first correlation degree feature set; A first energy consumption prediction model is trained according to the first correlation feature set, the first type of energy consumption data set, and the second type of energy consumption data set.

4. The airport energy consumption monitoring and management method according to claim 2 is characterized in that: If the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 in the first type of energy consumption data set are different from each other, and two of the second type of medium energy consumption data 1, the second type of medium energy consumption data 2 and the second type of medium energy consumption data 3 in the second type of energy consumption data set are the same, then the first type of energy consumption data set and the second type of energy consumption data set are subjected to corresponding correlation extraction and difference comparison combined features to train the second energy consumption prediction model, specifically, the following steps: If the first type medium energy consumption data 1, the first type medium energy consumption data 2 and the first type medium energy consumption data 3 are all different from each other, then the relationship between the second type medium energy consumption data 1, the second type medium energy consumption data 2 and the second type medium energy consumption data 3 is determined, and the determination information is output; According to the judgment information, if two of the second-category medium energy consumption data 1, the second-category medium energy consumption data 2 and the second-category medium energy consumption data 3 are the same, outputting one of the judgment results; Comparing the two identical medium energy consumption data in the first judgment result with the one different medium energy consumption data in the second judgment result to obtain a difference feature; According to the first judgment result, the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 are compared with each other to obtain the second difference feature; The first difference feature and the second difference feature are combined into a first difference feature set; The same two medium energy consumption data in the first judgment result are respectively corresponded to the corresponding two medium energy consumption data in the first type medium energy consumption data one, the first type medium energy consumption data two and the first type medium energy consumption data three to extract the correlation between them to obtain correlation degree four and correlation degree five; The association degree 4 and the association degree 5 are combined into a second association degree feature set; A second energy consumption prediction model is trained according to the first difference feature set, the second correlation feature set, the first type of energy consumption data set, and the second type of energy consumption data set.

5. The airport energy consumption monitoring and management method according to claim 2 is characterized in that: If two of the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 in the first type of energy consumption data set are identical to each other, and at least two of the second type of medium energy consumption data 1, the second type of medium energy consumption data 2 and the second type of medium energy consumption data 3 in the second type of energy consumption data set are identical to each other, then the step of calculating the second difference feature set is specifically as follows: If two of the first-category medium energy consumption data 1, the first-category medium energy consumption data 2 and the first-category medium energy consumption data 3 are the same, output the second judgment result; and if two of the second-category medium energy consumption data 1, the second-category medium energy consumption data 2 and the second-category medium energy consumption data 3 are the same, output the third judgment result; Comparing the energy consumption data of the two same media in the third judgment result to obtain the third difference feature; According to the same two medium energy consumption data in the third judgment result, the difference between the corresponding two medium energy consumption data in the second judgment result is compared to obtain the fourth difference feature; The third difference feature and the fourth difference feature are combined into a second difference feature set.

6. The airport energy consumption monitoring and management method according to claim 5 is characterized in that: If two of the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 in the first type of energy consumption data set are identical to each other, and the second type of medium energy consumption data 1, the second type of medium energy consumption data 2 and the second type of medium energy consumption data 3 in the second type of energy consumption data set are identical to each other, then a third difference feature set is statistically obtained, and the steps of training a third energy consumption prediction model according to the second difference feature set or the third difference feature set are specifically as follows: If two of the first-category medium energy consumption data 1, the first-category medium energy consumption data 2 and the first-category medium energy consumption data 3 are the same, output the second judgment result; and if the second-category medium energy consumption data 1, the second-category medium energy consumption data 2 and the second-category medium energy consumption data 3 are all the same, output the fourth judgment result; According to the same two medium energy consumption data in the second judgment result, the corresponding two medium energy consumption data in the fourth judgment result are respectively compared with another medium energy consumption data to obtain the fifth difference feature and the sixth difference feature; The fifth difference feature and the sixth difference feature are combined into a third difference feature set; The second difference feature set or the third difference feature set, the first type of energy consumption data set and the second type of energy consumption data set are used to train a third energy consumption prediction model.

7. The airport energy consumption monitoring and management method according to claim 6 is characterized in that: The steps of obtaining the energy consumption data of the same medium at the current detection moment, obtaining the energy consumption data of the same medium, obtaining three current energy consumption data of the same medium, performing relationship judgment on the three current energy consumption data of the same medium, selecting one of the corresponding models from the first energy consumption prediction model, the second energy consumption prediction model and the third energy consumption prediction model for energy consumption prediction, and finally performing the allocation adjustment control of the airport energy, and outputting the energy allocation adjustment control result, are specifically as follows: According to the same medium, the energy consumption data of each of the same mediums at the current detection moment is obtained to obtain the same medium energy consumption data to obtain three types of current same medium energy consumption data; Determine the relationship between the three current energy consumption data of the same medium. If the three current energy consumption data of the same medium are the same, input the three current energy consumption data of the same medium and the difference medium into the first energy consumption prediction model for testing, and output the energy consumption prediction result; If the three current energy consumption data of the same medium are all different, the three current energy consumption data of the same medium and the difference medium are input into the second energy consumption prediction model for testing, and the energy consumption prediction result is output; If two of the three current energy consumption data of the same medium are the same, the three current energy consumption data of the same medium and the difference medium are input into the third energy consumption prediction model for testing, and the energy consumption prediction result is output; Based on the energy consumption prediction results and the energy consumption data of three current identical media, the airport energy distribution adjustment control is performed, and the energy distribution adjustment control results are output.

8. An airport energy consumption monitoring and management system, characterized in that: The system is used to implement an airport energy consumption monitoring and management method as described in any one of claims 1 to 7, comprising: The historical energy consumption statistics unit is used to obtain three target detection areas for energy consumption detection, and to perform energy consumption statistics on the same media and the different media in the three target detection areas in the recent historical period to obtain the first type of energy consumption data set and the second type of energy consumption data set; The first model training unit is used to determine if the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 in the first type of energy consumption data set are the same, then extract the corresponding correlation between the first type of energy consumption data set and the second type of energy consumption data set to train a first energy consumption prediction model; The second model training unit is used to determine that if the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 in the first type of energy consumption data set are different from each other, and two of the second type of medium energy consumption data 1, the second type of medium energy consumption data 2 and the second type of medium energy consumption data 3 in the second type of energy consumption data set are the same, then the first type of energy consumption data set and the second type of energy consumption data set are subjected to corresponding correlation extraction and difference comparison combined features to train a second energy consumption prediction model; a data feature statistical unit, for determining that if two of the first-category medium energy consumption data 1, the first-category medium energy consumption data 2, and the first-category medium energy consumption data 3 in the first-category energy consumption data set are identical to each other, and at least two of the second-category medium energy consumption data 1, the second-category medium energy consumption data 2, and the second-category medium energy consumption data 3 in the second-category energy consumption data set are identical to each other, then a second difference feature set is statistically generated; The third model training unit is used to determine if two of the first type of medium energy consumption data 1, the first type of medium energy consumption data 2 and the first type of medium energy consumption data 3 in the first type of energy consumption data set are the same, and the second type of medium energy consumption data 1, the second type of medium energy consumption data 2 and the second type of medium energy consumption data 3 in the second type of energy consumption data set are the same, then a third difference feature set is statistically calculated, and a third energy consumption prediction model is trained according to the second difference feature set or the third difference feature set; The energy consumption prediction and energy control unit is used to obtain the energy consumption data of the same medium at the current detection moment to obtain the energy consumption data of the same medium to obtain three current energy consumption data of the same medium, make a relationship judgment on the three current energy consumption data of the same medium, and select one of the corresponding models from the first energy consumption prediction model, the second energy consumption prediction model and the third energy consumption prediction model to perform energy consumption prediction, and finally perform the distribution adjustment control of the airport energy, and output the energy distribution adjustment control result.

Images