Subway air conditioning system load prediction method and device based on passenger flow dynamic characteristics

Abstract

The invention provides a subway air-conditioning system load prediction method and device based on passenger flow dynamic characteristics, and the method comprises the steps: firstly obtaining in-out passenger flow parameters and the average passenger flow in a plurality of past preset time periods in real time, and then taking the average passenger flow in the plurality of past preset time periods as an original time sequence, the method comprises the steps that firstly, a single-parameter grey prediction model is established, the average passenger flow volume in a future preset time period is predicted through a single-parameter grey prediction model and an original time sequence, then load type classification is conducted on loads of the subway air conditioning system, and a simplified physical model containing various load parameters is established through a multiple linear regression method according to the average passenger flow volume in the future preset time period; and finally, initializing the simplified physical model, inputting the incoming and outgoing passenger flow parameters into the initialized simplified physical model, and predicting the load of the subway air conditioning system in a future preset time period. Compared with the prior art, the method has the advantages of being short in learning time, high in universality and high in prediction precision.

Description

Technical Field

[0001] The invention belongs to the technical field of subway air-conditioning load prediction, and in particular relates to a subway air-conditioning system load prediction method and device based on passenger flow dynamic characteristics. Background Art

[0002] The subway air conditioning system mainly includes the air conditioning cold source system, the public area ventilation and air conditioning system, and the equipment and management room air conditioning system. The subway air conditioning system mainly includes two modes during operation, namely the air conditioning mode based on mechanical refrigeration and the mechanical ventilation mode based on ventilation cooling. In the actual operation of the subway air conditioning system, the selection of the operation mode and the setting of specific operation parameters (such as supply and return water temperature, water pump frequency and fan frequency, etc.) have a significant impact on the operation energy consumption.

[0003] In the prior art, the corresponding air conditioning operation mode is usually scheduled according to the season. In the prior art, there are two technical solutions for the adjustment of each mode: 1. The setting parameters such as the supply and return water temperature, water pump frequency and fan frequency of the corresponding mode remain fixed; 2. PID feedback is used to adjust the fan or water pump frequency. The above two methods usually cannot pay attention to the real-time changes in passenger flow. In the actual operation process, there is often a situation where the temperature is adjusted too low or the operating load is too large. Therefore, it is of great significance to make accurate predictions for the subway air-conditioning system to assist the air-conditioning system in adjusting the set parameters.

[0004] In order to realize the load prediction of subway air conditioners, some scholars use data-driven algorithms to achieve it and design the operation mode of subway air conditioners according to the predicted load. However, the use of data-driven algorithms for air conditioning load prediction requires a large amount of historical data to train the model in advance, which takes a long time to learn and has high learning costs. In addition, the flow of people and the environment between different subway stations vary significantly, and retraining and learning are required for different station halls, which has poor versatility. Summary of the invention

[0005] The present invention provides a method and device for predicting the load of a subway air-conditioning system based on dynamic characteristics of passenger flow, which are used to solve the problems existing in the prior art.

[0006] In a first aspect, to achieve the above-mentioned purpose, the present invention provides a method for predicting the load of a subway air conditioning system based on dynamic characteristics of passenger flow, which comprises:

[0007] S10: obtaining in-and-out passenger flow parameters and average passenger flow in multiple past preset time periods in real time;

[0008] S20: taking the average passenger flow in multiple past preset time periods as the original time series, and using the single-parameter grey prediction model and the original time series to predict the average passenger flow in the future preset time period;

[0009] S30: classify the load types of the subway air conditioning system load, and establish a simplified physical model including various load parameters using a multivariate linear regression method based on the average passenger flow in a future preset time period;

[0010] The load parameters include short-term load, short-term load weight, long-term load, long-term load weight, dynamic load and dynamic load weight. The short-term load weight, long-term load weight and dynamic load weight are determined by the average passenger flow respectively.

[0011] S40: Initializing the simplified physical model, inputting the in-and-out passenger flow parameters into the initialized simplified physical model, and predicting the load of the subway air-conditioning system in a future preset time period.

[0012] Through the above technical solution, the present invention adopts the time series analysis method to take the average passenger flow in the past preset time period as the time series, and uses the gray system theory to predict the average passenger flow in the future preset time period as an important influencing factor affecting the air-conditioning system load, and then classifies the subway air-conditioning system load to construct a simplified physical model, and then inputs the in-and-out passenger flow parameters into the initialized simplified physical model, so as to predict the future load of the subway air-conditioning system. Compared with the prior art, the method of the present invention has the advantages of short learning time, strong versatility and high prediction accuracy.

[0013] Preferably, step S20 includes:

[0014] S21: taking the average passenger flow in multiple past preset time periods as the original time series;

[0015] S22: Generate a cumulative sequence based on the original time series;

[0016] S23: Based on the original time series, the adjacent mean sequence is calculated;

[0017] S24: Establish a single parameter grey prediction model based on the cumulative sequence and the adjacent mean sequence;

[0018] S25: The average passenger flow in a preset time period in the past is placed into a single-parameter grey prediction model for calculation to predict the average passenger flow in a future time period.

[0019] Preferably, the specific formula of step S21 is:

[0020] X (0) =(T1,......,T n )(1)

[0021] Among them, X (0) represents the original time series, n is the sequence number, T n is the average passenger flow corresponding to sequence number n.

[0022] Preferably, the specific formula of step S22 is:

[0023] X (1) =(y1,......,y n )(2)

[0024]

[0025] Among them, X (1) represents the cumulative sequence, k is the number of days in the preset time period, y n is the cumulative value of the n average passenger flows in the original time series.

[0026] Preferably, the specific formula of step S23 is:

[0027] Z (1) =(z2,......,z n )(4)

[0028] z n =0.5T k +0.5T k+1 ,k=2,......,n(5)

[0029] Among them, Z (1) represents a sequence close to the mean.

[0030] Preferably, the specific formula of step S24 is:

[0031]

[0032] Y=(T2,......,T n ) (8)

[0033]

[0034] Among them, a and b are the equation parameters of the single parameter grey prediction model, B is Z (1) The coefficient matrix obtained by transposing, Y is the true value vector of average passenger flow.

[0035] Preferably, the specific formula of step S25 is:

[0036]

[0037] Among them, T next Represents the average passenger flow in a future time period.

[0038] Preferably, step S30 includes:

[0039] S31: The load of the subway air conditioning system is divided into long-term load, short-term load and dynamic load;

[0040] S32: Calculate the short-term load weight, long-term load weight, and dynamic load weight using the multiple linear regression method based on the average passenger flow in the future time period to obtain a simplified physical model.

[0041] Preferably, the simplified physical model includes Formula (12) and Formula (13):

[0042] Q total = a1Q1 + a s Q S + a d Q d (12)

[0043]

[0044] Wherein, a1 is the weight coefficient of the long-term load, a s is the weight coefficient of the short-term load, a d is the weight coefficient of the dynamic load; Q total is the total load prediction value, Q1 is the long-term load, Q s is the short-term load, and Q d is the dynamic load; k is the number of days in the preset time period. To distinguish from the transpose symbol T and facilitate the understanding of the formula, the vector matrix of T next in Formula (13) is represented by x i , X represents the transpose vector of x i , Y is the true value vector of the average passenger flow, that is, the inbound and outbound passenger flow parameters; if k > 7, the calculated β' is assigned to a1; if 1 < k ≤ 7, the calculated β' is assigned to a s ; if k = 1, the calculated β' is assigned to a d ; Q1, Q s and Q d are the products of the daily average load and the corresponding k value.

[0045] Preferably, after step S40, the method of the present invention further includes:

[0046] S50: Regulate the working mode of the subway air conditioning system according to the load in the future preset time period.

[0047] Second, to achieve the invention purpose, the present invention also provides a subway air conditioning system load prediction device based on passenger flow dynamic characteristics. The system includes:

[0048] A data acquisition module for real-time acquiring the inbound and outbound passenger flow parameters and the average passenger flow in multiple past preset time periods;

[0049] The passenger flow prediction module is used to take the average passenger flow in multiple past preset time periods as the original time series, establish a single-parameter grey prediction model using the original time series, and use the single-parameter grey prediction model to predict the average passenger flow in the future preset time period;

[0050] The model building module is used to classify the load types of the subway air conditioning system load and establish a simplified physical model containing various load parameters using the multivariate linear regression method based on the average passenger flow in the future preset time period;

[0051] The load parameters include short-term load, short-term load weight, long-term load, long-term load weight, dynamic load and dynamic load weight. The short-term load weight, long-term load weight and dynamic load weight are determined by the average passenger flow respectively.

[0052] The load prediction module is used to initialize the simplified physical model, input the in-and-out passenger flow parameters into the initialized simplified physical model, and predict the load of the subway air-conditioning system in a preset time period in the future.

[0053] The present invention has at least the following advantages:

[0054] 1) The present invention adopts the time series analysis method to take the average passenger flow in the past preset time period as the time series, and uses the gray system theory to predict the average passenger flow in the future preset time period as an important influencing factor affecting the air-conditioning system load. Then the subway air-conditioning system load is classified to construct a simplified physical model, and then the inbound and outbound passenger flow parameters are input into the initialized simplified physical model to predict the future load of the subway air-conditioning system. Compared with the prior art, the method of the present invention has the advantages of short learning time, strong versatility and high prediction accuracy.

[0055] 2) The method of the present invention designs a single-parameter grey prediction model for subway air-conditioning systems. The model requires less modeling data, is easy to calculate and has high modeling accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 It is a flow chart of a method for predicting load of a subway air conditioning system based on dynamic characteristics of passenger flow in an embodiment of the present invention;

[0058] Figure 2It is a structural schematic diagram of a subway air conditioning system load prediction device based on passenger flow dynamic characteristics in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] In the description of the present invention, it should be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they cannot be understood as limitations on the present invention.

[0061] This embodiment discloses a method and device for predicting the load of a subway air-conditioning system based on dynamic characteristics of passenger flow, which are used to solve the problems existing in the prior art.

[0062] Embodiment 1

[0063] like Figure 1 As shown, this embodiment discloses a method for predicting the load of a subway air conditioning system based on dynamic characteristics of passenger flow, which includes:

[0064] S10: Obtain in-and-out passenger flow parameters and average passenger flow in multiple past preset time periods in real time.

[0065] In this embodiment, the in-and-out passenger flow parameters can be obtained through the camera in the station combined with the existing visual model recognition, or through the passenger's card swiping record. In this embodiment, the subway air-conditioning system operating parameters can also be obtained at the same time. The subway air-conditioning system operating parameters at least include the water supply temperature, return water temperature and water flow of the subway air-conditioning system. These parameters can be obtained through sensors installed in the air-conditioning system. The average passenger flow in multiple past preset time periods can be obtained through simple calculations based on the pre-determined time periods and passenger card swiping records. These three types of data can be obtained synchronously to improve the data processing efficiency of the method of the present invention.

[0066] S20: taking the average passenger flow in multiple past preset time periods as the original time series, and using the single-parameter grey prediction model and the original time series to predict the average passenger flow in a future preset time period.

[0067] Specifically, step S20 includes:

[0068] S21: The average passenger flow in multiple past preset time periods is used as the original time series:

[0069] X (0) =(T1,......,T n )(1)

[0070] Among them, X (0) represents the original time series, n is the sequence number, T n is the average passenger flow corresponding to sequence number n.

[0071] S22: Generate cumulative sequence based on the original time series:

[0072] X (1) =(y1,......,y n )(2)

[0073]

[0074] Among them, X (1) represents the cumulative sequence, k is the number of days in the preset time period, y n is the cumulative value of the n average passenger flows in the original time series.

[0075] S23: Based on the original time series, calculate the adjacent mean sequence:

[0076] Z (1) =(z2,......,z n )(4)

[0077] z n =0.5T k +0.5T k+1 ,k=2,......,n(5)

[0078] Among them, Z (1) represents a sequence close to the mean.

[0079] S24: According to the cumulative sequence and the adjacent mean sequence, a single parameter grey prediction model is established:

[0080]

[0081] Y=(T2,......,T n ) (8)

[0082]

[0083] Among them, a and b are the equation parameters of the single parameter grey prediction model, B is Z 1 The coefficient matrix obtained by transposing, Y is the real value vector of average passenger flow;

[0084] S25: The average passenger flow in a preset time period in the past is placed into a single-parameter grey prediction model for calculation to predict the average passenger flow in a future time period.

[0085] The specific calculation formula of the above step S25 is:

[0086]

[0087] Among them, T next Represents the average passenger flow in a future time period.

[0088] Through the above steps S21-S25 and their corresponding formulas (1)-(10), step S2 in the method of the present invention takes the average passenger flow in the preset time period as the original time series, and then generates a cumulative series and a sequence adjacent to the mean to establish a single-parameter grey prediction model, and then predicts the average passenger flow in the future preset time period based on the established single-parameter grey prediction model, so as to serve as the basis for predicting the load of the subway air-conditioning system, which greatly improves the accuracy of the prediction result.

[0089] After the above step S20, the method of the present invention further comprises:

[0090] S30: Classify the load types of the subway air-conditioning system loads, and use the multivariate linear regression method to establish a simplified physical model containing various load parameters based on the average passenger flow in a preset time period in the future.

[0091] In this embodiment, the load parameters include short-term load, short-term load weight, long-term load, long-term load weight, dynamic load and dynamic load weight. The short-term load weight, long-term load weight and dynamic load weight are respectively determined by the average passenger flow.

[0092] Specifically, the above step S30 includes:

[0093] S31: The subway air-conditioning system load is divided into long-term load, short-term load and dynamic load.

[0094] S32: According to the average passenger flow in the future time period, the short-term load weight, the long-term load weight and the dynamic load weight are calculated by using the multivariate linear regression method to obtain a simplified physical model.

[0095] In this embodiment, while executing the above step S30, multi-thread synchronization can also be used to calculate the real-time load Q of the subway air conditioning system. The specific calculation formula is as follows:

[0096] Q=mc p (t1-t2) (11)

[0097] Among them, Q is the real-time load, m is the water system flow rate, c p is the specific heat of water, t1 is the return water temperature, and t2 is the supply water temperature.

[0098] The simplified physical model of the above step S32 is specifically:

[0099] Q total = a1Q1 + a s Q S + a d Q d (12)

[0100] Among them, a1 is the weight coefficient of the long-term load, a s is the weight coefficient of the short-term load, a d is the weight coefficient of the dynamic load. Q total is the total load prediction value, Q1 is the long-term load, Q s is the short-term load and Q d is the dynamic load.

[0101] It should be explained that the calculation formulas of the above Q1, Q s and Q d are:

[0102]

[0103] Among them, a1 is the weight coefficient of the long-term load, a s is the weight coefficient of the short-term load, a d is the weight coefficient of the dynamic load; Q total is the total load prediction value, Q1 is the long-term load, Q s is the short-term load and Q d is the dynamic load; k is the number of days in the preset time period. To distinguish it from the transpose symbol T and facilitate the understanding of the formula, the vector matrix of T next in formula (13) is represented by x i denotes, X represents the transpose vector of x i and Y is the vector of the true value of the average passenger flow, that is, the inbound and outbound passenger flow parameters.

[0104] In this embodiment, if k > 7, the calculated β' is assigned to a1; if 1 < k ≤ 7, the calculated β' is assigned to a s ; if k = 1, the calculated β' is assigned to a d ; Q1, Q s and Q d are the products of the daily average load and the corresponding k value. Multiple k values can be set in segments to calculate a1, a s and a d .

[0105] In other embodiments, the above k value assignment condition can be adjusted according to actual conditions.

[0106] S40: Initializing the simplified physical model, inputting the in-and-out passenger flow parameters into the initialized simplified physical model, and predicting the load of the subway air-conditioning system in a future preset time period.

[0107] Through the above technical solution, the present invention adopts the time series analysis method to take the average passenger flow in the past preset time period as the time series, and uses the gray system theory to predict the average passenger flow in the future preset time period as an important influencing factor affecting the air-conditioning system load, and then classifies the subway air-conditioning system load to construct a simplified physical model, and then inputs the in-and-out passenger flow parameters into the initialized simplified physical model, so as to predict the future load of the subway air-conditioning system. Compared with the prior art, the method of the present invention has the advantages of short learning time, strong versatility and high prediction accuracy.

[0108] After the above step S40, the method of the present invention further includes:

[0109] S50: Adjusting the working mode of the subway air conditioning system according to the load in a future preset time period.

[0110] Specifically, the above step S50 includes:

[0111] S51: Extract the total load forecast value Q within the future preset time period total , outdoor air parameters and station public area air parameters in the in-and-out passenger flow parameters;

[0112] S52: Based on the total load forecast value Q total Calculate the required air volume G′ of the subway air conditioning system in ventilation mode, and calculate the required air volume G′ according to the maximum air volume G′ of the subway air conditioning system. max Determine the operating mode of the subway air conditioning system.

[0113] Through the above technical solution, according to the total load forecast value Q in the future preset time period total and the required air volume G′ to control the subway air-conditioning system to enter the corresponding working mode, thereby achieving precise control, energy saving and environmental protection.

[0114] Furthermore, the above step S52 specifically includes:

[0115] S521: Determine the required air volume G′ in the ventilation mode and the maximum air volume G of the subway air conditioning system max Whether the first preset condition is met, if yes, proceed to S522, otherwise, proceed to S523.

[0116] Among them, the first preset condition is: G′≤β*G max, β is the preset air volume coefficient, and β∈(0,1);

[0117] S522: Calculate a first predicted operating frequency f1 of the box fan of the subway air-conditioning system according to a relationship between the required air volume G′ and the box fan frequency f of the subway air-conditioning system, and determine whether the box fan of the subway air-conditioning system enters a natural ventilation mode or a mechanical ventilation mode according to the first predicted operating frequency f1;

[0118] Among them, the relationship between the required air volume G′ and the box fan frequency f of the subway air conditioning system is:

[0119] G′=a1*f2+b1*f+c1 (14)

[0120] Among them, a1 is the quadratic fitting coefficient of the ventilation mode, b1 is the linear fitting coefficient of the ventilation mode, and c1 is the fitting constant of the ventilation mode;

[0121] The above step S523 includes:

[0122] According to the total load prediction value Q total Calculate the required air volume G″ in air conditioning mode;

[0123] Calculate the indoor air enthalpy setting value h according to the indoor preset temperature and humidity n_set ;

[0124] And set the value h according to the indoor air enthalpy n_set and outdoor air enthalpy h w Make sure the fans of the subway air conditioning system enter the fresh air conditioning mode or the zero mechanical fresh air conditioning mode;

[0125] Among them, the required air volume G″ in air conditioning mode and the indoor air enthalpy setting value h n_set The relationship is:

[0126] G″=Q total / k2(h n_set -h s_set )(15)

[0127] Among them, h n_set is the indoor air enthalpy setting value, h s_set is the set value of supply air enthalpy, and k2 is the preset fixed coefficient.

[0128] The beneficial effects of the above steps S521-S522 are:

[0129] By judging the required air volume G′ in the ventilation mode and the maximum air volume G of the subway air conditioning system maxWhether the first preset condition is met, the subway air conditioning system can be controlled to enter the ventilation mode or the air conditioning mode, and then the fan of the subway air conditioning system can be controlled to enter the natural ventilation mode or the mechanical ventilation mode according to the first predicted operating frequency f1 corresponding to the required air volume G′ in the ventilation mode, or the internal air enthalpy value h corresponding to the required air volume G″ in the air conditioning mode can be set. n_set It ensures that the fans of the subway air-conditioning system enter the fresh air air-conditioning mode or the zero mechanical fresh air air-conditioning mode, and accurately switches the corresponding working modes for different environments, greatly improving the energy-saving efficiency of the subway air-conditioning system.

[0130] In the above step S522, determining whether the box fan of the subway air conditioning system enters the natural ventilation mode or the mechanical ventilation mode according to the first predicted operating frequency f1 is specifically as follows:

[0131] Determine the first predicted operating frequency f1 and the minimum fan frequency f of the subway air conditioning system _min Whether the second preset condition is met, if yes, proceed to step S524; otherwise, proceed to S525;

[0132] Among them, the second preset condition is: f1≤f _min ;

[0133] S524: Adjust the operating frequency of the fan of the subway air conditioning system to the minimum fan frequency f _min , and enter the natural ventilation mode;

[0134] S525: Adjust the operating frequency of the box fan of the subway air-conditioning system to the first predicted operating frequency f1, and enter the mechanical ventilation mode.

[0135] By comparing the first predicted operating frequency f1 with the minimum fan frequency f of the subway air conditioning system _min Whether the second preset condition is met, the operating frequency of the subway air-conditioning system box fan is adaptively adjusted, and the natural ventilation mode or mechanical ventilation mode is entered accordingly to achieve more refined control of the ventilation mode.

[0136] Furthermore, in the above steps, the value h is set according to the indoor air enthalpy value. n_set and outdoor air enthalpy h w Determining that the fans of the subway air conditioning system enter the fresh air conditioning mode or the zero mechanical fresh air conditioning mode specifically includes:

[0137] Determine the indoor air enthalpy setting value h n_set and outdoor air enthalpy h w Whether the third preset condition is met, if yes, proceed to step S526; otherwise, proceed to step S527;

[0138] Among them, the third preset condition is: h w≤γ≤h n_set , γ is the preset indoor air enthalpy coefficient.

[0139] S526: Calculate the second predicted operating frequency f2 of the box fan of the subway air-conditioning system according to the relationship between the required air volume G″ in the fresh air air-conditioning mode and the box fan frequency f of the subway air-conditioning system, and adjust the operating frequency of the box fan of the subway air-conditioning system to the second predicted operating frequency f2, and enter the fresh air air-conditioning mode;

[0140] Among them, the relationship between the required air volume G″ in the fresh air air conditioning mode and the box fan frequency f of the subway air conditioning system is:

[0141] G″=a2*f2+b2*f+c2 (16)

[0142] Among them, a2 is the quadratic fitting coefficient of the fresh air air-conditioning mode, b2 is the linear fitting coefficient of the fresh air air-conditioning mode, and c2 is the fitting constant of the fresh air air-conditioning mode.

[0143] The above step S527 specifically includes:

[0144] The third predicted operating frequency f3 of the box fan of the subway air conditioning system is calculated according to the relationship between the required air volume G″ in the zero mechanical fresh air air conditioning mode and the box fan frequency f of the subway air conditioning system;

[0145] The fourth predicted operating frequency f4 of the return fan of the subway air conditioning system is calculated according to the relationship between the required air volume G″ in the zero mechanical fresh air air conditioning mode and the return fan frequency of the subway air conditioning system;

[0146] Adjusting the operating frequency of the box fan of the subway air conditioning system to the third predicted operating frequency f3;

[0147] Adjusting the operating frequency of the return fan of the subway air conditioning system to the fourth predicted operating frequency f4, and entering the zero mechanical fresh air air conditioning mode;

[0148] The relationship between the required air volume G″ in the zero mechanical fresh air air conditioning mode and the fresh air fan frequency f of the subway air conditioning system is as follows:

[0149] G″=a3*f2+b3*f+c3 (17)

[0150] Among them, a3 is the quadratic fitting coefficient of the air supply fan in the zero mechanical fresh air air conditioning mode, b3 is the first-order fitting coefficient of the air supply fan in the zero mechanical fresh air air conditioning mode, and c3 is the fitting constant of the air supply fan in the zero mechanical fresh air air conditioning mode.

[0151] The relationship between the required air volume G″ in the zero mechanical fresh air air conditioning mode and the return fan frequency of the subway air conditioning system is as follows:

[0152] G″=a4*f2+b4*f+c4 (18)

[0153] Among them, a4 is the quadratic fitting coefficient of the return fan in the zero mechanical fresh air air conditioning mode, b4 is the first-order fitting coefficient of the return fan in the zero mechanical fresh air air conditioning mode, and c4 is the fitting constant of the return fan in the zero mechanical fresh air air conditioning mode.

[0154] The beneficial effects of the above steps S526-S527 are:

[0155] By comparing the indoor air enthalpy set value h n_set and outdoor air enthalpy h w Whether the third preset condition is met, the operating frequency of the box fan and return fan of the subway air-conditioning system can be calculated, and the corresponding fresh air air-conditioning mode or zero mechanical fresh air air-conditioning mode can be entered to achieve more refined control of the air-conditioning mode.

[0156] Furthermore, the method of the present invention further comprises the following steps:

[0157] S60: Collecting and storing the inbound and outbound passenger flow parameters and air conditioning system operation parameters;

[0158] S70: Online learning and updating of parameter information of the simplified physical model according to the in-and-out passenger flow parameters and the air-conditioning system operation parameters.

[0159] The beneficial effect of the above steps S60-S70 is that the prediction accuracy can be continuously improved and the adaptability between different subway stations is maintained.

[0160] Embodiment 2

[0161] like Figure 2 As shown, this embodiment discloses a subway air conditioning system load prediction device based on passenger flow dynamic characteristics, the system comprising:

[0162] A data acquisition module is used to obtain in- and out-of-station passenger flow parameters, subway air conditioning system operation parameters, and average passenger flow in multiple past preset time periods in real time;

[0163] The passenger flow prediction module is used to take the average passenger flow in multiple past preset time periods as the original time series, establish a single-parameter grey prediction model using the original time series, and use the single-parameter grey prediction model to predict the average passenger flow in the future preset time period;

[0164] The model building module is used to classify the load types of the subway air conditioning system load and establish a simplified physical model containing various load parameters using the multivariate linear regression method based on the average passenger flow in the future preset time period;

[0165] The load parameters include short-term load, short-term load weight, long-term load, long-term load weight, dynamic load and dynamic load weight. The short-term load weight, long-term load weight and dynamic load weight are determined by the average passenger flow respectively.

[0166] The load prediction module is used to initialize the simplified physical model, input the inbound and outbound passenger flow parameters into the initialized simplified physical model, and predict the load of the subway air conditioning system in a preset time period in the future;

[0167] The control module is used to control the working mode of the subway air-conditioning system according to the load in a preset time period in the future.

[0168] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc. Various media that can store program codes.

[0169] For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and concepts described above, and all of these changes and deformations should fall within the protection scope of the claims of the present invention.

Claims

1. A method for predicting subway air conditioning system load based on passenger flow dynamic characteristics, characterized in that: include: S10: obtaining in-and-out passenger flow parameters and average passenger flow in multiple past preset time periods in real time; S20: taking the average passenger flow in multiple past preset time periods as original time series, establishing a single-parameter grey prediction model using the original time series, and using the single-parameter grey prediction model to predict the average passenger flow in a future preset time period; S30: classify the load types of the subway air conditioning system load, and establish a simplified physical model including various load parameters using a multivariate linear regression method based on the average passenger flow in a future preset time period; The load parameters include short-term load, short-term load weight, long-term load, long-term load weight, dynamic load and dynamic load weight, and the short-term load weight, long-term load weight and dynamic load weight are respectively determined by the average passenger flow; S40: Initializing the simplified physical model, inputting the in-and-out passenger flow parameters into the initialized simplified physical model, and predicting the load of the subway air-conditioning system in a future preset time period.

2. The method for predicting subway air conditioning system load based on passenger flow dynamic characteristics according to claim 1, characterized in that: The step S20 comprises: S21: taking the average passenger flow in multiple past preset time periods as the original time series; S22: Generate a cumulative sequence according to the original time series; S23: Calculate and obtain a close-by mean sequence according to the original time series; S24: establishing a single-parameter grey prediction model according to the cumulative sequence and the adjacent mean sequence; S25: The average passenger flow in a preset time period in the past is placed into the single-parameter grey prediction model for calculation to predict the average passenger flow in a future time period.

3. The method for predicting subway air conditioning system load based on passenger flow dynamic characteristics according to claim 2, characterized in that: The specific formula of step S21 is: X (0) =(T1,......,X n )(1) Among them, X (0) represents the original time series, n is the sequence number, T n is the average passenger flow corresponding to sequence number n.

4. The method for predicting subway air conditioning system load based on passenger flow dynamic characteristics according to claim 3 is characterized in that: The specific formula of step S22 is: X (1) =(y1,......,y n )(2) Among them, X (1) represents the cumulative sequence, k is the number of days in the preset time period, y n is the cumulative value of the n average passenger flows in the original time series.

5. The method for predicting subway air conditioning system load based on passenger flow dynamic characteristics according to claim 4 is characterized in that: The specific formula of step S23 is: WITH (1) =(z2,......,z n )(4) z n =0.5T k +0.5T k+1 ,k=2,......,n(5) Among them, Z (1) represents a sequence close to the mean.

6. The method for predicting subway air conditioning system load based on passenger flow dynamic characteristics according to claim 5, characterized in that: The specific formula of step S24 is: Y=(T2,......,T n ) (8) Among them, a and b are the equation parameters of the single parameter grey prediction model, B is Z (1) The coefficient matrix obtained by transposing, Y is the true value vector of average passenger flow.

7. The method for predicting subway air conditioning system load based on passenger flow dynamic characteristics according to claim 6, characterized in that: The specific formula of step S25 is: Among them, T next Represents the average passenger flow in a future time period.

8. The method and device for predicting subway air conditioning system load based on passenger flow dynamic characteristics according to claim 7, characterized in that: The step S30 comprises: S31: The load of the subway air conditioning system is divided into long-term load, short-term load and dynamic load; S32: According to the average passenger flow in the future time period, the short-term load weight, the long-term load weight and the dynamic load weight are calculated by using a multivariate linear regression method to obtain the simplified physical model.

9. The method for predicting subway air conditioning system load based on passenger flow dynamic characteristics according to claim 8, characterized in that: The simplified physical model includes formula (12) and formula (13): Q total =a1Q1+a s Q S +a d Q d (12) Among them, a1 is the weight coefficient of the long-term load, a s is the weight coefficient of the short-term load, a d is the weight coefficient of the dynamic load; Q total is the total load prediction value, Q1 is the long-term load, Q s is the short-term load, and Q d is the dynamic load; k is the number of days in the preset time period. To distinguish it from the transpose symbol T and facilitate the understanding of the formula, the vector matrix of T in formula (13) next is represented by x i denotes, X represents the transpose vector of x i and Y is the vector of the true value of the average passenger flow, that is, the inbound and outbound passenger flow parameters; if k > 7, the calculated β' is assigned to a1; if 1 < k ≤ 7, the calculated β' is assigned to a s ; if k = 1, the calculated β' is assigned to a d ; Q1, Q s and Q d are the products of the daily average load and the corresponding k value.

10. A subway air conditioning system load prediction device based on passenger flow dynamic characteristics, characterized in that: include: A data acquisition module is used to obtain in- and out-of-station passenger flow parameters and average passenger flow in multiple past preset time periods in real time; A passenger flow prediction module is used to use the average passenger flow in multiple past preset time periods as an original time series, establish a single-parameter grey prediction model using the original time series, and use the single-parameter grey prediction model to predict the average passenger flow in a future preset time period; The model building module is used to classify the load types of the subway air conditioning system load and establish a simplified physical model containing various load parameters using the multivariate linear regression method based on the average passenger flow in the future preset time period; The load parameters include short-term load, short-term load weight, long-term load, long-term load weight, dynamic load and dynamic load weight, and the short-term load weight, long-term load weight and dynamic load weight are respectively determined by the average passenger flow; The load prediction module is used to initialize the simplified physical model, input the in-and-out passenger flow parameters into the initialized simplified physical model, and predict the load of the subway air-conditioning system in a future preset time period.

Images