Urban Rail Engineering Building Energy Saving Design Method and System Based on Deep Learning

Abstract

The energy-saving design method and system for urban rail engineering buildings based on deep learning proposed in this application take multiple entrances and exits corresponding to each urban rail station as analysis targets, generate multiple first building portraits according to multiple first buildings corresponding to each first entrance and exit, and thus obtain multiple first entrance and exit portraits based on the multiple first building portraits. Combine the historical travel patterns of the crowd at the first entrance and exit to obtain the first lighting control strategy. In addition, based on the multiple first entrance and exit portraits and the comparison of the clothing of passengers when getting on and off the train, obtain the first air-conditioning control strategy, and thus integrate the first lighting control strategy and the first air-conditioning control strategy in the dimension of time intervals to obtain the first urban rail energy-saving adjustment strategy. Through the technical solution of this application, the lighting strategy and air-conditioning operation strategy in the urban rail station can be made more in line with the actual needs of users, thereby effectively reducing the energy consumption of the urban rail station.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy-saving control, and particularly relates to an energy-saving design method and system for urban rail transit engineering buildings based on deep learning. Background Art

[0002] In the field of urban rail transit engineering buildings, lighting equipment and air-conditioning equipment are the main factors of energy consumption. At present, during the operation stage of urban rail transit, the opening and closing of lighting equipment and the operating state of air-conditioning equipment are fixed and there is a certain mismatch with the actual lighting demand and the demand for the opening and closing of air-conditioning. If the operation of lighting and air-conditioning equipment can be adjusted adaptively according to the characteristics and laws of people's travel, it can not only meet the actual needs but also achieve the technical effect of energy conservation and consumption reduction. Summary of the Invention

[0003] The purpose of the present invention is to provide an energy-saving design method and system for urban rail transit engineering buildings based on deep learning, so as to solve the technical problem that the lighting and air-conditioning equipment in urban rail transit stations cannot be adjusted adaptively in the prior art.

[0004] An energy-saving design method for urban rail transit engineering buildings based on deep learning, characterized in that the method includes:

[0005] S1: According to the first corresponding relationship between the buildings around the first urban rail transit station and multiple first entrances and exits, obtain multiple first entrance and exit portraits corresponding to the first urban rail transit station;

[0006] Wherein, the first entrance and exit portrait is used to characterize the information about the characteristics of the travel crowd corresponding to the first entrance and exit;

[0007] S2: Determine the first lighting control strategy corresponding to each first entrance and exit according to the first historical crowd travel information and multiple first entrance and exit portraits;

[0008] S3: Integrate multiple first entrance and exit portraits to obtain a first station travel portrait;

[0009] S4: Identify the first clothing state information and the second clothing state information of multiple first passengers within a second preset time interval, and obtain multiple first comparison information according to the multiple first clothing state information and the multiple second clothing state information;

[0010] Wherein, the first clothing state information refers to the clothing information of the first passenger when waiting for the train, and the second clothing state information refers to the clothing information of the first passenger after getting on the train;

[0011] S5: Determine the first air-conditioning control strategy according to the multiple first comparison information and the first station travel portrait;

[0012] S6: Obtain the first urban rail energy-saving adjustment strategy according to the first lighting control strategy and the first air-conditioning control strategy.

[0013] Preferably, the S1 includes the following sub-steps:

[0014] S11: Obtain multiple first entrances and exits corresponding to the first urban rail station;

[0015] S12: Taking each of the first entrances and exits as a reference, obtain multiple first buildings within the first preset distance range according to the multiple first corresponding relationships;

[0016] S13: Determine multiple first building portraits according to multiple first historical behavior data corresponding to each of the first buildings;

[0017] S14: Generate the first entrance and exit portrait according to the multiple first building portraits.

[0018] Preferably, the S13 includes the following sub-steps:

[0019] S131: For each of the first buildings, obtain multiple first historical behavior data within the first preset time interval;

[0020] S132: Determine the first building portrait according to the multiple first historical behavior data and the first travel pattern data.

[0021] Preferably, the S2 includes the following sub-steps:

[0022] S21: Obtain the first historical population travel information of each of the first entrances and exits within the first preset time interval;

[0023] S22: Establish a first mapping relationship between the first building portrait and the first historical population travel information corresponding to each of the first entrances and exits, so as to obtain multiple first mapping relationships;

[0024] S23: Input each of the first mapping relationships into the first population travel pattern prediction model one by one to obtain multiple first population travel pattern information;

[0025] S24: Determine the first lighting control strategy according to the first population travel pattern information.

[0026] Preferably, the S3 includes the following sub-steps:

[0027] S31: Extract multiple first entrance and exit label information from the multiple first entrance and exit portraits;

[0028] S32: Obtain the first label union of the multiple first entrance and exit label information;

[0029] S33: Obtain the travel portrait of the first site according to the union of the first tags.

[0030] Preferably, the S4 includes the following sub-steps:

[0031] S41: Obtain the first clothing status information and the second clothing status information of each of the first passengers within a second preset time interval;

[0032] S42: For each of the first passengers, input the first clothing status information and the second clothing status information into a first clothing comparison model to obtain a plurality of the first comparison information.

[0033] The present application also proposes a subway engineering building energy-saving design system based on deep learning for implementing the above-mentioned subway engineering building energy-saving design method based on deep learning.

[0034] The subway engineering building energy-saving design method and system proposed by the present application take multiple first entrances and exits corresponding to each subway station as analysis targets, generate multiple first building portraits according to multiple first buildings corresponding to each first entrance and exit, thereby obtain multiple first entrance and exit portraits according to the multiple first building portraits, combine the historical population travel rules of the first entrance and exit to obtain a first lighting control strategy. In addition, according to the multiple first entrance and exit portraits and the clothing comparison situation when passengers get on and off the train, obtain a first air-conditioning control strategy, thereby integrating the first lighting control strategy and the first air-conditioning control strategy in the dimension of time intervals to obtain the first subway energy-saving adjustment strategy. Through the technical solution of the present application, the lighting strategy and the air-conditioning operation strategy in the subway station can be made more in line with the actual needs of users, thereby effectively reducing the energy consumption of the subway station. Description of the Drawings

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only exemplary, and for those of ordinary skill in the art, other implementation drawings can be obtained according to the provided drawings without creative efforts.

[0036] Figure 1 is the execution flowchart of the subway engineering building energy-saving design method based on deep learning of the present invention.

[0037] Figure 2 is a schematic diagram of the corresponding relationship between the first subway station, the multiple first entrances and exits, and the multiple first buildings in the present invention. Detailed Embodiments

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0039] The following will detail the present invention in combination with the accompanying drawings and specific embodiments, where the illustrative embodiments and explanations are only used to explain the present invention, but not to limit the present invention.

[0040] The following will detail the energy-saving design method and system for urban rail transit engineering buildings based on deep learning of the present invention.

[0041] This embodiment proposes an energy-saving design method for urban rail transit engineering buildings based on deep learning, and the method flow is as Figure 1 shown, specifically including the following steps:

[0042] S1: According to the first corresponding relationship between the buildings around the first urban rail transit station and multiple first entrances and exits, obtain multiple first entrance and exit portraits corresponding to the first urban rail transit station.

[0043] An urban rail transit station usually corresponds to multiple entrances and exits, and each entrance and exit corresponds to multiple surrounding buildings. The characteristics of the mobile population corresponding to each surrounding building may not be the same. For example, they can be classified into office buildings, shopping malls or residential buildings according to the building type. For each type of building, other dimensions can be further divided. For example, for residential buildings, they can be divided into multiple dimensions according to the degree of newness and value. Thus, according to the multiple surrounding buildings corresponding to each first entrance and exit, a first entrance and exit portrait used to characterize the characteristics of the traveling population is determined.

[0044] The first urban rail transit station, the multiple first entrances and exits, and the multiple first buildings are in a corresponding relationship, and the specific relationship is shown in Figure 2 .

[0045] The S1 includes the following sub-steps:

[0046] S11: Obtain multiple first entrances and exits corresponding to the first urban rail transit station.

[0047] The first entrances and exits can be obtained from the planning map of the first urban rail transit station.

[0048] S12: Based on each first entrance and exit, obtain multiple first buildings within a first preset distance range according to the multiple first corresponding relationships.

[0049] In order to obtain the travel characteristics of the population corresponding to each of the first inlets and outlets, in this step, it is necessary to take each of the first inlets and outlets as a reference and obtain a plurality of first buildings within a preset distance range.

[0050] Preferably, each of the first inlets and outlets can be saved in one-to-one correspondence with a plurality of first buildings within the first preset distance range to obtain a plurality of the first corresponding relationships.

[0051] For example, if there are 4 buildings within a radius of 3 kilometers around the first inlet and outlet, namely Office Building 1, Office Building 2, Residence 1, and Residence 2, the above four buildings can be used as the plurality of first buildings.

[0052] S13: Determine a plurality of first building portraits according to a plurality of first historical behavior data corresponding to each of the first buildings.

[0053] In order to determine the travel characteristics of the population corresponding to each of the first buildings, the first historical behavior data including those related to shared transportation or receiving takeaways can be obtained, so as to determine the first building portrait for each of the first buildings, and then comprehensively consider the plurality of first buildings corresponding to each of the first inlets and outlets, and finally obtain the first inlet and outlet portrait.

[0054] The S13 may include the following sub-steps:

[0055] S131: For each of the first buildings, obtain a plurality of first historical behavior data within a first preset time interval.

[0056] This step is used to obtain the order receipt quantity of shared transportation tool orders or takeaway receipt addresses with the first building as the departure place or destination within the first preset time interval.

[0057] Since the common transportation tools for office workers to reach their workplaces from urban rail stations are shared transportation tools such as shared bicycles, and orders such as takeaways are usually completed during the corresponding personnel's stay in the first building, the travel time pattern of the personnel in the first building can be obtained through the above first historical behavior data.

[0058] S132: Determine the first building portrait according to the plurality of first historical behavior data and the first travel pattern data.

[0059] For each piece of the first historical behavior data, it indicates that the personnel corresponding to the first building have made an arrival or departure action.

[0060] To further clarify the inbound and outbound actions, it is necessary to perform similarity matching between multiple pieces of the first historical behavior data and the first travel pattern data. The first travel pattern data refers to general travel patterns. For example, for office buildings, arrival actions generally occur in the morning and departure actions occur in the afternoon, while for residential buildings, departure actions generally occur in the morning and arrival actions occur in the afternoon.

[0061] In the first building portrait, the inbound and outbound times and the number of people corresponding to the first building can be relatively clearly characterized. For example, 8 am for work and 100 people can be used as two labels to generate the first building portrait.

[0062] In addition, since the first historical behavior data can also reflect the age group information of the people, the age group information can also be used as part of the first building portrait.

[0063] S14: Generate the first import and export portrait based on multiple first building portraits.

[0064] Since one import and export of the urban rail transit station corresponds to multiple first buildings, the multiple first building portraits corresponding to the one import and export obtained in S13 can be integrated to generate the first import and export portrait.

[0065] Preferably, multiple first labels can be extracted from multiple first building portraits, and the union of the multiple first labels can be used as the first import and export portrait.

[0066] Preferably, according to the frequency of each first label appearing, corresponding weights can be assigned to each first label in the first import and export portrait to enhance the influence of the first label.

[0067] S2: Determine the first lighting control strategy corresponding to each first import and export according to the first historical population travel information and multiple first import and export portraits.

[0068] To a certain extent, each first import and export portrait can reflect the population travel pattern corresponding to the first import and export. However, due to the influence of other factors, it is also necessary to adjust the prediction result of the population travel pattern according to the historical population travel information of the first import and export station.

[0069] S2 includes the following sub-steps:

[0070] S21: Obtain the first historical population travel information of each first import and export within the first preset time interval.

[0071] Since multiple of the first building portraits are established based on the personnel travel data within the first preset time interval, the first historical population travel information should also select the personnel travel data within the first preset time interval.

[0072] S22: Establish a first mapping relationship between the first building portrait corresponding to each first import and export and the first historical population travel information, thereby obtaining multiple first mapping relationships.

[0073] Since it is necessary to predict the population travel pattern of each first import and export based on the first building portrait and the first historical population travel information corresponding to each first import and export, it is necessary to store them associatively in this step.

[0074] S23: Input each of the first mapping relationships into the first population travel pattern prediction model one by one to obtain multiple first population travel pattern information.

[0075] The first population travel pattern prediction model is obtained by training a convolutional neural network model based on historical data. The specific training process is as follows:

[0076] First, use the real historical data of known urban rail station entrances and exits as training sample data. For example, for each historical urban rail station entrance and exit, use the building portrait and historical population travel information of the historical urban rail station entrance and exit within the preset time interval as input data, and use the actual passenger flow data of the historical urban rail station entrance and exit within the future preset time interval as output data to train the convolutional neural network model.

[0077] Second, use the test samples to test the trained model. If the similarity between the output data and the actual data exceeds the preset value, the training of the first population travel pattern prediction model is completed.

[0078] The first population travel pattern information reflects the passenger flow situation of each first import and export. For example, the first building portrait may have a peak passenger flow from 7:00 to 8:00, with a passenger flow of 500 people, and the first historical population travel information may have a peak passenger flow from 7:05 to 8:10, with a passenger flow of 700. Then, according to the specified operation strategy and fitting rule, the first population travel pattern information can be obtained based on the first building portrait and the first historical population travel information.

[0079] S24: Determine the first lighting control strategy according to the first population travel pattern information.

[0080] The first lighting control strategy is formulated according to the first population travel pattern information.

[0081] Preferably, a preset correspondence relationship can be established in advance between the travel pattern information of the first population and the number of lighting devices turned on, so that according to the passenger flow and travel time interval information reflected in the travel pattern information of the first population, a specified number of lighting devices can be controlled to turn on and off according to the travel time interval information. Thus, the matching degree between the passenger flow and the lighting devices is improved, so as to achieve the technical effect of saving energy and reducing consumption.

[0082] S3: Integrate the multiple first import and export portraits to obtain a first station travel portrait.

[0083] Since the passenger flow data of taking the urban rail within the preset time interval is aggregated from multiple import and export ports corresponding to an urban rail station, in order to formulate the air-conditioning adjustment strategy on the urban rail, it is necessary to integrate multiple first import and export portraits to obtain the first station travel portrait of the first urban rail station.

[0084] The step S3 may include the following sub-steps:

[0085] S31: Extract multiple first import and export label information from the multiple first import and export portraits.

[0086] For example, the first import and export label information may be that the peak passenger flow is from 7:00 to 8:00 or the passenger flow is 500 people, and the main population age group is from 25 to 35 years old.

[0087] S32: Obtain a first label union for the multiple first import and export label information.

[0088] The first label union may be to obtain the travel passenger flow of the first urban rail station at different times according to the multiple first import and export label information, so as to obtain a passenger flow sequence within multiple time periods.

[0089] For example, within the time period from 7:00 to 8:00, the passenger flows of the first import and export and the second import and export are 500 people and 200 people respectively. Then, within this time period, the passenger flow of the first urban rail station is 700 people. Therefore, 7:00 to 8:00, 700 people can be used as the label data in the first label union.

[0090] S33: Obtain the first station travel portrait according to the first label union.

[0091] The first station travel portrait is obtained after describing the passenger flow sequence within the multiple time periods.

[0092] S4: Identify the first clothing state information and the second clothing state information of multiple first passengers within the second preset time interval, and obtain multiple first comparison information according to the multiple first clothing state information and the multiple second clothing state information.

[0093] In order to verify whether the operating state of the air conditioner matches the actual needs of the passengers, in this step, pattern recognition technology is required to compare the status information of the passengers before and after boarding.

[0094] The S4 includes the following sub-steps:

[0095] S41: Within a second preset time interval, obtain the first clothing status information and the second clothing status information of each of the first passengers.

[0096] The second preset time interval refers to the preset time interval before the time node when the air conditioner operating state needs to be adjusted.

[0097] Among them, each of the first passengers corresponds to one piece of the first clothing status information and one piece of the second clothing status information. The first clothing status information refers to the clothing information of the first passenger when waiting for the vehicle, and the second clothing status information refers to the clothing information of the first passenger after getting on the vehicle.

[0098] For example, for a first passenger, the first clothing status information may be wearing shorts and a short-sleeved shirt when waiting for the vehicle, and the second clothing status information may be wearing long pants and a short-sleeved shirt after getting on the vehicle.

[0099] S42: For each of the first passengers, input the first clothing status information and the second clothing status information into the first clothing comparison model to obtain a plurality of the first comparison information.

[0100] The first clothing comparison model is used to compare the clothing status of each of the first passengers before and after getting on the vehicle, so as to further analyze and obtain the matching degree between the operating state of the air conditioner and the passengers.

[0101] The first clothing comparison model is preferably also obtained by training a convolutional neural network model with historical sample data.

[0102] The first comparison information can be divided into two ways: adding clothes and removing clothes. Adding clothes indicates that the operating temperature of the air conditioner is too low, and removing clothes indicates that the operating temperature of the air conditioner is too high.

[0103] S5: Determine the first air conditioner control strategy according to a plurality of the first comparison information and the first station travel portrait.

[0104] Since people of different age groups have different preferences for the operating state of the air conditioner, based on the plurality of the first comparison information determined in S4, it is also necessary to combine the age information of the travel population reflected in the first station travel portrait to generate the final first air conditioner control strategy.

[0105] Preferably, the convolutional neural network model can be trained according to the historical sample information composed of the comparison information and the site travel portrait, so as to obtain the first air-conditioning control strategy determination model. Wherein, in the training process, the comparison information and the site travel portrait are used as input data, and the air-conditioning control strategy composed of parameters such as the temperature and wind speed of the air conditioner is used as output data.

[0106] Thereby, a plurality of the first comparison information and the first site travel portrait are input into the first air-conditioning control strategy determination model to obtain the first air-conditioning control strategy.

[0107] S6: Obtain the first urban rail energy-saving adjustment strategy according to the first lighting control strategy and the first air-conditioning control strategy.

[0108] In this step, according to the specified time series, the first lighting control strategy and the first air-conditioning control strategy corresponding to each time interval are integrated to obtain the first urban rail energy-saving adjustment strategy.

[0109] Preferably, from 7:00 to 8:00, the first urban rail energy-saving adjustment strategy can be that 50% of the lighting equipment is turned on and 50% of the air-conditioning cooling is turned on.

[0110] The present application also proposes an urban rail engineering building energy-saving design system based on deep learning, which is used to execute the above-mentioned urban rail engineering building energy-saving design method based on deep learning.

[0111] The urban rail engineering building energy-saving design method and system proposed in the present application take multiple first entrances and exits corresponding to each urban rail station as analysis targets, generate multiple first building portraits according to multiple first buildings corresponding to each first entrance and exit, and thus obtain multiple first entrance and exit portraits according to the multiple first building portraits. Combining the historical population travel rules of the first entrance and exit to obtain the first lighting control strategy. In addition, according to the multiple first entrance and exit portraits and the comparison of the clothing of passengers when getting on and off the train, the first air-conditioning control strategy is obtained, and thus the first lighting control strategy and the first air-conditioning control strategy are integrated in the dimension of time intervals to obtain the first urban rail energy-saving adjustment strategy. Through the technical solution of the present application, the lighting strategy and air-conditioning operation strategy in the urban rail station can be made more in line with the actual needs of users, thereby effectively reducing the energy consumption of the urban rail station.

[0112] The above is only the preferred embodiment of the present invention, so all equivalent changes or modifications made according to the structure, features and principles described in the scope of the present invention patent application are included in the scope of the present invention patent application.

Claims

1. An energy-saving design method for urban rail transit engineering buildings based on deep learning, characterized in that The method includes: S1: Obtain multiple first entrance and exit portraits corresponding to the first urban rail transit station according to the first corresponding relationship between the buildings around the first urban rail transit station and multiple first entrances and exits; wherein, the first entrance and exit portrait is used to characterize the travel population characteristic information corresponding to the first entrance and exit; S2: Determine the first lighting control strategy corresponding to each first entrance and exit according to the first historical population travel information and multiple first entrance and exit portraits; S3: Integrate multiple first entrance and exit portraits to obtain a first station travel portrait; S4: Identify the first clothing status information and the second clothing status information of multiple first passengers within a second preset time interval, and obtain multiple first comparison information according to the multiple first clothing status information and the multiple second clothing status information; wherein, the first clothing status information refers to the clothing information of the first passenger when waiting for the train, and the second clothing status information refers to the clothing information of the first passenger after getting on the train; S5: Determine the first air-conditioning control strategy according to the multiple first comparison information and the first station travel portrait; S6: Obtain the first urban rail transit energy-saving adjustment strategy according to the first lighting control strategy and the first air-conditioning control strategy.

2. The energy-saving design method for urban rail engineering buildings based on deep learning according to claim 1, characterized in that, The S1 includes the following sub-steps: S11: Obtain multiple first entrances and exits corresponding to the first urban rail transit station; S12: Taking each first entrance and exit as a reference, obtain multiple first buildings within a first preset distance range according to the multiple first corresponding relationships; S13: Determine multiple first building portraits according to multiple first historical behavior data corresponding to each first building; S14: Generate the first entrance and exit portrait according to the multiple first building portraits.

3. The energy-saving design method for urban rail transit engineering buildings based on deep learning according to claim 2, characterized in that, The S13 includes the following sub-steps: S131: For each first building, obtain multiple first historical behavior data within a first preset time interval; S132: Determine the first building portrait according to the multiple first historical behavior data and the first travel rule data.

4. The energy-saving design method for urban rail engineering buildings based on deep learning according to claim 3, characterized in that, The S2 includes the following sub-steps: S21: Obtain the first historical population travel information of each first entrance and exit within a first preset time interval; S22: Establish a first mapping relationship between the first building portrait corresponding to each first entrance and exit and the first historical population travel information, so as to obtain multiple first mapping relationships; S23: Input each first mapping relationship into the first population travel rule prediction model one by one to obtain multiple first population travel rule information; S24: Determine the first lighting control strategy according to the first population travel rule information.

5. The energy-saving design method for urban rail engineering buildings based on deep learning according to claim 4, characterized in that The S3 includes the following sub-steps: S31: Extract multiple first entrance and exit label information from multiple first entrance and exit portraits; S32: Obtain the first label union of the multiple first entrance and exit label information; S33: Obtain the first station travel portrait according to the first label union.

6. The energy-saving design method for urban rail transit engineering buildings based on deep learning according to claim 5, characterized in that The S4 includes the following sub-steps: S41: Within the second preset time interval, obtain the first clothing status information and the second clothing status information of each first passenger; S42: For each of the first passengers, input the first clothing status information and the second clothing status information into the first clothing comparison model to obtain a plurality of the first comparison information.

7. An urban rail transit engineering building energy-saving design system based on deep learning, which is used to implement the urban rail transit engineering building energy-saving design method according to claims 1-6.

Images