A method for controlling an air conditioning system of a rail transit station based on MPC

Abstract

The application discloses a kind of based on MPC's rail transit station air conditioning system control method, comprising the following steps: step 1, the average enthalpy of outdoor obtained by day-ahead prediction calculation is compared with set enthalpy, according to the comparison result to determine the operation condition of subway air conditioning system, then the switch state of different fan and air valve is controlled;Step 2, the station hall cold load is compared to determine the station hall cold load prediction value, then the station hall air supply valve opening degree value is calculated;Step 3, the cold load of sub-system is compared to determine the sub-system cold load prediction value, then the frequency of fan is calculated;Step 4, the cold load of sub-system is compared to determine the sub-system cold load prediction value, then the opening degree of group air cooling water valve is calculated.It solves the existing subway station air conditioning control system, does not associate the factors such as passenger flow, weather, departure logarithm, holiday and so on influencing load change, leading to poor system energy-saving effect.

Description

[Technical Field]

[0001] This invention belongs to the field of air conditioning system control technology, specifically relating to a control method for an air conditioning system in a rail transit station based on MPC. [Background Technology]

[0002] With the gradual implementation of dual-carbon targets and the continuous development of intelligent technologies in the rail transit industry, green and intelligent technologies have become new directions for the industry's development. As the second largest energy-consuming system in subway stations, the ventilation and air conditioning system has significant potential for energy saving due to its design conditions deviating significantly from actual operating conditions and its relatively outdated control modes.

[0003] The cooling load in subway station public areas is significantly affected by outdoor meteorological parameters, piston wind, and changes in passenger flow, resulting in substantial hourly variations. Existing solutions, whether using traditional univariate feedback regulation or purely data-driven passive algorithm control, lack mechanistic modeling. There is no zoned control for platforms and concourses, no time-based control for peak and off-peak passenger flow, no separate control for weekdays and weekends, and no separate system control for public area systems A and B. Under low-load conditions, uneven cooling distribution, over-cooling, and indoor temperatures below the lower limit of the thermal comfort zone occur between the two air-conditioned zones, leading to increased system energy consumption. [Summary of the Invention]

[0004] The purpose of this invention is to provide a control method for air conditioning systems in rail transit stations based on MPC, in order to solve the problem that existing subway station air conditioning control systems do not take into account factors that affect load changes, such as passenger flow, weather, number of train departures, and holidays, resulting in poor energy-saving performance of the system.

[0005] This invention adopts the following technical solution: a control method for an air conditioning system in a rail transit station based on MPC, comprising the following steps:

[0006] Step 1: Compare the outdoor average enthalpy value obtained from the previous day's forecast with the set enthalpy value, determine the operating conditions of the subway air conditioning system based on the comparison results, and control the on / off status of different fans and air valves in the system according to the operating conditions.

[0007] Step 2: Compare the station hall cooling load predicted at time T for each node before the day with the station hall cooling load predicted at time T during the day. Determine the predicted value of the station hall cooling load based on the comparison results, and calculate the valve opening value of the station hall air supply valve based on the predicted value of the station hall cooling load.

[0008] Step 3: Compare the subsystem cooling load predicted at time T the day before with the subsystem cooling load predicted at time T by the intraday rolling forecast. Determine the subsystem cooling load forecast value based on the comparison result, and calculate the fan frequency based on the subsystem cooling load forecast value.

[0009] Step 4: Compare the subsystem cooling load predicted at time T before the day with the subsystem cooling load predicted at time T during the day. Determine the subsystem cooling load prediction value based on the comparison result, and calculate the opening degree of the air-cooled water valve based on the subsystem cooling load prediction value.

[0010] Furthermore, the specific method for step 1 is as follows:

[0011] When the average outdoor enthalpy is greater than or equal to the set enthalpy, the subway air conditioning system will operate under the small fresh air condition.

[0012] When the average outdoor enthalpy is less than the set enthalpy:

[0013] If the average outdoor temperature is greater than or equal to the temperature threshold, the subway air conditioning system will operate under 100% fresh air condition.

[0014] If the average outdoor temperature is less than the temperature threshold, the subway air conditioning system will operate under the transitional season ventilation conditions.

[0015] Furthermore, the specific method for step 2 is as follows:

[0016] When the station hall cooling load error is less than the error threshold and the station hall monitoring temperature is less than the lower limit of the station hall thermal comfort temperature, the valve opening value of the station hall air supply valve calculated based on the station hall cooling load at time T predicted by each node before the day will be directly issued.

[0017] When the cooling load error of the station hall is less than the error threshold and the monitored temperature of the station hall is within the thermal comfort temperature range of the station hall, the air supply valve of the station hall will not operate.

[0018] When the station hall cooling load error is greater than or equal to the error threshold, the station hall air supply valve opening value is calculated using the station hall cooling load predicted on a rolling basis at time T within the day.

[0019] Among them, the station hall cooling load error is the difference between the station hall cooling load predicted at time T on a daily basis and the station hall cooling load predicted on a rolling basis at time T within the day; the station hall thermal comfort temperature range is 28 to 30℃.

[0020] Furthermore, the specific method for step 3 is as follows:

[0021] When the subsystem cooling load error is less than the error threshold and the station monitoring temperature is less than the lower limit of the station thermal comfort temperature, the frequency of the group air fan and return exhaust fan is directly issued based on the subsystem cooling load calculation at time T predicted a day before.

[0022] When the subsystem cooling load error is greater than or equal to the error threshold, the fan frequency value is calculated using the subsystem cooling load predicted at time T during the day's rolling forecast.

[0023] The subsystem cooling load error is the difference between the subsystem cooling load predicted at time T the day before and the subsystem cooling load predicted at time T during the day's rolling forecast; the platform thermal comfort temperature is 27-29℃.

[0024] Furthermore, the specific method for step 4 is as follows:

[0025] When the cooling load error of the subsystem is less than the error threshold and the real-time supply air temperature is less than the set lower limit of the supply air temperature, the opening degree of the air-cooled water valve is directly issued based on the cooling load calculation of the subsystem at time T predicted a day before.

[0026] When the cooling load error of the subsystem is greater than or equal to the error threshold, the cooling load calculation group air-cooled water valve opening value of the subsystem predicted at time T within the day shall be used.

[0027] The subsystem cooling load error is the difference between the subsystem cooling load predicted at time T a day prior and the subsystem cooling load predicted at time T within the day.

[0028] Furthermore, in step 2, the calculation method for the station hall cooling load is as follows:

[0029] Q i =(Q c，o +Q occ +Q sl +Q lgt +Q eqp ) / 1000,

[0030] In the formula, Q c，o Q occ Q sl Q lgt Q eqp These are the mechanical fresh air load, personnel load, infiltration load, lighting load, and equipment load in the station hall or platform.

[0031] Furthermore, in steps 3 and 4, the method for calculating the cooling load of the subsystem is as follows:

[0032] Q sys =θΣMAX(Q i ,0),

[0033] In the formula, θ is the cooling load ratio coefficient; Q i This refers to the cooling load of the station hall or platform.

[0034] The beneficial effects of this invention are:

[0035] 1. This invention considers multiple influencing factors and combines them with passenger flow prediction algorithms. It adopts the model predictive control (MPC) method to optimize the control of the air conditioning system of rail transit stations on the basis of the original BAS system, thereby improving the energy-saving effect of the system.

[0036] 2. This invention employs a segmented and zoned system mechanism modeling method to predict the cold load of large systems, effectively reducing the data requirements of the prediction model. It features flexible configuration, interpretability, and descriptivity, facilitating systematic understanding of system operation by maintenance and management personnel, and contributing to intelligent and efficient operation and maintenance.

[0037] 3. This invention enables system load prediction by category and zone, and achieves zoned control of air volume at the terminal of the air conditioning system based on the predicted load, thus solving the problem of poor cooling capacity distribution in the platform and concourse. Based on the subsystem load, it enables variable air volume control and group air-cooled water valve control for the A and B end air conditioning systems, solving the problem of equipment control being unrelated to dynamic load.

[0038] 4. This invention employs time-based strategy control, applying different optimized control strategies for weekdays, weekends, peak passenger flow periods, and off-peak passenger flow periods, thereby better matching the load characteristics of the large system. Through intelligent energy management of green stations, the subway air conditioning system can achieve energy-efficient, low-carbon, and low-cost operation. [Attached Image Description]

[0039] Figure 1(a) and Figure 1(b) are the passenger flow prediction results for entering and exiting the station in an embodiment of the present invention, respectively.

[0040] Figure 2 This is a predicted load diagram of a subway station concourse and platform in an embodiment of the present invention;

[0041] Figure 3 This is a schematic diagram of the architecture of an air conditioning control system in a subway station, as shown in the embodiment.

Detailed Implementation Methods

[0042] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0043] This invention provides a control method for an air conditioning system in a rail transit station based on MPC (Model Predictive Control), which is a time-sharing and zone-based system control of the air conditioning system based on a zone-specific load model. Rail transit stations generally refer to subway stations. The method specifically includes the following steps:

[0044] Step 1: Compare the outdoor average enthalpy value obtained from the previous day's forecast with the set enthalpy value, determine the operating conditions of the subway air conditioning system based on the comparison results, and control the on / off status of different fans and air valves in the system according to the operating conditions.

[0045] Step 2: Compare the station hall cooling load predicted at time T for each node before the day with the station hall cooling load predicted at time T during the day. Determine the predicted value of the station hall cooling load based on the comparison results, and calculate the valve opening value of the station hall air supply valve based on the predicted value of the station hall cooling load.

[0046] Step 3: Compare the subsystem cooling load predicted at time T the day before with the subsystem cooling load predicted at time T by the intraday rolling forecast. Determine the subsystem cooling load forecast value based on the comparison result, and calculate the fan frequency based on the subsystem cooling load forecast value.

[0047] Step 4: Compare the subsystem cooling load predicted at time T before the day with the subsystem cooling load predicted at time T during the day. Determine the subsystem cooling load prediction value based on the comparison result, and calculate the opening degree of the air-cooled water valve based on the subsystem cooling load prediction value.

[0048] In some embodiments, the specific method of step 1 is as follows:

[0049] When the average outdoor enthalpy is greater than or equal to the set enthalpy, the subway air conditioning system will operate under a small fresh air condition. For example, the set enthalpy can be 67.65 kJ / kg, and the temperature threshold can be 19.5℃.

[0050] When the average outdoor enthalpy is less than the set enthalpy:

[0051] If the average outdoor temperature is greater than or equal to the temperature threshold, the subway air conditioning system will operate under 100% fresh air condition.

[0052] If the average outdoor temperature is less than the temperature threshold, the subway air conditioning system will operate under the transitional season ventilation conditions.

[0053] In some embodiments, step 2 is specifically performed as follows:

[0054] When the station hall cooling load error is less than the error threshold and the station hall monitoring temperature is less than the lower limit of the station hall thermal comfort temperature, the valve opening value of the station hall air supply valve calculated based on the station hall cooling load at time T predicted by each node before the day will be directly issued.

[0055] When the cooling load error of the station hall is less than the error threshold and the monitored temperature of the station hall is within the thermal comfort temperature range of the station hall, the air supply valve of the station hall will not operate.

[0056] When the station hall cooling load error is greater than or equal to the error threshold, the station hall air supply valve opening value is calculated using the station hall cooling load predicted on a rolling basis at time T within the day.

[0057] The station hall cooling load error is the difference between the station hall cooling load predicted at time T on a daily basis and the station hall cooling load predicted on a rolling basis at time T within the day; the station hall thermal comfort temperature range is 28–30℃. The error threshold is set based on empirical values ​​and can be 5–10%.

[0058] In some embodiments, step 3 is specifically implemented as follows:

[0059] When the subsystem cooling load error is less than the error threshold and the station monitoring temperature is less than the lower limit of the station thermal comfort temperature, the frequency of the group air fan and return exhaust fan is directly issued based on the subsystem cooling load calculation at time T predicted a day before.

[0060] When the subsystem cooling load error is greater than or equal to the error threshold, the fan frequency value is calculated using the subsystem cooling load predicted at time T during the day's rolling forecast.

[0061] The subsystem cooling load error is the difference between the subsystem cooling load predicted at time T the day before and the subsystem cooling load predicted at time T in the intraday rolling forecast; the platform thermal comfort temperature is 27–29℃. The error threshold is set based on empirical values ​​and can be 5–10%.

[0062] The subway ventilation and air conditioning system is divided into a water system and a wind system. The wind system is further divided into a large system and a small system. The large system and the small system are each composed of an independent subsystem. An independent subsystem consists of a fresh air system, a supply air system, and a return air system. The main equipment units include: combined air handling units, return and exhaust fans, small fresh air fans, terminal two-way valves, and regulating air valves.

[0063] In some embodiments, step 4 is specifically implemented as follows:

[0064] When the cooling load error of the subsystem is less than the error threshold and the real-time supply air temperature is less than the set lower limit of the supply air temperature, the opening degree of the air-cooled water valve is directly issued based on the cooling load calculation of the subsystem at time T predicted a day before.

[0065] When the cooling load error of the subsystem is greater than or equal to the error threshold, the cooling load calculation group air-cooled water valve opening value of the subsystem predicted at time T within the day shall be used.

[0066] The subsystem's cooling load error is the difference between the subsystem's cooling load predicted at time T a day prior and the subsystem's cooling load predicted at time T within the day. The supply air temperature is set according to actual needs. The error threshold is set based on empirical values, such as 5-10%.

[0067] In some embodiments, the method for calculating the cooling load of the station hall in step 2 is as follows:

[0068] Q i =(Q c，o +Q occ +Q sl +Q lgt +Q eqp ) / 1000,

[0069] In the formula, Q c，o Q occ Q sl Q Igt Qeqp These are the mechanical fresh air load, personnel load, infiltration load, lighting load, and equipment load in the station hall or platform.

[0070] Among them, ① the calculation method for mechanical fresh air load is as follows:

[0071] The system obtains the on / off status of the fresh air unit. When the fresh air unit is on, it calculates the fresh air load. It obtains the outdoor temperature and relative humidity at time t+1, calculates the outdoor air enthalpy using the enthalpy calculation module, calculates the system load according to the fresh air load formula, and then calculates the mechanical fresh air load of the station hall and platform separately according to the distribution ratio of the air supply volume between the station hall and the platform. The calculation formulas are as follows:

[0072] Q c，o =ρ*G o (h w -h n ) / 3.6,

[0073] In the formula, G o For mechanical fresh air volume, m 3 / h; Volume of the air-conditioned area in the station hall, m 3 ρ is the outdoor air density, kg / m³ 3 h n h w The values ​​are the indoor and outdoor air enthalpy, respectively, in kJ / kg.

[0074] ②The calculation method for personnel load is as follows:

[0075] First, calculate the number of passengers on the platform and in the concourse based on the predicted passenger flow data for entering and exiting the station. Then, calculate the passenger load on the platform and in the concourse based on the load calculation formula, as follows:

[0076]

[0077] In the formula, A1 and A2 are the passenger flow entering and exiting the station per unit time, respectively; α is the peak passenger flow coefficient; a1 and a2 are the dwell time of entering passengers in the station hall and on the platform, respectively, in min; Δt is the time step (data sampling interval), in min; b1 and b2 are the dwell time of exiting passengers in the station hall and on the platform, respectively, in min;

[0078]

[0079] In the formula, q is the cluster coefficient; s q l For sensible and latent heat loss in adult males, W; G i Calculate the number of people in the station hall or platform.

[0080] ③ The calculation method for permeation load is as follows:

[0081] The system obtains the number of air changes per minute and the volume of the air-conditioned area in the station hall, and calculates the infiltration air volume at the station hall entrances and exits. It also obtains the outdoor temperature and relative humidity at time t+1, calculates the outdoor air enthalpy using the enthalpy calculation module, and then calculates the station hall infiltration load using the following formula:

[0082] G sl，hall =nV,

[0083] Q slhall =ρ*G sl，hall (h w -h c ) / 3.6,

[0084] In the formula, G slhall For the infiltration air volume into the station hall at the entrance / exit, m 3 / h; n is the number of air changes per hour; volume of the air-conditioned area in the station hall, m³. 3 ρ is the outdoor air density, kg / m³ 3 h w Outdoor air enthalpy, kJ / kg; h c The enthalpy of air in the station hall under the target temperature and humidity conditions is expressed in kJ / kg.

[0085] The logarithmic data of weekday and weekend departures are obtained from the network operation management system. Engineers configure the fitting coefficients and gap areas to calculate the air infiltration volume from the tunnel to the platform. The infiltration load of the platform screen doors is then calculated using the following formula:

[0086] G slpsd =(an 3 +bn 2 -cn+d*f / 1.7,

[0087] Q slpsd =ρ*G al *(h t -h p ) / 3.6,

[0088] In the formula, G slpsd To account for the airflow infiltration into the platform through the platform screen doors, m 3 / h; n is the number of train departures, trains / hour; a, b, c, d are polynomial fitting coefficients; f is the area of ​​the gap in the platform screen door, m² 2 ρ represents the density of the leaking air from the shielded door, in kg / m³. 3 h t The enthalpy of air in the tunnel section is expressed in kJ / kg; h p The value is the air enthalpy in the platform area, in kJ / kg.

[0089] ④ The calculation methods for equipment load and lighting load are as follows:

[0090] Based on the drawings and on-site information regarding the lighting area and the quantity of heat dissipation equipment, the engineer will configure the utilization factor and unit heat dissipation index to calculate the equipment and lighting load in the station hall and platform. The calculation formula is as follows:

[0091]

[0092] In the formula, These are the hourly usage factors for lighting and equipment, respectively; e i The heat dissipation index per unit of equipment in the i-lighting zone, W / m 2 A i Let i be the area of ​​the lighting zone, in meters. 2 ;q i i represents the unit heat dissipation index of device i, in W / unit; n i The number of cooling devices is 1 unit.

[0093] In some embodiments, the method for calculating the cooling load of the subsystem in steps 3 and 4 is as follows:

[0094]

[0095] In the formula, θ is the cooling load ratio coefficient; Q i This refers to the cooling load of the station hall or platform.

[0096] In addition, during the commissioning and testing phase of the subway air conditioning system, engineers continuously calibrated the system load models for each component and zone based on indoor temperature and humidity and actual cooling capacity to ensure that the deviation between the actual cooling capacity and the model-predicted load is less than 10% under the same indoor temperature and humidity.

[0097] Example

[0098] Step 1: Prediction of passenger flow in and out of the station and average outdoor enthalpy.

[0099] Taking a specific subway station as an example, its control system architecture is as follows: Figure 3 As shown, the part indicated by the dashed box is the original subway control system structure, including the large system control cabinet, the small system control cabinet and the water system group control cabinet. The large system control cabinet and the small system control cabinet are used to connect and control the corresponding fresh air fans, electric air valves, return and exhaust fans, air conditioning units and environmental parameter sensors. The water system group control cabinet is used to connect and control the cooling tower, electric water valves, chilled water pumps and chiller units.

[0100] Figure 3The solid-line box indicates the structure of the newly added subway control system. The new control system includes a frequency converter control cabinet for the ventilation system and a frequency converter control cabinet for the water system. The ventilation system frequency converter control cabinet connects to and controls the electrically adjustable dampers in the station hall, and also installs frequency converters for exhaust fans on all return and exhaust fans, and frequency converters for group air fans on all air conditioning units. The water system frequency converter control cabinet connects to and controls the main pipe cooling capacity meter and temperature sensor, and has frequency converters on all cooling water pumps and chilled water pumps. The new control system is connected to the edge-side control cabinet via a smart network. This edge-side control cabinet is connected to the meteorological module and the energy-saving system workstation. The edge-side control cabinet is also connected to the existing system's BAS system via a smart gateway. BAS system refers to the Building Automation System, i.e., the environmental and equipment monitoring system. The sub-item and zone-specific system load model of this invention is placed in the edge-side controller, enabling optimized control of the original system.

[0101] The specific optimization control method is as follows:

[0102] Step 1: Using a time-series prediction model LSTM based on two months of data with a 30-minute time step, the first 80% of the data was used for training, and the last 20% was used for testing. The delay step was set to 96, and the maximum number of training iterations was 200. The inbound and outbound passenger flows at time t+1 were predicted. The prediction results for inbound and outbound passenger flows are shown in Figure 1(a) and Figure 1(b). The error is less than 5%, indicating that the results of the time-series prediction model LSTM are accurate.

[0103] Based on third-party meteorological data, outdoor temperature and humidity data at time t+1 are obtained, and the predicted outdoor air enthalpy at time t+1 is obtained through the enthalpy calculation module.

[0104] Step 2: Calculate the load by region, item, and system, specifically the cooling load at time t+1.

[0105] 1) Determine the status of the fresh air unit based on the actual operating conditions. In the example, the fresh air unit is shut down and the mechanical fresh air load is 0.

[0106] 2) Calculate the passenger load in the station hall and the passenger load on the platform using the passenger load calculation formula. Then, input the passenger flow into the following formula to calculate the number of people in the station hall and on the platform at time t+1. Finally, calculate the passenger load in the station hall and on the platform using the formula.

[0107] 3) Calculate the infiltration air volume using the formula for calculating the infiltration air volume in the station hall and platform, and then obtain the infiltration load in the station hall and platform using the formula for calculating the infiltration load.

[0108] 4) Obtain the lighting load of the station hall and platform according to the lighting calculation formula, obtain the total heat dissipation load of the public area according to the equipment heat dissipation load calculation formula, and calculate the equipment load of the station hall and platform by combining the proportion of heat dissipation load of the platform and station hall equipment.

[0109] 5) Obtain the cooling load of the station hall and platform according to the zoned load calculation formula.

[0110] 6) Obtain the cooling load of the subsystem according to the subsystem load calculation formula, such as... Figure 2 The load shown exhibits a bimodal characteristic.

[0111] Step 3: Control Strategy Generation

[0112] 3.1 Calculate the valve opening based on the station hall load. If the valve opening and station hall load under rated operating conditions are ω0 and Q0 respectively, the lower limit constraint for valve opening is ω. min When the average temperature of the sensors in the station hall is below 28℃, calculate the valve opening at time t+1.

[0113] ω1=ω0*Q0 / Q1,

[0114] ω t+1 =Max(ω1, ω min ),

[0115] Where ω1 is the calculated opening degree of the station hall air valve at time t, ω t+1 Calculate the opening degree of the station hall ventilation valve at time t+1.

[0116] 3.2 Calculate the water valve opening based on the load of system A / B. For example, under rated operating conditions, the load of system A and the frequency of the blower are Q... A.0 f A.0 The lower limit constraint for valve opening is ω. min When the average temperature of the platform sensors is below 27℃, calculate the fan frequency of the blower at time t+1 in this system.

[0117] 3.3. During off-peak hours, when the air supply temperature is below 21℃, or during peak hours, when the air supply temperature is below 18℃, calculate the opening degree of the air cooling water valve at time t+1.

[0118] f A.1 =f A.0 *Q A.0 / Q A.1 ,

[0119] f A.+1 =Max(f A.1 f A.min ),

[0120]

[0121] in, Calculate the valve opening degree of the air-cooled water valve at end A at time t. Calculate the opening degree of the air-cooled water valve at end A at time t+1. Q A.0 , The cooling load and water valve opening of system A under rated operating conditions. To constrain the lower limit of valve opening for the air-cooled water valve group.

[0122] 3.4 The frequency of the return and exhaust fans is adjusted dynamically based on the frequency of the supply fans. The frequency of the return and exhaust fans at time t+1 is calculated using the frequency change rate of the supply fans.

[0123]

[0124] Among them, f Ah.t+1 Calculate the fan frequency for the return exhaust fan at point A at time t, f A.1 Calculate the fan frequency for the blower at end A at time t+1.

[0125] In summary, the subway air conditioning system is continuously optimized and controlled using the above control method with a step size as the frequency.

[0126] The fan frequencies of the air supply fans at end A and end B in the original system and this system are shown in Table 1 below. The original system refers to the air conditioning system of a subway station without an optimization strategy, while this system refers to the air conditioning system of a subway station using the control method of this invention.

[0127] Table 1. Frequency Comparison of A-end and B-end B B Ventilation Fans in the Original System and This System

[0128]

[0129] As can be seen from Table 1, the original system always maintained a fixed frequency. This system can achieve time-sharing and zone-based optimized control of the frequency based on multiple variables such as peak passenger flow and weather, which saves energy, reduces operating costs, and achieves on-demand energy supply.

[0130] In summary, this invention considers multiple influencing factors and combines passenger flow prediction algorithms and expert experience models, adopting a model-based predictive control method to replace the original single-variable feedback control method, thereby improving the system's energy-saving effect. This invention uses an expert-experience-based, segmented, and zoned system mechanism modeling method to predict the cooling load of the large system, effectively reducing the data requirements of the prediction model; it is flexible in configuration, interpretable, and describable, facilitating operation and maintenance personnel to systematically grasp the system's operating status and contributing to intelligent and efficient operation and maintenance. This invention achieves segmented and zoned system load prediction, realizing zoned control of the air conditioning system's terminal air volume based on zoned predicted load, solving the problem of poor cooling distribution in platforms and concourses; it also realizes variable air volume control and group air-cooled water valve control for A-end and B-end air conditioning systems based on subsystem load, solving the problem of equipment control being unrelated to dynamic load. This invention adopts time-based strategy control, employing different optimized control strategies for weekdays, weekends, peak passenger flow periods, and off-peak passenger flow periods, thereby better matching the load characteristics of the large system; through green station intelligent energy management, it achieves energy-efficient, low-carbon, and low-cost operation of the subway air conditioning system.

Claims

1. A method for controlling an air conditioning system of a rail transit station based on MPC, characterized in that, The method comprises the following steps: Step 1, comparing the outdoor average enthalpy value calculated by the day-ahead prediction with the set enthalpy value, determining the operation condition of the subway air conditioning system according to the comparison result, and controlling the on-off state of different fans and air valves in the system according to the operation condition; Step 2, comparing the station hall cold load at T time predicted by the day-ahead node-by-node prediction with the station hall cold load at T time predicted by the day-ahead rolling prediction, determining the station hall cold load prediction value according to the comparison result, and calculating the station hall air supply valve opening value according to the station hall cold load prediction value; The specific method of step 2 is: When the station hall cold load error is less than the error threshold value and the station hall monitoring temperature is less than the lower limit value of the station hall thermal comfort temperature, the station hall air supply valve opening value calculated based on the station hall cold load at T time predicted by the day-ahead node-by-node prediction is directly issued; When the station hall cold load error is less than the error threshold value and the station hall monitoring temperature is within the station hall thermal comfort temperature range, the station hall air supply valve is not actuated; When the station hall cold load error is greater than or equal to the error threshold value, the station hall air supply valve opening value is calculated based on the station hall cold load at T time predicted by the day-ahead rolling prediction; Wherein, the station hall cold load error is the difference between the station hall cold load at T time predicted by the day-ahead node-by-node prediction and the station hall cold load at T time predicted by the day-ahead rolling prediction; and the station hall thermal comfort temperature range is 28-30℃. The calculation method of the station hall cold load is: ， wherein, , , , , are the mechanical fresh air load, the personnel load, the infiltration load, the lighting load and the equipment load of the station hall or platform, respectively; Step 3, comparing the subsystem cold load at T time predicted by the day-ahead prediction with the subsystem cold load at T time predicted by the day-ahead rolling prediction, determining the subsystem cold load prediction value according to the comparison result, and calculating the fan frequency according to the subsystem cold load prediction value; Step 4, comparing the subsystem cold load at T time predicted by the day-ahead prediction with the subsystem cold load at T time predicted by the day-ahead rolling prediction, determining the subsystem cold load prediction value according to the comparison result, and calculating the opening of the group air cooling water valve according to the subsystem cold load prediction value.

2. The MPC-based control method for an air conditioning system of a rail transit station according to claim 1, wherein, The specific method of step 1 is: When the outdoor average enthalpy value is greater than or equal to the set enthalpy value, the subway air conditioning system operates in the small fresh air condition; When the outdoor average enthalpy value is less than the set enthalpy value: If the outdoor average temperature is greater than or equal to the temperature threshold value, the subway air conditioning system operates in the full fresh air condition; If the outdoor average temperature is less than the temperature threshold value, the subway air conditioning system operates in the transition season ventilation condition.

3. The MPC-based control method for air conditioning system in rail transit station according to claim 1, wherein, The specific method of step 3 is: When the subsystem cold load error is less than the error threshold value and the platform monitoring temperature is less than the lower limit value of the platform thermal comfort temperature, the frequency of the group air fan and the return air fan calculated based on the subsystem cold load at T time predicted by the day-ahead prediction is directly issued; When the subsystem cold load error is greater than or equal to the error threshold value, the fan frequency value is calculated based on the subsystem cold load at T time predicted by the day-ahead rolling prediction; Wherein, the subsystem cold load error is the difference between the subsystem cold load at T time predicted by the day-ahead prediction and the subsystem cold load at T time predicted by the day-ahead rolling prediction; and the platform thermal comfort temperature is 27-29℃.

4. The MPC-based control method for air conditioning system in rail transit station according to claim 1, characterized in that, The specific method of step 4 is: When the subsystem cold load error is less than the error threshold value and the real-time air supply temperature is less than the lower limit value of the set air supply temperature, the opening of the group air cooling water valve calculated based on the subsystem cold load at T time predicted by the day-ahead prediction is directly issued; When the error of the cold load of the sub-system is greater than or equal to the error threshold, the valve opening value of the group air cooling water is calculated according to the predicted cold load of the sub-system at the time T in the day. The error of the cold load of the sub-system is the difference between the predicted cold load of the sub-system at the time T in the day and the predicted cold load of the sub-system at the time T in the day.

5. The MPC-based control method for air conditioning system in rail transit station according to claim 1, characterized in that, In the step 3 and the step 4, the calculation method of the cold load of the sub-system is as follows: ， In the formula, is a cold load ratio coefficient; is the cold load of the station hall or platform.

Images