CN115854501A - Airport terminal room temperature large-lag prediction control method based on passenger flow prediction - Google Patents
Airport terminal room temperature large-lag prediction control method based on passenger flow prediction Download PDFInfo
- Publication number
- CN115854501A CN115854501A CN202310032559.1A CN202310032559A CN115854501A CN 115854501 A CN115854501 A CN 115854501A CN 202310032559 A CN202310032559 A CN 202310032559A CN 115854501 A CN115854501 A CN 115854501A
- Authority
- CN
- China
- Prior art keywords
- passenger
- prediction
- room temperature
- model
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000009826 distribution Methods 0.000 claims abstract description 54
- 238000004378 air conditioning Methods 0.000 claims abstract description 23
- 238000010438 heat treatment Methods 0.000 claims abstract description 17
- 238000013528 artificial neural network Methods 0.000 claims abstract description 14
- 238000009423 ventilation Methods 0.000 claims abstract description 10
- 239000012530 fluid Substances 0.000 claims abstract description 7
- 230000015654 memory Effects 0.000 claims abstract description 6
- 230000006870 function Effects 0.000 claims description 17
- 238000005457 optimization Methods 0.000 claims description 15
- 230000007787 long-term memory Effects 0.000 claims description 14
- 230000006403 short-term memory Effects 0.000 claims description 11
- 238000012546 transfer Methods 0.000 claims description 11
- 239000002245 particle Substances 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 7
- 230000009471 action Effects 0.000 claims description 6
- 230000010006 flight Effects 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 4
- 230000007246 mechanism Effects 0.000 claims description 4
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 claims description 3
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 claims description 3
- 230000004913 activation Effects 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 230000014759 maintenance of location Effects 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000000306 recurrent effect Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000009827 uniform distribution Methods 0.000 claims description 3
- 238000011161 development Methods 0.000 abstract description 5
- 229910052799 carbon Inorganic materials 0.000 abstract description 4
- 238000005265 energy consumption Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 3
- 238000011217 control strategy Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a passenger flow prediction-based method for controlling the large-lag prediction of the room temperature of an airport terminal, which comprises the following steps: s1, building a probability model of airport terminal passenger arrival based on chi-square distribution, and realizing passenger flow space-time distribution prediction of a terminal passenger streamline by applying a fluid dynamics principle; s2, establishing a room temperature large-lag prediction model of the airport terminal based on the long-short term memory neural network and the prediction result of passenger space-time distribution; and S3, establishing a model prediction theory of the heating, ventilating and air conditioning system of the airport terminal building based on the model prediction control theory and the room temperature large lag prediction result, and giving an optimal input sequence of control variables of the heating, ventilating and air conditioning system. The method predicts the passenger flow volume of the terminal building through the flight dynamic information, and uses the passenger flow volume as an input parameter for predicting the room temperature of the terminal building, thereby realizing the energy-saving control of the heating, ventilation and air conditioning system of the terminal building, and providing an important technical support for friendly, convenient, green, low-carbon, intelligent and efficient development of the airport terminal building.
Description
Technical Field
The invention relates to the technical field of intelligent control of an airport terminal environment control system, in particular to a method for predicting and controlling the room temperature of an airport terminal based on passenger flow prediction.
Background
The airport terminal building is used as an important urban traffic hub and has important strategic significance on the sustainable development and urbanization construction of cities. Due to the characteristics of multiple functions of a service system, large personnel flow scale, long annual running time and the like, the terminal buildings become important places for energy consumption, the average energy consumption intensity of the terminal buildings is 2.9 times of that of common public buildings and 8.0 times of that of town residential buildings, and the terminal buildings belong to typical high-energy-consumption and high-emission buildings (DOI: 10.1016/j.buildenv.2019.03.011; DOI:10.1016/j.scs.2021.103619; DOI:10.1016/j.enbenv.2022.06.006; DOI: 10.1016/j.buildenv.2018.02.009). As important equipment for energy consumption, the heating, ventilation and air conditioning system occupies 40-80% of the energy consumption of the whole airport terminal, and the implementation of an efficient and energy-saving environment control strategy becomes an urgent need for green and low-carbon development of the airport terminal. Due to the fact that the space structure is coherent, the system scale is large, personnel flow is severe, the influence factors of the indoor environment of the airport terminal building are many, the hysteresis effect is strong, and the dynamic change rule is difficult to represent by using a traditional mechanism modeling method. According to the existing research results, fresh air, illumination and equipment loads in the energy consumption of the station building heating, ventilating and air conditioning system are closely related to personnel behaviors. (DOI: 10.1016/j.buildenv.2019.03.011; DOI:10.1016/j.scs.2021.103619; DOI: 10.1016/j.buildenv.2018.02.009). In 2018, the cold load index of the terminal building is estimated by the actual passenger flow density in Liuhua university and the actual cold load is only 31% of the design value. In 2022, the Beijing architectural design institute, kuo-Liang, simulated the energy consumption of the Daxing airport terminal using the passenger flow density estimated in the divided areas, and showed that the simulated value was 11.3% smaller than the design value. However, because of the difficulty of forecasting the passenger flow distribution of the terminal, the above research only introduces the passenger flow density of a certain area or the whole terminal into the energy consumption evaluation stage, which proves that the passenger flow volume is helpful for the energy-saving operation of the terminal, and does not really introduce the passenger flow distribution forecasting into the terminal environment control system to dig the correlation between the passenger flow of the terminal and the indoor environment. Therefore, a new method must be developed from a control theory and a technical scheme, a machine learning intelligent prediction method is adopted to predict the passenger flow of the airport terminal building, the indoor environmental parameter prediction of the terminal building is further realized, a reference basis is provided for the control of a heating ventilation air conditioning system, and the method is an effective means for solving the problem of the control of the overlarge space environment of the terminal building.
Disclosure of Invention
Aiming at the problem of complex regulation and control of the station building heating ventilation air conditioning system caused by nonlinearity, large lag, randomness and distribution of the indoor temperature of the station building, the invention provides the airport station building room temperature large lag prediction control method based on passenger flow prediction, which promotes the energy-saving intelligent operation of the station building environment control system, by adopting a statistical analysis theory and a machine learning method. The invention mainly establishes a probability model of airport terminal passenger arrival through chi-square distribution, realizes the prediction of the space-time distribution of the airport terminal passengers, establishes an indoor environmental parameter prediction model of the airport terminal by utilizing a long-short term memory cyclic neural network, and finally realizes the intelligent control of the airport terminal heating, ventilating and air conditioning system by adopting a model prediction control method. The establishment of a suitable indoor environment with minimum energy consumption is an important target for the control of the heating, ventilation and air conditioning system, and the indoor temperature is a key index for representing the indoor environment comfort degree of the terminal building and influencing the energy consumption of the heating, ventilation and air conditioning system. The intelligent energy-saving control system of the airport terminal building promotes the energy-saving intelligent operation of the environment control system of the airport terminal building, and provides an important technical support for the friendly, convenient, green, low-carbon, intelligent and efficient development of the airport terminal building.
The technical scheme of the invention is as follows:
a large-lag prediction control method for room temperature of an airport terminal based on passenger flow prediction comprises the following steps:
s1, forecasting the space-time distribution of passengers in an airport terminal: based on chi-square distribution, an airport terminal passenger arrival probability model is established, and the passenger flow space-time distribution prediction of a terminal passenger streamline is realized by applying a fluid dynamics thought, and the method specifically comprises the following steps:
s1.1, station building passenger port probability chi-square distribution model
Extracting flight information and passenger security check information, wherein the earliest time and the latest time of arrival of passengers are respectively represented as t EA And t LA (ii) a Setting the sampling interval as epsilon, counting the number of the safety check people of each flight passenger, and converting the port-reaching percentage of the corresponding flight passenger; introducing a transformation factor, and fitting the passenger port probability by adopting chi-square distribution;
wherein f (t) represents the passenger port percentage; Γ (·) represents a gamma function; t represents the passenger arrival time; t is t SD 、t EA And t LA Respectively representing scheduled departure time of flights and earliest and latest arrival time of passengers; d is a degree of freedom; s is a transform factor;
s1.2, space-time distribution prediction model of passengers in terminal building
Establishing a passenger flow space-time distribution prediction model of a passenger flow line of the terminal building based on a terminal building passenger probability chi-square distribution model by applying a fluid dynamics thought; setting the prediction range of passenger space-time distribution as 24 hours, adopting relative time in one day by the model, and assuming that passengers board to obey uniform distribution;
wherein, Z j A number representing the jth spatial cell; g f,i A gate number indicating flight i;
representing the number of passengers in the jth space unit at the time t; c f,i Representing the passenger capacity of flight i; l is a radical of an alcohol in,j And L out,j The distance between the inlet and the outlet of the jth space unit and the security inspection channel is shown; m is the total number of flights in the prediction range; p is the passenger attendance; v is the average pace of the passenger; g (t) is a passenger boarding probability distribution model; t is t SB And t EB Representing a start boarding time and an end boarding time;
if gate G of flight i f,i In space unit Z j If the passenger is in the middle, the passenger enters the space unit until the passenger boarding and leaving; if gate G of flight i f,i Not in spatial cell Zj, then the passenger only passes through spatial cell; starting boarding time t of flight SB Possibly earlier than the latest arrival time t of passengers LA 。
S1.3, passenger space-time distribution prediction model identification and calibration
The prediction model of passenger space-time distribution comprises passenger attendance rate p, passenger average pace speed v and passenger boarding starting time t SB And a time t of boarding EB Identifying and calibrating four unknown parameters by using a model; defining evaluation indexes of passenger space-time distribution prediction model, including root mean square error RMSE, mean absolute error MAPE and correlation index R 2 (ii) a Monitoring the actual passenger flow of the terminal building, and solving unknown parameters of the model by adopting a particle swarm optimization algorithm;
wherein the content of the first and second substances,
and &>
Respectively representing the position and velocity of the alpha particle in the tau iteration; />
And gbest τ Representing the individual optima and the global optima in the τ th iteration; a is 1 And a 2 Is a learning factor; r is 1 And r 2 Is a random number between 0 and 1; w is the inertial weight; y is t And &>
Respectively representing the real-time value and the average value of the actual passenger flow of the terminal building; err represents an error vector of the actual passenger flow of the terminal building and the model predicted value; n represents the number of samples;
s2, predicting the large room temperature hysteresis of the airport terminal: establishing a room temperature large-lag prediction model of the airport terminal based on the long-short term memory neural network and the prediction result of passenger space-time distribution, and specifically comprising the following steps:
s2.1, analysis of cause of temperature lag of building in airport terminal
Evaluating the degree of association between each influence factor and the room temperature of the terminal building by adopting a transfer entropy analysis method, and providing guarantee for reasonably selecting input parameters of a room temperature neural network prediction model;
TE(X→Y)＝MI(X - ；Y + |Y - ) (9)
MI(X - ；Y + |Y - )＝H(X - ,Y - )+H(Y + ,Y - )-H(Y - )-H(Y + ,X - ,Y - ) (10)
X＝[x 1 ,…x i ,…x n ] (12)
wherein, TE, MI (: B; respectively representing transfer entropy, mutual information, condition mutual information and information entropy; x and Y represent system input variables and output variables; y is + Represents the future result of the output variable Y; x - And Y - Past observations representing an input variable X and an output variable Y;
the larger the transfer entropy is, the more sufficient the information transfer between the influence factors and the prediction variables is, and the closer the correlation degree is;
s2.2, station building room temperature neural network prediction model
Based on the room temperature hysteresis cause, establishing a big room temperature hysteresis prediction model of the terminal building by using a long and short term memory network; the long-term and short-term memory network is a variant of a recurrent neural network and consists of a forgetting gate, an input gate and an output gate, and a gate control mechanism of the long-term and short-term memory network can control the retention or the abandonment of information, so that the long-term and short-term memory network has long-term memory capability;
Γ f,t ＝σ(W f h t-1 +U f X i,t ) (13)
Γ i,t ＝σ(W i h t-1 +U i X i,t ) (14)
Γ o,t ＝σ(W o h t-1 +U o X i,t )
h t ＝Γ o,t ⊙tanh(c t )
Y o,t ＝g(W p h t ) (17)
err t ＝T t -Y o,t (18)
wherein, gamma is f,t 、Γ i,t And Γ o,t Respectively show a forgetting door,Outputs of the input gate and the output gate;
c t and h t Representing candidate cell state, cell state and implicit state, respectively; x is the number of i,t 、Y o,t 、T t And err t Respectively representing input parameters, output parameters, target parameters and error vectors of the circulation unit; w f 、W i 、W c 、W o 、U f 、U i 、U c 、U o And W p Respectively representing weight matrixes of the forgetting gate, the input gate, the output gate and the output unit; σ (-) and g (-) denote sigmoid and linear activation function, respectively; as indicates the Hadamard product;
s3, room temperature model prediction control of the airport terminal: establishing a model prediction theory of the heating, ventilation and air conditioning system of the airport terminal building based on a model prediction control theory and a room temperature large-lag prediction result, and specifically comprising the following steps of:
s3.1, forecasting and controlling cost function of room temperature of terminal building
The room temperature prediction control cost function can be defined as the weight sum of tracking error and control action, so that the room temperature prediction value is close to the set value on one hand, and the fluctuation range of the input variable is ensured to be as small as possible on the other hand;
fit(τ)＝||Q(Y set,t -Y o,t )|| 2 +||RX i,t || 2 (19)
wherein, Y set,t To output a set value; q and R represent the tracking error weight and the control action weight respectively;
s3.2, on-line optimization of airport terminal room temperature prediction control
And solving the optimal control sequence of the heating, ventilation and air conditioning system by adopting a particle swarm optimization algorithm and taking the minimum cost function as an optimization target.
The method is also suitable for indoor temperature prediction control of transportation hubs such as railway stations, high-speed railway stations, bus stations and the like.
Compared with the prior art, the invention has the beneficial effects that: the indoor temperature prediction control method of the terminal building related to flight and passenger information is provided, the passenger flow volume of the terminal building is predicted through flight dynamic information, and the predicted passenger flow volume is used as an input parameter for predicting the indoor temperature of the terminal building, so that the energy-saving control of a heating, ventilating and air-conditioning system of the terminal building is realized, and an important technical support is provided for friendly, convenient, green, low-carbon and intelligent efficient development of airport terminal buildings.
Drawings
Fig. 1 is a block diagram of a large-lag prediction control method for the room temperature of a terminal building based on passenger flow prediction.
Fig. 2 is a logic diagram of the large-lag prediction control of the terminal building room temperature based on passenger flow prediction.
Fig. 3 is a flow chart of a station building room temperature large-lag predictive control method based on passenger flow prediction.
FIG. 4 is a block diagram of a probability chi-square distribution model of airport terminal passenger arrival.
Fig. 5 is a schematic diagram of the analysis of the room temperature hysteresis cause of the airport terminal.
Fig. 6 is a schematic diagram of a room temperature large-lag prediction model of an airport terminal.
Fig. 7 is a predictive control schematic diagram of an airport terminal building heating, ventilating and air conditioning system.
Detailed Description
The following detailed description of the invention will be made in conjunction with the accompanying drawings and equations that describe the summary of the invention.
Referring to fig. 1-3, the invention is a method for predicting and controlling the large delay of the room temperature of an airport terminal based on passenger flow prediction, taking the large delay prediction and control of the room temperature of a certain airport terminal in Guangzhou city as an example, the method comprises the following steps:
s1, forecasting the space-time distribution of passengers in an airport terminal: referring to fig. 3, based on chi-square distribution, an airport terminal building passenger arrival probability model is established, and the passenger flow space-time distribution prediction of the terminal building passenger streamline is realized by applying the fluid dynamics thought, and the method comprises the following specific steps:
s1.1, station building passenger arrival probability chi-square distribution model
The passenger flow of the airport terminal is closely related to the time of a flight shift, and the airport terminal has remarkable planning and predictability; referring to FIG. 4, flight information and passenger security information are extracted and storedThe earliest and latest arrival time of the passenger are 400 minutes and 20 minutes before the predicted departure time of the flight respectively, namely t SD -t EA ＝400min，t SD -t LA =20min; setting the sampling interval epsilon as 5 minutes, counting the number of passengers for security check of each flight, and analyzing the association relation between the arrival probability of passengers at the terminal and the number of flights; adopting a chi-square distribution probability density function to establish a station building passenger port-arrival probability model;
wherein f (t) represents the percentage of passengers arriving in the port; Γ (·) represents a gamma function; t represents the passenger arrival time; t is t EA And t LA Respectively representing the earliest and latest arrival time of passengers; d is a degree of freedom; s is a transform factor;
the card square distribution model freedom degree d and the conversion factor s of an airport terminal in Guangzhou city are respectively 7 and 0.0974;
s1.2, space-time distribution prediction model of passengers in terminal building
Establishing a passenger flow space-time distribution prediction model of a station building passenger streamline based on a station building passenger arrival probability chi-square distribution model by using a fluid dynamics thought; setting the prediction range of the passenger time-space distribution to be 24 hours, adopting the relative time of a day by the model, and assuming that the passenger boarding obeys uniform distribution;
wherein Z is j A number representing the jth spatial cell; g f,i A gate number indicating flight i;
denotes the jth of time tThe number of passengers in a spatial cell; c f,i Representing the passenger capacity of flight i; l is in,j And L out,j The distance between the inlet and the outlet of the jth space unit and the security inspection channel is shown; m is the total number of flights in the prediction range; p is the passenger attendance; v is the average pace of the passenger; g (t) is a passenger boarding probability distribution model; t is t SB And t EB Representing a start boarding time and an end boarding time;
if gate G of flight i f,i In space unit Z j If the passenger is in the middle, the passenger enters the space unit until the passenger boarding and leaving; if gate G of flight i f,i Out of space unit Z j In, then the passenger only passes through the space cell; starting boarding time t of flight SB Possibly earlier than the latest arrival time t of passengers LA Therefore, the passenger boarding probability distribution cannot be simplified to g (t) = 1/(t) EB -t SB )；
S1.3, passenger space-time distribution prediction model identification and calibration
The prediction model of passenger space-time distribution comprises passenger attendance rate p, passenger average pace speed v and passenger boarding starting time t SB And a time t of boarding EB Identifying and calibrating four unknown parameters by using a model; defining evaluation indexes of passenger space-time distribution prediction model, including root mean square error RMSE, mean absolute error MAPE and correlation index R 2 (ii) a Monitoring the actual passenger flow of the terminal building, and solving unknown parameters of the model by adopting a particle swarm optimization algorithm;
wherein the content of the first and second substances,
and &>
Represents the position and velocity of the alpha particle in the tau iteration; />
And gbest τ Representing the individual optima and the global optima in the τ th iteration; a is 1 And a 2 For learning factor, 2 can be taken; r is 1 And r 2 Is a random number between 0 and 1; w is the inertial weight; y is t And &>
Respectively representing the real-time value and the average value of the actual passenger flow of the terminal building; err represents an error vector of the actual passenger flow of the terminal building and the model predicted value; n represents the number of samples;
the Wi-Fi indoor positioning technology is adopted by a certain airport terminal building in Guangzhou city to calibrate the assumed parameters of a passenger space-time distribution prediction model, the passenger attendance rate p is 0.84, the passenger average pace v is 1.21m/s, and the passenger boarding starting time t SB And a time t of boarding EB 42 minutes and 23 minutes before flight take-off, respectively;
s2, predicting the large room temperature lag of the airport terminal building: referring to fig. 3, a room temperature large-lag prediction model of the airport terminal is established based on the long-short term memory neural network and the prediction result of the passenger space-time distribution, and the method comprises the following specific steps:
s2.1, analysis of cause of temperature lag of building in airport terminal
Referring to fig. 5, as the envelope of the station building is transparent, personnel flow is violent, and the system scale is huge, the factors of temperature lag of the station building are numerous; evaluating the degree of association between each influence factor and the room temperature of the terminal building by adopting a transfer entropy analysis method, and providing guarantee for reasonably selecting input parameters of a room temperature neural network prediction model;
TE(X→Y)＝MI(X - ；Y + |Y - ) (9)
MI(X - ；Y + |Y - )＝H(X - ,Y - )+H(Y + ,Y - )-H(Y - )-H(Y + ,X - ,Y - ) (10)
X＝[x 1 ,…x i ,…x n ] (12)
wherein, TE, MI (:; the delivery entropy, the mutual information, the condition mutual information and the information entropy are respectively expressed by H; x and Y represent system input variables and output variables; y is + Represents the future result of the output variable Y; x - And Y - Past observations representing an input variable X and an output variable Y;
the larger the transfer entropy is, the more sufficient the information transfer between the influence factors and the prediction variables is, and the closer the correlation degree is; the room temperature of a certain airport terminal building in Guangzhou city is delayed, so that the passenger flow, the outdoor temperature, the solar radiation intensity, the air supply temperature of an air conditioning system and the air supply volume of the air conditioning system are caused;
s2.2, station building room temperature neural network prediction model
Referring to fig. 6, a large-lag prediction model of the airport terminal room temperature is established by using a long-short term memory network based on the lag cause of the room temperature; the long-term and short-term memory network is a variant of a recurrent neural network and consists of a forgetting gate, an input gate and an output gate, and a gate control mechanism of the long-term and short-term memory network can control the retention or the abandonment of information, so that the long-term and short-term memory network has long-term memory capability;
Γ f,t ＝σ(W f h t-1 +U f X i,t ) (13)
Γ i,t ＝σ(W i h t-1 +U i X i,t ) (14)
Γ o,t ＝σ(W o h t-1 +U o X i,t )
h t ＝Γ o,t ⊙tanh(c t )
Y o,t ＝g(W p h t ) (17)
err t ＝T t -Y o,t (18)
wherein, gamma is f,t 、Γ i,t And Γ o,t Respectively representing the outputs of the forgetting gate, the input gate and the output gate;
c t and h t Representing candidate cell state, cell state and implicit state, respectively; x i,t 、Y o,t 、T t And err t Respectively representing input parameters, output parameters, target parameters and error vectors of the circulation unit; w f 、W i 、W c 、W o 、U f 、U i 、U c 、U o And W p Respectively representing weight matrixes of the forgetting gate, the input gate, the output gate and the output unit; σ (-) and g (-) denote sigmoid and linear activation function, respectively; as indicates the Hadamard product;
the root mean square error of the room temperature prediction result of an airport terminal in Guangzhou city is less than 0.5 ℃, and the prediction control requirement of a central air-conditioning system is met;
s3, room temperature model prediction control of the airport terminal: referring to fig. 3, the model prediction theory of the central air-conditioning system of the airport terminal building is established based on the model prediction control theory and the room temperature large hysteresis prediction result, and the specific steps are as follows:
s3.1, forecasting and controlling cost function of room temperature of terminal building
Model predictive control has been widely used in the heating, ventilation and air conditioning field because of the handling of time-varying, nonlinear, constrained optimization problems. The room temperature prediction control cost function can be defined as the weight sum of tracking error and control action, so that the room temperature prediction value is close to the set value on one hand, and the fluctuation range of the input variable is ensured to be as small as possible on the other hand;
fit(τ)＝||Q(Y set,t -Y o,t )|| 2 +||RX i,t || 2 (19)
wherein, Y set,t To output a set value; q and R represent the tracking error weight and the control action weight respectively;
the input variable of the indoor temperature model predictive control of a certain airport terminal in Guangzhou city is the air output of the central air conditioning system;
s3.2, on-line optimization of airport terminal room temperature prediction control
Referring to fig. 7, model predictive control is a rolling optimization control strategy, and online optimization is a key link for ensuring room temperature predictive control effect; and similarly, a particle swarm optimization algorithm is adopted, the minimum cost function is taken as an optimization target, and the solved air output of the central air-conditioning system can enable the room temperature variation range of an airport terminal building in Guangzhou city to be maintained within +/-0.5 ℃ of the set value.
Claims (1)
1. The method for predicting and controlling the room temperature of the airport terminal building based on passenger flow prediction is characterized by comprising the following steps of:
s1, forecasting the space-time distribution of passengers in an airport terminal: based on chi-square distribution, an airport terminal passenger arrival probability model is established, and the passenger flow space-time distribution prediction of a terminal passenger streamline is realized by applying a fluid dynamics thought, and the method specifically comprises the following steps:
s1.1, station building passenger arrival probability chi-square distribution model
Extracting flight information and passenger security check informationThe early and late arrival times are denoted t EA And t LA (ii) a Setting the sampling interval as epsilon, counting the number of the passengers in the security check of each flight, and converting the port-returning percentage of the corresponding flight; introducing a transformation factor, and fitting the passenger port probability by adopting chi-square distribution;
wherein f (t) represents the percentage of passengers arriving in the port; Γ (·) represents a gamma function; t represents the passenger arrival time; t is t SD 、t EA And t LA Respectively representing scheduled departure time of flights and earliest and latest arrival time of passengers; d is the degree of freedom; s is a transformation factor;
s1.2, space-time distribution prediction model of passengers in terminal building
Establishing a passenger flow space-time distribution prediction model of a station building passenger streamline based on a station building passenger arrival probability chi-square distribution model by using a fluid dynamics thought; setting the prediction range of the passenger time-space distribution to be 24 hours, adopting the relative time of a day by the model, and assuming that the passenger boarding obeys uniform distribution;
wherein Z is j A number representing the jth spatial cell; g f,i A gate number indicating flight i;
representing the number of passengers in the jth space unit at the time t; c f,i Representing the passenger capacity of flight i; l is in,j And L out,j The distance between the inlet and the outlet of the jth space unit and the security inspection channel is shown; m is the prediction rangeTotal number of flights in; p is the passenger attendance; v is the average pace of the passenger; g (t) is a passenger boarding probability distribution model; t is t SB And t EB Representing a start boarding time and an end boarding time;
if gate G of flight i f,i In space unit Z j If the passenger is in the middle, the passenger enters the space unit until the passenger boarding and leaving; if gate G of flight i f,i Out of space unit Z j In, then the passenger only passes through the space cell; starting boarding time t of flight SB Possibly earlier than the latest arrival time t of passengers LA ；
S1.3, passenger space-time distribution prediction model identification and calibration
The prediction model of passenger space-time distribution comprises passenger attendance rate p, passenger average pace speed v and passenger boarding starting time t SB And a time t of boarding EB Identifying and calibrating four unknown parameters by using a model; defining evaluation indexes of passenger space-time distribution prediction model, including root mean square error RMSE, mean absolute error MAPE and correlation index R 2 (ii) a Monitoring the actual passenger flow of the terminal building, and solving unknown parameters of the model by adopting a particle swarm optimization algorithm;
wherein the content of the first and second substances,
and &>
Respectively representing the position and the speed of the alpha particle in the tau iteration; />
And gbest τ Representing the individual optima and the global optima in the τ th iteration; a is 1 And a 2 Is a learning factor; r is 1 And r 2 Is a random number between 0 and 1; w is the inertial weight; y is t And &>
Respectively representing the real-time value and the average value of the actual passenger flow of the terminal building; err represents an error vector of the actual passenger flow of the terminal building and the model predicted value; n represents the number of samples;
s2, predicting the large room temperature hysteresis of the airport terminal: establishing a room temperature large-lag prediction model of the airport terminal based on the long-short term memory neural network and the prediction result of passenger space-time distribution, and specifically comprising the following steps:
s2.1, analysis of cause of temperature lag of building in airport terminal
Evaluating the degree of association between each influence factor and the room temperature of the terminal building by adopting a transfer entropy analysis method, and providing guarantee for reasonably selecting input parameters of a room temperature neural network prediction model;
TE(X→Y)＝MI(X - ；Y + |Y - ) (9)
MI(X - ；Y + |Y - )＝H(X - ,Y - )+H(Y + ,Y - )-H(Y - )-H(Y + ,X - ,Y - ) (10)
X＝[x 1 ,…x i ,…x n ] (12)
wherein, TE, MI (: B; respectively representing transfer entropy, mutual information, condition mutual information and information entropy; x and Y represent system input variables and output variables; y is + Represents the future outcome of the output variable Y; x - And Y - Past observations representing an input variable X and an output variable Y;
the larger the transfer entropy is, the more sufficient the information transfer between the influence factors and the prediction variables is, and the closer the correlation degree is;
s2.2, station building room temperature neural network prediction model
Based on the room temperature hysteresis cause, establishing a big room temperature hysteresis prediction model of the terminal building by using a long and short term memory network; the long-term and short-term memory network is a variant of a recurrent neural network and consists of a forgetting gate, an input gate and an output gate, and a gate control mechanism of the long-term and short-term memory network can control the retention or the discard of information, so that the long-term and short-term memory network has long-term memory capability;
Γ f,t ＝σ(W f h t-1 +U f X i,t ) (13)
Γ i,t ＝σ(W i h t-1 +U i X i,t ) (14)
Γ o,t ＝σ(W o h t-1 +U o X i,t )
h t ＝Γ o,t ⊙tanh(c t )
Y o,t ＝g(W p h t ) (17)
err t ＝T t -Y o,t (18)
wherein, gamma is f,t 、Γ i,t And Γ o,t Respectively representing the outputs of the forgetting gate, the input gate and the output gate;
c t and h t Representing candidate cell state, cell state and implicit state, respectively; x i,t 、Y o,t 、T t And err t Respectively representing input parameters, output parameters, target parameters and error vectors of the circulation unit; w f 、W i 、W c 、W o 、U f 、U i 、U c 、U o And W p Respectively representing weight matrixes of the forgetting gate, the input gate, the output gate and the output unit; σ (-) and g (-) denote sigmoid and linear activation function, respectively; as indicates the Hadamard product;
s3, room temperature model prediction control of the airport terminal: establishing a model prediction theory of the heating, ventilation and air conditioning system of the airport terminal building based on a model prediction control theory and a room temperature large-lag prediction result, and specifically comprising the following steps of:
s3.1, forecasting and controlling cost function of room temperature of terminal building
The room temperature prediction control cost function can be defined as the weight sum of tracking error and control action, so that the room temperature prediction value is close to the set value on one hand, and the fluctuation range of the input variable is ensured to be as small as possible on the other hand;
fit(τ)＝||Q(Y set,t -Y o,t )|| 2 +||RX i,t || 2 (19)
wherein, Y set,t To output a set value; q and R represent the tracking error weight and the control action weight respectively;
s3.2, on-line optimization of airport terminal room temperature prediction control
And solving the optimal control sequence of the heating, ventilating and air conditioning system by adopting a particle swarm optimization algorithm and taking the minimum cost function as an optimization target.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310032559.1A CN115854501B (en) | 2023-01-10 | 2023-01-10 | Airport terminal room temperature large hysteresis prediction control method based on passenger flow prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310032559.1A CN115854501B (en) | 2023-01-10 | 2023-01-10 | Airport terminal room temperature large hysteresis prediction control method based on passenger flow prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115854501A true CN115854501A (en) | 2023-03-28 |
| CN115854501B CN115854501B (en) | 2024-03-19 |
Family
ID=85657240
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310032559.1A Active CN115854501B (en) | 2023-01-10 | 2023-01-10 | Airport terminal room temperature large hysteresis prediction control method based on passenger flow prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115854501B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109130767A (en) * | 2017-06-28 | 2019-01-04 | 北京交通大学 | The intelligent control method of rail traffic station ventilation and air conditioning system based on passenger flow |
| CN112783044A (en) * | 2020-12-31 | 2021-05-11 | 新奥数能科技有限公司 | Energy control system and energy control method |
| WO2022025819A1 (en) * | 2020-07-27 | 2022-02-03 | Hitachi, Ltd. | System and method of controlling an air-conditioning and/or heating system |
| CN114492967A (en) * | 2022-01-17 | 2022-05-13 | 河海大学 | Urban rail station passenger flow prediction method and medium based on CEEMDAN and BLSTM combined model |
| CN115130312A (en) * | 2022-07-07 | 2022-09-30 | 大连理工大学 | Heating ventilation air conditioning system lag phase estimation method based on information theory framework data driving |
| CN115407658A (en) * | 2022-08-29 | 2022-11-29 | 广东机场白云信息科技有限公司 | Method, device and medium for determining causal relationship between heating, ventilating and air conditioning system time sequences |
-
2023
- 2023-01-10 CN CN202310032559.1A patent/CN115854501B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109130767A (en) * | 2017-06-28 | 2019-01-04 | 北京交通大学 | The intelligent control method of rail traffic station ventilation and air conditioning system based on passenger flow |
| WO2022025819A1 (en) * | 2020-07-27 | 2022-02-03 | Hitachi, Ltd. | System and method of controlling an air-conditioning and/or heating system |
| CN112783044A (en) * | 2020-12-31 | 2021-05-11 | 新奥数能科技有限公司 | Energy control system and energy control method |
| CN114492967A (en) * | 2022-01-17 | 2022-05-13 | 河海大学 | Urban rail station passenger flow prediction method and medium based on CEEMDAN and BLSTM combined model |
| CN115130312A (en) * | 2022-07-07 | 2022-09-30 | 大连理工大学 | Heating ventilation air conditioning system lag phase estimation method based on information theory framework data driving |
| CN115407658A (en) * | 2022-08-29 | 2022-11-29 | 广东机场白云信息科技有限公司 | Method, device and medium for determining causal relationship between heating, ventilating and air conditioning system time sequences |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115854501B (en) | 2024-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Deb et al. | Using artificial neural networks to assess HVAC related energy saving in retrofitted office buildings | |
| Ben-Nakhi et al. | Cooling load prediction for buildings using general regression neural networks | |
| CN107942960B (en) | A kind of intelligentized information processing system | |
| Kwok et al. | An intelligent approach to assessing the effect of building occupancy on building cooling load prediction | |
| Kumar et al. | Energy analysis of a building using artificial neural network: A review | |
| CN110332647B (en) | Load prediction method for air conditioning system of underground station of subway and air conditioning system | |
| Kim et al. | Sequence-to-sequence deep learning model for building energy consumption prediction with dynamic simulation modeling | |
| CN107239874A (en) | A kind of quality of power supply and energy-saving analysis system towards track traffic | |
| CN112434787A (en) | Terminal space energy consumption prediction method based on building total energy consumption, medium and equipment | |
| Song et al. | An indoor temperature prediction framework based on hierarchical attention gated recurrent unit model for energy efficient buildings | |
| Alamin et al. | An Artificial Neural Network (ANN) model to predict the electric load profile for an HVAC system | |
| CN112734101B (en) | Sharing bicycle intelligent allocation method based on vehicle demand prediction | |
| CN108197404A (en) | A kind of building load Forecasting Methodology based on time hereditary capacity | |
| Wani et al. | Estimating thermal parameters of a commercial building: A meta-heuristic approach | |
| Zhao et al. | An artificial intelligence (AI)-driven method for forecasting cooling and heating loads in office buildings by integrating building thermal load characteristics | |
| CN116204954A (en) | Air terminal building fresh air energy-saving optimization control method based on passenger space-time distribution prediction | |
| CN115983487B (en) | Airport terminal passenger space-time distribution prediction method based on chi-square distribution | |
| Godahewa et al. | Simulation and optimisation of air conditioning systems using machine learning | |
| CN115854501A (en) | Airport terminal room temperature large-lag prediction control method based on passenger flow prediction | |
| Li et al. | Passenger spatiotemporal distribution prediction in airport terminals based on insect intelligent building architecture and its contribution to fresh air energy saving | |
| Guo et al. | Research on short-term traffic demand of taxi in large cities based on BP neural network algorithm | |
| CN114200839A (en) | Office building energy consumption intelligent control model for dynamic monitoring of coupled environmental behaviors | |
| Pombeiro et al. | Linear, fuzzy and neural networks models for definition of baseline consumption: Early findings from two test beds in a University campus in Portugal | |
| Junghans et al. | Introduction of a plug and play model predictive control to predict room temperatures | |
| Liu et al. | A hybrid model of AR and PNN method for building thermal load forecasting |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |