CN113821902A - Active disturbance rejection control system for static optimization of central air-conditioning refrigeration station - Google Patents

Active disturbance rejection control system for static optimization of central air-conditioning refrigeration station Download PDF

Info

Publication number
CN113821902A
CN113821902A CN202110672979.7A CN202110672979A CN113821902A CN 113821902 A CN113821902 A CN 113821902A CN 202110672979 A CN202110672979 A CN 202110672979A CN 113821902 A CN113821902 A CN 113821902A
Authority
CN
China
Prior art keywords
model
unit
central air
optimization
refrigerating
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
Application number
CN202110672979.7A
Other languages
Chinese (zh)
Other versions
CN113821902B (en
Inventor
李壮举
郭虹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing University of Civil Engineering and Architecture
Original Assignee
Individual
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to CN202110672979.7A priority Critical patent/CN113821902B/en
Publication of CN113821902A publication Critical patent/CN113821902A/en
Application granted granted Critical
Publication of CN113821902B publication Critical patent/CN113821902B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/06—Power analysis or power optimisation

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Air Conditioning Control Device (AREA)

Abstract

The invention discloses an active disturbance rejection control system for static optimization of a central air-conditioning refrigeration station, wherein an air-conditioning load prediction unit is used for predicting load through a BP (back propagation) neural network; the central air-conditioning refrigeration station model calculation unit is used for executing one of a single model modeling mode and a multi-model fusion modeling mode; the system working point optimization unit is used for calculating the optimal working point of the system by using an optimization algorithm; and the central control refrigeration station control unit is used for tracking control by adopting an active disturbance rejection controller. The invention also discloses an active disturbance rejection control method for the static optimization of the central air-conditioning refrigeration station. The system and the method determine the optimal working point of the system closest to the actual value by accurately establishing a static model and predicting the cold load, and adopt the active disturbance rejection controller to carry out tracking control on the optimal working point of the system, thereby realizing the optimization and energy conservation of the central air-conditioning refrigeration station.

Description

Active disturbance rejection control system for static optimization of central air-conditioning refrigeration station
Technical Field
The invention relates to the field of automatic control, in particular to an active disturbance rejection control system and method for static optimization of a central air-conditioning refrigeration station.
Background
At present, various large shopping malls, shopping supermarkets, office buildings and the like are distributed in various large and small cities across the country. The large public building is a high-energy-consumption building different from other civil buildings, and naturally becomes a key project for energy conservation and emission reduction transformation of an environmental protection department and an electric power department. In large public buildings, the energy consumption of a central air conditioning (HVAC) system generally accounts for over 50% of the energy consumption of the entire building, while the energy consumption of a central air conditioning refrigeration station accounts for over 70% of the energy consumption of the central air conditioning system. At present, most of heating ventilation air-conditioning systems in China are in an inefficient operation state, so that the operation efficiency is low and the energy waste is serious. The reason for this phenomenon is manifold, because the central air-conditioning unit operates under partial load most of the time, this causes the operating condition of large flow, little temperature difference to appear in most central air-conditioning refrigerating systems, can't make the air conditioner in the best operating condition, cause the enormous loss of energy; in the aspect of control, the main machine and part of tail end devices of the central air conditioner are provided with automatic control devices, but the automatic control level is low, and when external influence factors change, energy is wasted due to too slow adjustment time of a control system.
Disclosure of Invention
To solve the above problems, the present invention is achieved by the following means.
The invention provides an active disturbance rejection control system for static optimization of a central air-conditioning refrigeration station, which comprises: the system comprises an air conditioner load prediction unit, a central air conditioner refrigerating station model calculation unit, a system working point optimization unit, a central control refrigerating station control unit and a central air conditioner refrigerating station;
the air conditioner load prediction unit is used for predicting the load through a BP neural network;
the central air-conditioning refrigeration station model calculation unit is used for executing one of a single model modeling mode and a multi-model fusion modeling mode;
the system working point optimization unit is used for calculating the optimal working point of the system by using an optimization algorithm;
and the central control refrigeration station control unit is used for tracking control by adopting an active disturbance rejection controller.
Preferably, the central air-conditioning refrigeration station comprises four subsystems, specifically a chilled water system, a refrigeration unit, a cooling water system and a cooling tower.
Preferably, the central air-conditioning refrigeration station model calculation unit adopts weighted multi-model fusion modeling to establish a static energy consumption model of the system for the refrigeration unit with stronger nonlinearity; and for a chilled water system, a cooling water system and a cooling tower, a single model modeling mode is adopted.
Preferably, the modeling process of the multi-model fusion model of the refrigerating unit executed by the central air-conditioning refrigerating station model calculation unit specifically includes:
(1) dividing the working points of the refrigeration unit samples;
(2) establishing a sub-model of the refrigerating unit;
(3) and performing weighted fusion on the sub-models.
Preferably, the sample operating points of the refrigeration unit are specifically divided into: and clustering the modeling data by using a k-means algorithm, dividing the modeling data into 3 groups, randomly selecting 3 objects as initial clustering centers, calculating the distance between each object and each sub-clustering center, and allocating each object to the nearest clustering center. The cluster centers and the objects assigned to them represent a cluster.
Preferably, the specific processing of the central air-conditioning refrigeration station model calculation unit on the refrigeration unit with stronger nonlinearity is as follows: establishing a static energy consumption model of the system by adopting weighted multi-model fusion modeling, and predicting the obtained refrigerating capacity Q according to the load so as to reduce the energy consumption of the refrigerating unit
Figure 161325DEST_PATH_IMAGE001
The minimum is taken as a target, and the outlet water temperature of the chilled water is calculated by utilizing an optimization algorithm
Figure 657028DEST_PATH_IMAGE002
And the outlet temperature of the cooling water
Figure 186230DEST_PATH_IMAGE003
The value of (a).
Preferably, unlike the conventional dynamic optimization system, the active disturbance rejection control system for the static optimization of the central air conditioning refrigeration station performs data acquisition and optimization calculation processing at long time intervals without real-time processing.
The invention also provides an active disturbance rejection control method for the static optimization of the central air-conditioning refrigeration station, which specifically comprises the following steps:
step 1, predicting the refrigerating capacity Q obtained by accurate load prediction of a central air conditioner through a BP neural network, and controlling a refrigerating unit to generate enough refrigerating capacity according to the refrigerating capacity Q obtained by load prediction;
2, calculating variable parameters by using different models aiming at different equipment or systems, and establishing a static energy consumption model of the system by adopting weighted multi-model fusion modeling for a refrigerating unit with stronger nonlinearity;
step 3, performing variable calculation on the refrigerating water system and the cooling water system by adopting a single model modeling mode;
step 4, for the cooling tower, performing variable calculation by adopting a single model modeling mode;
step 5, verifying the variables obtained in the step 2-4 according to the objective function and the constraint condition to finally obtain a calculated optimal working point;
and 6, tracking the optimal working point of the control system by adopting the active disturbance rejection controller.
Due to the adoption of the technical scheme, the invention can achieve the following beneficial effects: the system and the method determine the optimal working point of the system closest to the actual value by accurately establishing a static model and predicting the cold load, and adopt the active disturbance rejection controller to carry out tracking control on the optimal working point of the system, thereby realizing the optimization and energy conservation of the central air-conditioning refrigeration station. The system and the method greatly reduce the workload of the processor, and the used optimization method is relatively simple and has small calculation amount, thereby greatly reducing the load of the system and improving the processing speed and the use efficiency.
Drawings
FIG. 1 is a schematic diagram of an active disturbance rejection control system;
FIG. 2 is a schematic diagram of the operation of the air conditioning water circulation system;
FIG. 3 is a block diagram of the optimization control of the optimal operating point of the central air conditioning refrigeration system;
FIG. 4 is a comparison graph of the predicted value-actual value of the neural network of the measured building cold load;
FIG. 5 is a scatter plot of the k-means algorithm clustering of modeled data;
fig. 6 is a flowchart of an active disturbance rejection control method.
FIG. 7 is a block diagram of the structure of the active disturbance rejection control process
These drawings and written description are not intended to limit the scope of the inventive concepts in any manner, but rather to illustrate the inventive concepts to those skilled in the art by reference to specific embodiments.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
An embodiment of the present invention provides an active disturbance rejection control system for static optimization of a central air conditioning refrigeration station, as shown in fig. 1, the system includes: the system comprises an air conditioner load prediction unit, a central air conditioner refrigerating station model calculation unit, a system working point optimization unit, a central control refrigerating station control unit and a central air conditioner refrigerating station. The central air-conditioning refrigeration station comprises four subsystems, specifically a chilled water system, a refrigeration unit, a cooling water system and a cooling tower.
And the air conditioner load prediction unit is used for predicting the load through a BP neural network.
The central air-conditioning refrigeration station model calculation unit is used for executing one of a single model modeling mode and a multi-model fusion modeling mode; for a refrigerating unit with stronger nonlinearity, a static energy consumption model of the system is established by adopting weighted multi-model fusion modeling; and for a chilled water system, a cooling water system and a cooling tower, a single model modeling mode is adopted.
And the system working point optimization unit is used for calculating the optimal working point of the system by utilizing an optimization algorithm.
And the central control refrigeration station control unit is used for tracking control by adopting an active disturbance rejection controller.
As shown in fig. 2, the air conditioning water circulation system is composed of a refrigerating unit, a chilled water pump, a cooling tower, and the like. The refrigerating unit mainly has the functions of generating cold, providing cold for chilled water through the heat absorption/release function of a refrigerant to supply to the tail end of an air conditioner, releasing heat for the chilled water and dissipating heat through a cooling tower.
The chilled water exchanges heat with the refrigerant of the refrigeration unit in the evaporator through a heat exchanger.
The cooling water exchanges heat with the refrigerant of the refrigeration unit in the condenser through a heat exchanger.
The cooling water dissipates heat to the atmosphere through a fan of the cooling tower.
The refrigerant absorbs the heat of the chilled water in the evaporator of the refrigerating unit through gasification, so that the temperature of the chilled water is reduced, the chilled water absorbs heat at the tail end of the air conditioner through the circulation of the chilled water pump and flows back to the evaporator to release heat, and the cold energy is transmitted from the refrigerating unit to the tail end of the air conditioner.
The refrigerant gasified in the evaporator is compressed into high-pressure high-temperature gas when passing through a compressor of the refrigerating unit, is changed into low-temperature low-pressure gas by cooling water when passing through a condenser, is changed into low-temperature high-pressure liquid again after passing through an expansion valve, and circularly enters the evaporator to be gasified, thereby completing one Carnot cycle.
The heat generated by the refrigerating unit exchanges heat with cooling water in the condenser, is circularly conveyed to the cooling tower through the cooling water pump, and is dissipated to the atmosphere through the cooling tower fan.
In order to achieve the optimization goal, the active disturbance rejection control system for the static optimization of the central air-conditioning refrigeration station completes the optimization control of the optimal working point of the central air-conditioning refrigeration system through the processing of the air-conditioning load prediction unit, the central air-conditioning refrigeration station model calculation unit, the system working point optimization unit and the central control refrigeration station control unit in sequence, as shown in fig. 3.
In addition, different from the traditional dynamic optimization system, the active disturbance rejection control system for the static optimization of the central air-conditioning refrigeration station carries out data acquisition and optimization calculation processing every longer time, such as 20-30 minutes, does not need real-time processing, and greatly reduces the workload of a processor. In addition, the optimization method used by the active disturbance rejection control system is relatively simple, the calculation amount is small, the load of the system is reduced to a great extent, and the processing speed and the use efficiency are improved.
The method specifically comprises the following steps: the air conditioner load prediction unit predicts the refrigerating capacity Q obtained by accurate load prediction of the central air conditioner through a BP neural network, and controls the refrigerating unit to generate enough refrigerating capacity according to the refrigerating capacity Q obtained by load prediction so as to meet the refrigerating capacity requirement at the tail end of the air conditioner and the loss in the refrigerating capacity transportation process.
The air conditioner load prediction unit firstly collects environmental parameters and system parameters, sets BP neural network parameters, and obtains refrigerating capacity Q obtained by load prediction through BP neural network prediction according to the collected parameters and the set parameters.
The environment parameters comprise outdoor dry bulb temperature, outdoor solar radiation, outdoor relative humidity and the like, the system parameters are a current actual load value Q, and the BP neural network parameters comprise the number n of neurons in an input layer, the number m of neurons in an output layer and the number l of neurons in a hidden layer.
For example, the number of input layer neurons is n = 4; the number of neurons in the output layer is m = 1; taking the coefficient a =10, the number of the neurons in the hidden layer is obtained
Figure 573525DEST_PATH_IMAGE004
。
For load prediction, a BP neural network prediction method is generally used. The BP neural network is a prediction method with high prediction accuracy, but simultaneously, the response speed of the BP neural network is prolonged due to the complexity of a modeling process of the BP neural network, however, for short-term load prediction, the prediction accuracy of the BP neural network is high, and the response speed is also within the bearing range of general engineering application.
After BP neural network prediction using the specific parameters of the inputs mentioned above, the rootThe daily load prediction accuracy rate is obtained according to calculation
Figure 162769DEST_PATH_IMAGE005
Percent daily pass point
Figure 880189DEST_PATH_IMAGE006
As shown in fig. 4.
And the central air-conditioning refrigeration station model calculation unit calculates variable parameters by using different models aiming at different equipment or systems.
The central air-conditioning refrigeration station model calculation unit specifically comprises a chilled water system calculation module, a cooling tower calculation module and a refrigeration unit calculation module.
The method specifically comprises the following steps:
for the refrigerating unit with stronger nonlinearity, the refrigerating unit calculation module utilizes the multi-model fusion model (1) of the refrigerating unit, adopts the weighted multi-model fusion modeling to establish a static energy consumption model of the system, predicts the obtained refrigerating capacity Q according to the load and uses the energy consumption of the refrigerating unit
Figure 213082DEST_PATH_IMAGE001
The minimum is taken as a target, and the outlet water temperature of the chilled water is calculated by utilizing an optimization algorithm
Figure 463672DEST_PATH_IMAGE002
And the outlet temperature of the cooling water
Figure 223818DEST_PATH_IMAGE003
The value of (a).
Figure 428534DEST_PATH_IMAGE007
（1）
Wherein w, m and n are fitting parameters,
Figure 565117DEST_PATH_IMAGE008
is the average of the power in the sample data of the three submodels,
Figure 226693DEST_PATH_IMAGE018
，
Figure 434558DEST_PATH_IMAGE019
，
Figure 385196DEST_PATH_IMAGE020
and fitting parameters in the sub-model weighting process.
In addition, the first and second substrates are,
Figure 298795DEST_PATH_IMAGE012
the model is a DOE-2 model, the model is a performance curve model, the performance curve is a mathematical model mainly obtained by taking the input/output relationship of the refrigerating unit, only parameter fitting is needed to be carried out according to the operation sample data of the refrigerating unit, the modeling method is similar to that of an artificial neural network, but an accurate mathematical expression of the model can be obtained, and therefore, the optimal control of the refrigerating unit is easier to carry out. The DOE-2 energy consumption model selected herein is shown as formula (2):
Figure 759863DEST_PATH_IMAGE013
（2）
wherein a, b, c, d, e and f are fitting parameters,
Figure 360346DEST_PATH_IMAGE014
is the temperature of the outlet water of the chilled water.
The modeling process of the multi-model fusion model of the refrigerating unit specifically comprises the following steps:
(1) and dividing the working points of the refrigeration unit samples.
As shown in FIG. 5, the invention selects a k-means algorithm to cluster the modeling data, divides the modeling data into 3 groups, randomly selects 3 objects as initial clustering centers, then calculates the distance between each object and each sub-clustering center, and assigns each object to the clustering center closest to the object. The cluster centers and the objects assigned to them represent a cluster.
(2) Establishing a sub-model of a refrigerating unit
Establishing a sub-model of the refrigerating unit by using the DOE-2 model method, setting the maximum iteration number to be 1000 by using a Marquardt algorithm and a general global optimization algorithm according to the divided three working condition sample data, and obtaining three sub-models by using the DOE-2 energy consumption model formula (2)
Figure 539655DEST_PATH_IMAGE015
、
Figure 18041DEST_PATH_IMAGE016
And
Figure 333615DEST_PATH_IMAGE017
。
(3) sub-model weighted fusion
Setting submodels
Figure 373509DEST_PATH_IMAGE015
、
Figure 774535DEST_PATH_IMAGE016
And
Figure 56611DEST_PATH_IMAGE017
fitting parameters are w, m and n, introducing the concept of sample center, considering the difference value between the total sample and the center of the subsample, giving different weights to each subsample according to the distance between the total sample and the center of the subsample, and fitting the coefficient of the sample center
Figure 226693DEST_PATH_IMAGE018
、
Figure 434558DEST_PATH_IMAGE019
、
Figure 385196DEST_PATH_IMAGE020
. And then obtaining an energy consumption model to be fitted, then bringing all sample data of the refrigerating unit into the energy consumption model, and fitting by using a particle swarm algorithm to obtain a multi-model fusion model (1) of the refrigerating unit.
For chilled water systems, the chillingThe calculation module of the frozen water system utilizes a single model (3) according to the flow of the frozen water
Figure 997782DEST_PATH_IMAGE026
And the outlet water temperature of the chilled water
Figure 657028DEST_PATH_IMAGE002
Energy consumption of chilled water system
Figure 72551DEST_PATH_IMAGE025
Minimum as target, calculating flow of chilled water
Figure 997782DEST_PATH_IMAGE026
And the outlet water temperature of the chilled water
Figure 657028DEST_PATH_IMAGE002
。
Figure 21419DEST_PATH_IMAGE024
(3)
Wherein, a, b, c, d, e and f are fitting parameters, and the energy consumption of the chilled water system is used
Figure 72551DEST_PATH_IMAGE025
The minimum target is based on the preset
Figure 997782DEST_PATH_IMAGE026
And
Figure 658308DEST_PATH_IMAGE002
and a reference range, and finding a best match value close to the reference value by calculation and experiment.
For cooling water systems, the cooling water system calculation module uses a single model (4) according to the flow of cooling water
Figure 358836DEST_PATH_IMAGE032
And the outlet water temperature of the cooling water
Figure 186230DEST_PATH_IMAGE003
Cooling water system energy consumption
Figure 471575DEST_PATH_IMAGE031
Minimum target, calculating the flow of cooling water
Figure 358836DEST_PATH_IMAGE032
And the outlet water temperature of the cooling water
Figure 186230DEST_PATH_IMAGE003
。
Figure 220798DEST_PATH_IMAGE030
(4)
Wherein a, b, c, d, e and f are fitting parameters, and the energy consumption of the cooling water system is used
Figure 792725DEST_PATH_IMAGE031
The minimum target is based on the preset
Figure 358836DEST_PATH_IMAGE032
And
Figure 96722DEST_PATH_IMAGE003
and a reference range, and finding a best match value close to the reference value by calculation and experiment.
For cooling towers, the cooling tower calculation module uses a single model (5) based on the temperature difference of the cooling water
Figure 208246DEST_PATH_IMAGE038
Air flow rate influenced by cooling tower fan
Figure 719782DEST_PATH_IMAGE034
Energy consumption of cooling tower
Figure 27266DEST_PATH_IMAGE035
Minimum target, calculate coolingTower fan affected air flow
Figure 719782DEST_PATH_IMAGE034
Figure 357327DEST_PATH_IMAGE037
(5)
Wherein a, b, c, d, e and f are fitting parameters, and the energy consumption of the cooling tower is used
Figure 903846DEST_PATH_IMAGE035
The minimum target is based on the preset
Figure 749442DEST_PATH_IMAGE034
And
Figure 760123DEST_PATH_IMAGE038
and a reference range, and finding a best match value close to the reference value by calculation and experiment.
In addition, the system working point optimization unit is used for verifying a plurality of variables obtained by the central air-conditioning refrigeration station model calculation unit according to an objective function and a constraint condition to finally obtain a calculated optimal working point so as to realize the optimization and energy conservation of the central air-conditioning refrigeration station.
Wherein the variables include: refrigerating capacity Q obtained by load prediction, chilled water supply/discharge temperature
Figure 164298DEST_PATH_IMAGE039
、
Figure 198113DEST_PATH_IMAGE040
Temperature of cooling water supply/discharge
Figure 847400DEST_PATH_IMAGE041
、
Figure 650271DEST_PATH_IMAGE042
(ii) a Flow rate of chilled water
Figure 228276DEST_PATH_IMAGE026
(ii) a Flow rate of cooling water
Figure 749387DEST_PATH_IMAGE032
(ii) a Air flow rate
Figure 202365DEST_PATH_IMAGE034
。
An objective function of
Figure 859743DEST_PATH_IMAGE043
The minimum is that the sum of the energy consumption of each link is minimum.
The constraint conditions are determined by the performance of each element of the refrigerating unit, and specifically comprise the following steps:
Figure 605720DEST_PATH_IMAGE044
Figure 614127DEST_PATH_IMAGE045
Figure 605217DEST_PATH_IMAGE046
the optimization algorithm for calculating the optimal working point mainly comprises a Particle Swarm Optimization (PSO), an Ant Colony Optimization (ACO), an artificial bee colony Algorithm (ABC), a Bacterial Foraging Optimization (BFO), a Firefly Algorithm (FA) and the like.
And finally obtaining the numerical values of the variables corresponding to the optimal working point.
And the central control refrigeration station control unit is used for tracking and controlling the optimal working point of the system by adopting the active disturbance rejection controller.
Dynamic simulation analysis shows that when the anti-interference performance of the control system is consistent, the predictive control is often high in response speed and small in overshoot, and is especially suitable for unstable objects. However, when the model is disturbed, the control effect of the predictive control is influenced to some extent. The active disturbance rejection controller has the advantages of having a strong disturbance rejection effect for a central air conditioning system with strong disturbance, and therefore having strong robustness. Since the central air-conditioning system studied in the text has the characteristic of strong disturbance, an active disturbance rejection controller is adopted to track the optimal working point of the control system so as to realize the optimization and energy saving of the central air-conditioning system.
The invention also provides an active disturbance rejection control method for static optimization of a central air-conditioning refrigeration station, as shown in fig. 6, the method specifically includes:
step 1, predicting the refrigerating capacity Q obtained by accurate load prediction of the central air conditioner through a BP neural network, and controlling a refrigerating unit to generate enough refrigerating capacity according to the refrigerating capacity Q obtained by load prediction so as to meet the refrigerating capacity requirement at the tail end of the air conditioner and the loss in the refrigerating capacity transportation process.
The method specifically comprises the following steps:
the method comprises the steps of firstly collecting environmental parameters and system parameters, setting BP neural network parameters, and obtaining refrigerating capacity Q obtained by load prediction through BP neural network prediction according to the collected parameters and the set parameters.
The environment parameters comprise outdoor dry bulb temperature, outdoor solar radiation, outdoor relative humidity and the like, the system parameters are a current actual load value Q, and the BP neural network parameters comprise the number n of neurons in an input layer, the number m of neurons in an output layer and the number l of neurons in a hidden layer.
For example, the number of input layer neurons is n = 4; the number of neurons in the output layer is m = 1; taking the coefficient a =10, the number of the neurons in the hidden layer is obtained
Figure 382680DEST_PATH_IMAGE047
。
The BP neural network is a prediction method with the highest prediction accuracy, but at the same time, the response speed of the BP neural network is prolonged due to the complexity of a modeling process of the BP neural network, but for short-term and medium-term load prediction, the prediction accuracy of the BP neural network is high, and the response speed is within the bearing range of general engineering application.
2, calculating variable parameters by using different models aiming at different equipment or systems, wherein the different models are one of a single model modeling mode and a multi-model fusion modeling mode; and for the refrigerating unit with stronger nonlinearity, a static energy consumption model of the system is established by adopting weighted multi-model fusion modeling.
The method specifically comprises the following steps: for the refrigerating unit with stronger nonlinearity, a multi-model fusion model (1) of the refrigerating unit is utilized, a static energy consumption model of the system is established by adopting weighted multi-model fusion modeling, the obtained refrigerating capacity Q is predicted according to the load, and the energy consumption of the refrigerating unit is used
Figure 161325DEST_PATH_IMAGE001
The minimum is taken as a target, and the outlet water temperature of the chilled water is calculated by utilizing an optimization algorithm
Figure 657028DEST_PATH_IMAGE002
And the outlet temperature of the cooling water
Figure 186230DEST_PATH_IMAGE003
The value of (a).
Figure 327393DEST_PATH_IMAGE049
（1）
Wherein w, m and n are fitting parameters,
Figure 959362DEST_PATH_IMAGE050
is the average of the power in the sample data of the three submodels,
Figure 226693DEST_PATH_IMAGE018
，
Figure 434558DEST_PATH_IMAGE019
，
Figure 385196DEST_PATH_IMAGE020
and fitting parameters in the sub-model weighting process.
In addition, the first and second substrates are,
Figure 115089DEST_PATH_IMAGE012
is a DOE-2 model which is a performance curve modelThe performance curve is a mathematical model mainly obtained by the input/output relationship of the refrigerating unit, only parameter fitting needs to be carried out according to the operation sample data of the refrigerating unit, the modeling method is similar to an artificial neural network, but an accurate model mathematical expression can be obtained, and therefore the optimization control of the refrigerating unit is easier to carry out. The DOE-2 energy consumption model selected herein is shown as formula (2):
Figure 671972DEST_PATH_IMAGE051
（2）
wherein a, b, c, d, e and f are fitting parameters,
Figure 569697DEST_PATH_IMAGE052
is the temperature of the outlet water of the chilled water.
The modeling process of the multi-model fusion model of the refrigerating unit specifically comprises the following steps:
(1) and dividing the working points of the refrigeration unit samples.
The invention selects a k-means algorithm to cluster modeling data, divides the modeling data into 3 groups, randomly selects 3 objects as initial clustering centers, then calculates the distance between each object and each sub-clustering center, and allocates each object to the nearest clustering center. The cluster centers and the objects assigned to them represent a cluster.
(2) Establishing a sub-model of a refrigerating unit
Establishing a sub-model of the refrigerating unit by using the DOE-2 model method, setting the maximum iteration number to be 1000 by using a Marquardt algorithm and a general global optimization algorithm according to the divided three working condition sample data, and obtaining three sub-models by using the DOE-2 energy consumption model formula (2)
Figure 373509DEST_PATH_IMAGE015
、
Figure 774535DEST_PATH_IMAGE016
And
Figure 56611DEST_PATH_IMAGE017
。
(3) sub-model weighted fusion
Setting submodels
Figure 231174DEST_PATH_IMAGE056
、
Figure 109132DEST_PATH_IMAGE057
And
Figure 570200DEST_PATH_IMAGE058
fitting parameters are w, m and n, introducing the concept of sample center, considering the difference value between the total sample and the center of the subsample, giving different weights to each subsample according to the distance between the total sample and the center of the subsample, and fitting the coefficient of the sample center
Figure 44DEST_PATH_IMAGE018
、
Figure 618501DEST_PATH_IMAGE019
、
Figure 628045DEST_PATH_IMAGE020
. And then obtaining an energy consumption model to be fitted, then bringing all sample data of the refrigerating unit into the energy consumption model, and fitting by using a particle swarm algorithm to obtain a multi-model fusion model (1) of the refrigerating unit.
And 3, performing variable calculation on the refrigerating water system and the cooling water system by adopting a single model modeling mode.
The method specifically comprises the following steps: for chilled water systems, the chilled water system calculation module utilizes a single model (3) based on the flow of chilled water
Figure 997782DEST_PATH_IMAGE026
And the outlet water temperature of the chilled water
Figure 657028DEST_PATH_IMAGE002
Energy consumption of chilled water system
Figure 401725DEST_PATH_IMAGE025
Minimum as target, calculating flow of chilled water
Figure 997782DEST_PATH_IMAGE026
And the outlet water temperature of the chilled water
Figure 657028DEST_PATH_IMAGE002
。
Figure 808677DEST_PATH_IMAGE024
(3)
Wherein, a, b, c, d, e and f are fitting parameters, and the energy consumption of the chilled water system is used
Figure 401725DEST_PATH_IMAGE025
The minimum target is based on the preset
Figure 18651DEST_PATH_IMAGE026
And
Figure 512081DEST_PATH_IMAGE002
and a reference range, and finding a best match value close to the reference value by calculation and experiment.
For cooling water systems, the cooling water system calculation module uses a single model (4) according to the flow of cooling water
Figure 358836DEST_PATH_IMAGE032
And the outlet water temperature of the cooling water
Figure 186230DEST_PATH_IMAGE003
Cooling water system energy consumption
Figure 471575DEST_PATH_IMAGE031
Minimum target, calculating the flow of cooling water
Figure 358836DEST_PATH_IMAGE032
And the outlet water temperature of the cooling water
Figure 186230DEST_PATH_IMAGE003
。
Figure 808119DEST_PATH_IMAGE030
(4)
Wherein a, b, c, d, e and f are fitting parameters, and the energy consumption of the cooling water system is used
Figure 471575DEST_PATH_IMAGE031
The minimum target is based on the preset
Figure 470755DEST_PATH_IMAGE032
And
Figure 896051DEST_PATH_IMAGE003
and a reference range, and finding a best match value close to the reference value by calculation and experiment.
And 4, performing variable calculation on the cooling tower in a single model modeling mode.
For cooling towers, the cooling tower calculation module uses a single model (5) based on the temperature difference of the cooling water
Figure 208246DEST_PATH_IMAGE038
Air flow rate influenced by cooling tower fan
Figure 507216DEST_PATH_IMAGE034
Energy consumption of cooling tower
Figure 157454DEST_PATH_IMAGE035
Minimum target, calculate air flow for cooling tower fan effect
Figure 719782DEST_PATH_IMAGE034
。
Figure 591343DEST_PATH_IMAGE037
(5)
Wherein a, b, c, d, e and f are fitting parameters, and the energy consumption of the cooling tower is used
Figure 157454DEST_PATH_IMAGE035
The minimum target is based on the preset
Figure 909989DEST_PATH_IMAGE034
And
Figure 208246DEST_PATH_IMAGE038
and a reference range, and finding a best match value close to the reference value by calculation and experiment.
And 5, verifying the variables obtained in the steps 2-4 according to the objective function and the constraint condition to finally obtain the optimal working point so as to realize the optimization and energy conservation of the central air-conditioning refrigeration station.
Wherein the variables include: refrigerating capacity Q obtained by load prediction, chilled water supply/discharge temperature
Figure 533048DEST_PATH_IMAGE039
、
Figure 840533DEST_PATH_IMAGE040
Temperature of cooling water supply/discharge
Figure 167347DEST_PATH_IMAGE041
、
Figure 902085DEST_PATH_IMAGE042
(ii) a Flow rate of chilled water
Figure 448604DEST_PATH_IMAGE026
(ii) a Flow rate of cooling water
Figure 559779DEST_PATH_IMAGE032
(ii) a Air flow rate
Figure 9608DEST_PATH_IMAGE034
。
An objective function of
Figure 915248DEST_PATH_IMAGE043
The minimum is that the sum of the energy consumption of each link is minimum.
The constraint conditions are determined by the performance of each element of the refrigerating unit, and specifically comprise the following steps:
Figure 949063DEST_PATH_IMAGE060
Figure 660667DEST_PATH_IMAGE045
Figure 463538DEST_PATH_IMAGE061
the optimization algorithm for calculating the optimal working point mainly comprises a Particle Swarm Optimization (PSO), an Ant Colony Optimization (ACO), an artificial bee colony Algorithm (ABC), a Bacterial Foraging Optimization (BFO), a Firefly Algorithm (FA) and the like.
And finally obtaining the numerical values of the variables corresponding to the optimal working point.
And 6, tracking the optimal working point of the control system by adopting the active disturbance rejection controller.
The active disturbance rejection controller arranges the transition process of the system by adding a differential tracker TD at the input end, so that the set value of the system is increased in a curve form, and overshoot caused by the error between the set value and the output value is effectively eliminated; the extended state observer is used for monitoring the internal disturbance and the external disturbance of the system in real time, regarding the internal disturbance and the external disturbance of the system as a total disturbance, expanding the total disturbance into a new state variable and acting on the system, thereby realizing the monitoring and elimination of the total disturbance of the system; the feedback control rate is to adjust the control performance of the active disturbance rejection controller through parameter setting, and the three parts directly influence the subsequent disturbance compensation and the control performance of the active disturbance rejection controller through the synergistic effect.
Building active disturbance rejection control in SimulinkThe block diagram of the controller is shown in fig. 7, the number of important parameters to be set in the active disturbance rejection controller is mainly six, and the parameters are parameters in the feedback control rate respectively
Figure 38613DEST_PATH_IMAGE062
、
Figure 559724DEST_PATH_IMAGE063
(ii) a Parameters b1, b2, b3 in the extended state observer; bandwidth of state observer
Figure 747123DEST_PATH_IMAGE064
(ii) a And estimation of the system input gain b
Figure 670080DEST_PATH_IMAGE065
If the transfer function of the controlled system is known, the parameter is a known number. The parameters for arranging the transition (tracking differentiator TD) can be selected according to practical conditions and experience.
Dynamic simulation analysis shows that when the anti-interference performance of the control system is consistent, the predictive control is often high in response speed and small in overshoot, and is especially suitable for unstable objects. However, when the model is disturbed, the control effect of the predictive control is influenced to some extent. The active disturbance rejection controller has the advantages of having a strong disturbance rejection effect for a central air conditioning system with strong disturbance, and therefore having strong robustness. Since the central air-conditioning system studied in the text has the characteristic of strong disturbance, an active disturbance rejection controller is adopted to track the optimal working point of the control system so as to realize the optimization and energy saving of the central air-conditioning system.
The preferred embodiments of the present disclosure are described above with reference to the drawings, but the present disclosure is of course not limited to the above examples. Various changes and modifications within the scope of the appended claims may be made by those skilled in the art, and it should be understood that these changes and modifications naturally will fall within the technical scope of the present disclosure.
For example, a plurality of functions included in one unit may be implemented by separate devices in the above embodiments. Alternatively, a plurality of functions implemented by a plurality of units in the above embodiments may be implemented by separate devices, respectively. In addition, one of the above functions may be implemented by a plurality of units. Needless to say, such a configuration is included in the technical scope of the present disclosure.
In this specification, the steps described in the flowcharts include not only the processing performed in time series in the described order but also the processing performed in parallel or individually without necessarily being performed in time series. Further, even in the steps processed in time series, needless to say, the order can be changed as appropriate.
Although the embodiments of the present disclosure have been described in detail with reference to the accompanying drawings, it should be understood that the above-described embodiments are merely illustrative of the present disclosure and do not constitute a limitation of the present disclosure. It will be apparent to those skilled in the art that various modifications and variations can be made in the above-described embodiments without departing from the spirit and scope of the disclosure. Accordingly, the scope of the disclosure is to be defined only by the claims appended hereto, and by their equivalents.

Claims (10)

1. An active disturbance rejection control system for static optimization of a central air conditioning refrigeration station, comprising: the system comprises an air conditioner load prediction unit, a central air conditioner refrigerating station model calculation unit, a system working point optimization unit, a central control refrigerating station control unit and a central air conditioner refrigerating station;
the air conditioner load prediction unit is used for predicting the load through a BP neural network;
the central air-conditioning refrigeration station model calculation unit is used for executing one of a single model modeling mode and a multi-model fusion modeling mode;
the system working point optimization unit is used for calculating the optimal working point of the system by using an optimization algorithm;
and the central control refrigeration station control unit is used for tracking control by adopting an active disturbance rejection controller.
2. The system as claimed in claim 1, wherein said central air conditioning refrigeration station model calculation unit employs weighted multi-model fusion modeling to establish a static energy consumption model of the system for the refrigeration unit with the stronger nonlinearity; and for a chilled water system, a cooling water system and a cooling tower, a single model modeling mode is adopted.
3. The system as set forth in claim 2 wherein the modeling process of the multi-model fusion model of the chiller unit performed by the central air conditioning chiller station model calculation unit is specifically:
(1) dividing the working points of the refrigeration unit samples;
(2) establishing a sub-model of the refrigerating unit;
(3) and performing weighted fusion on the sub-models.
4. The system of claim 3, wherein the refrigeration unit sample operating point classifications are embodied as: the method comprises the steps of clustering modeling data by using a k-means algorithm, dividing the modeling data into 3 groups, randomly selecting 3 objects as initial clustering centers, calculating the distance between each object and each sub-clustering center, allocating each object to the nearest clustering center, and enabling the clustering centers and the objects allocated to the clustering centers to represent a cluster.
5. The system as set forth in claim 2 wherein said central air conditioning refrigeration station model calculation unit is configured to process the more non-linear refrigeration unit by: and establishing a static energy consumption model of the system by adopting weighted multi-model fusion modeling, and calculating the values of the outlet water temperature T _ wo of the chilled water and the outlet water temperature T _ co of the cooling water by utilizing an optimization algorithm aiming at the minimum refrigeration capacity Q obtained by load prediction and the energy consumption P of the refrigerating unit (Chiller-s).
6. The system of claim 2, wherein step (3) comprises in particular: for the chilled water system, the chilled water system calculation module calculates the flow m _ w of the chilled water and the outlet water temperature T _ wo of the chilled water by using a single model according to the flow m _ w of the chilled water and the outlet water temperature T _ wo of the chilled water and by taking the minimum energy consumption P _ (pump _ w) of the chilled water system as a target.
7. The system of claim 2, wherein step (3) comprises in particular: for the cooling water system, the cooling water system calculation module calculates the flow rate m _ c of the cooling water and the outlet water temperature T _ co of the cooling water by using a single model according to the flow rate m _ c of the cooling water and the outlet water temperature T _ co of the cooling water, and taking the minimum energy consumption P _ (pump _ c) of the cooling water system as a target.
8. The system of claim 2, wherein step (3) comprises in particular: for the cooling Tower, the cooling Tower calculation module calculates the air flow m _ a influenced by the cooling Tower fan according to the temperature difference T _ c of the cooling water and the air flow m _ a influenced by the cooling Tower fan by using the single model and aiming at the minimum energy consumption P _ of the cooling Tower (cooling _ Tower).
9. The system of claim 1, wherein the active disturbance rejection control system for the static optimization of the central air conditioning refrigeration station performs the data acquisition and optimization calculation processes at longer intervals without real-time processing, unlike the conventional dynamic optimization system.
10. An active disturbance rejection control method for a central air conditioning refrigeration station static optimization by using the system of any one of claims 1 to 9, the method specifically comprises the following steps:
step 1, predicting the refrigerating capacity Q obtained by accurate load prediction of a central air conditioner through a BP neural network, and controlling a refrigerating unit to generate enough refrigerating capacity according to the refrigerating capacity Q obtained by load prediction;
2, calculating variable parameters by using different models aiming at different equipment or systems, and establishing a static energy consumption model of the system by adopting weighted multi-model fusion modeling for a refrigerating unit with stronger nonlinearity;
step 3, performing variable calculation on the refrigerating water system and the cooling water system by adopting a single model modeling mode;
step 4, for the cooling tower, performing variable calculation by adopting a single model modeling mode;
step 5, verifying the variables obtained in the step 2-4 according to the objective function and the constraint condition to finally obtain a calculated optimal working point;
and 6, tracking the optimal working point of the control system by adopting the active disturbance rejection controller.
CN202110672979.7A 2021-06-17 2021-06-17 Active disturbance rejection control system for static optimization of central air-conditioning refrigeration station Active CN113821902B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110672979.7A CN113821902B (en) 2021-06-17 2021-06-17 Active disturbance rejection control system for static optimization of central air-conditioning refrigeration station

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110672979.7A CN113821902B (en) 2021-06-17 2021-06-17 Active disturbance rejection control system for static optimization of central air-conditioning refrigeration station

Publications (2)

Publication Number Publication Date
CN113821902A true CN113821902A (en) 2021-12-21
CN113821902B CN113821902B (en) 2022-07-05

Family

ID=78912534

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110672979.7A Active CN113821902B (en) 2021-06-17 2021-06-17 Active disturbance rejection control system for static optimization of central air-conditioning refrigeration station

Country Status (1)

Country Link
CN (1) CN113821902B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103912966A (en) * 2014-03-31 2014-07-09 武汉科技大学 Optimal control method for ground source heat pump refrigerating system
CN104515271A (en) * 2013-09-30 2015-04-15 代建国 Energy conservation optimization control system and method for central air conditioner refrigeration station system
CN105571063A (en) * 2015-12-31 2016-05-11 深圳市同鑫热力技术有限公司 Shallow geothermal energy source management system and implementation method thereof
CN108489012A (en) * 2018-01-30 2018-09-04 深圳市新环能科技有限公司 Cold source of air conditioning energy efficiency model control method based on load prediction and constraint
CN109945402A (en) * 2019-03-07 2019-06-28 东南大学 A kind of central air conditioning water system power-economizing method
CN110288164A (en) * 2019-07-02 2019-09-27 广州市特沃能源管理有限公司 A kind of building air conditioning refrigeration station system forecast Control Algorithm
CN112084707A (en) * 2020-09-02 2020-12-15 西安建筑科技大学 Refrigeration machine room energy-saving optimization method and system based on variable flow decoupling of chilled water and cooling water

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104515271A (en) * 2013-09-30 2015-04-15 代建国 Energy conservation optimization control system and method for central air conditioner refrigeration station system
CN103912966A (en) * 2014-03-31 2014-07-09 武汉科技大学 Optimal control method for ground source heat pump refrigerating system
CN105571063A (en) * 2015-12-31 2016-05-11 深圳市同鑫热力技术有限公司 Shallow geothermal energy source management system and implementation method thereof
CN108489012A (en) * 2018-01-30 2018-09-04 深圳市新环能科技有限公司 Cold source of air conditioning energy efficiency model control method based on load prediction and constraint
CN109945402A (en) * 2019-03-07 2019-06-28 东南大学 A kind of central air conditioning water system power-economizing method
CN110288164A (en) * 2019-07-02 2019-09-27 广州市特沃能源管理有限公司 A kind of building air conditioning refrigeration station system forecast Control Algorithm
CN112084707A (en) * 2020-09-02 2020-12-15 西安建筑科技大学 Refrigeration machine room energy-saving optimization method and system based on variable flow decoupling of chilled water and cooling water

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
LI ZHUANGJU 等: "Research on energy saving control of chilled water system of central air conditioning system", 《2016 28TH CHINESE CONTROL AND DECISION CONFERENCE》 *
LINGYAN CAO 等: "Electrical load prediction of healthcare buildings through single and ensemble learning", 《ENERGY REPORTS》 *
周栋梁 等: "多模型融合用于中央空调最优控制策略", 《中国新通信》 *
孙崇国 等: "基于自抗扰的变风量空调房间温湿度解耦控制", 《电工技术》 *
安建月 等: "基于负荷预测的中央空调制冷系统能耗综合优化", 《建筑热能通风空调》 *
李强 等: "自抗扰控制与广义预测控制的性能鲁棒性比较研究", 《第三十一届中国控制会议论文集》 *
郭虹 等: "基于中央空调的几种负荷预测方法对比分析", 《建筑热能通风空调》 *

Also Published As

Publication number Publication date
CN113821902B (en) 2022-07-05

Similar Documents

Publication Publication Date Title
Mohanraj et al. Applications of artificial neural networks for refrigeration, air-conditioning and heat pump systems—A review
CN102283283B (en) Intelligentized superficial geothermal energy low-temperature grain storage system
Du et al. Evaluation of operation and control in HVAC (heating, ventilation and air conditioning) system using exergy analysis method
CN110107989A (en) Small-sized based on chilled water return water temperature optimum set point determines frequency water cooler and becomes temperature control method of water
Huang et al. Multi-zone temperature prediction in a commercial building using artificial neural network model
Li et al. Recent advances in dynamic modeling of HVAC equipment. Part 1: Equipment modeling
CN113326651A (en) Refrigerating station load and energy efficiency ratio dynamic modeling method based on T-S fuzzy model
CN116398994B (en) Water chilling unit group control optimization method based on load prediction
CN113268913B (en) Intelligent building air conditioner cooling machine system operation optimization method based on PSO-ELM algorithm
CN111649457A (en) Dynamic predictive machine learning type air conditioner energy-saving control method
CN114543274B (en) Temperature and humidity optimization control method and system for building central air conditioner
Zhang et al. The model predictive control strategy of the transcritical CO2 air conditioning system used in railway vehicles
CN113028610B (en) Method and device for global optimization and energy-saving control of dynamic load of central air conditioner
Jiang et al. Multi-objective optimization of the refrigerant-direct convective-radiant cooling system considering the thermal and economic performances
Adnan et al. Modeling baseline electrical energy use of chiller system by artificial neural network
Ma et al. Test and evaluation of energy saving potentials in a complex building central chilling system using genetic algorithm
CN113821902B (en) Active disturbance rejection control system for static optimization of central air-conditioning refrigeration station
Baglivo et al. Air-source heat pump (ASHP) under very climate change scenarios: a numerical analysis
CN113757852B (en) Multi-split air conditioning unit control method and system based on digital twinning technology
Kim et al. Application of artificial neural network model for optimized control of condenser water temperature set-point in a chilled water system
Yang et al. Hybrid artificial neural network− genetic algorithm technique for condensing temperature control of air-cooled chillers
Song et al. Chiller System Modeling using PSO Optimization based NARX approach
CN117928052B (en) Energy-saving control method and system for central air conditioner
Zheng Research on energy-saving control and optimisation of air conditioning system based on genetic algorithm
Zhou et al. The application of artificial neural network in HVAC system

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
TR01 Transfer of patent right

Effective date of registration: 20230920

Address after: 100044 No. 1, Xicheng District exhibition museum, Beijing

Patentee after: Beijing University of Civil Engineering and Architecture

Address before: Beijing University of architecture, No. 15, Yongyuan Road, Huangcun Town, Daxing District, Beijing 102600

Patentee before: Li Zhuangju

Patentee before: Guo Hong

TR01 Transfer of patent right