CN111043720B - Low-cost robustness adjustment strategy making method of refrigeration system under load uncertainty - Google Patents

Low-cost robustness adjustment strategy making method of refrigeration system under load uncertainty Download PDF

Info

Publication number
CN111043720B
CN111043720B CN201910997972.5A CN201910997972A CN111043720B CN 111043720 B CN111043720 B CN 111043720B CN 201910997972 A CN201910997972 A CN 201910997972A CN 111043720 B CN111043720 B CN 111043720B
Authority
CN
China
Prior art keywords
adaptation
robustness
cost
low
load
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.)
Active
Application number
CN201910997972.5A
Other languages
Chinese (zh)
Other versions
CN111043720A (en
Inventor
丁研
宿皓
王翘楚
鄢睿
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.)
Tianjin University
Original Assignee
Tianjin University
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 Tianjin University filed Critical Tianjin University
Priority to CN201910997972.5A priority Critical patent/CN111043720B/en
Publication of CN111043720A publication Critical patent/CN111043720A/en
Application granted granted Critical
Publication of CN111043720B publication Critical patent/CN111043720B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Biomedical Technology (AREA)
  • Computational Linguistics (AREA)
  • Mechanical Engineering (AREA)
  • Chemical & Material Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Signal Processing (AREA)
  • Biophysics (AREA)
  • Combustion & Propulsion (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Fuzzy Systems (AREA)
  • Air Conditioning Control Device (AREA)

Abstract

The invention discloses a method for making a low-cost robustness adjustment strategy of a refrigeration system under load uncertainty. And dividing the system working condition through a part of load rate in the prediction interval. And establishing a refrigeration system model, a variable frequency water pump model and a circulating pipe network model which reflect the characteristics of the system. And determining a low-cost adaptation method of each working condition by using an optimization algorithm. The low-cost adjustment method mainly controls the chilled water supply temperature, the water pump motor frequency and the pipe network pressure difference. Finally, a multi-target decision method is used for carrying out decision making on the adaptation methods under different working conditions to determine the adaptation strategy which meets the requirements of robustness and low energy consumption.

Description

Low-cost robustness adjustment strategy making method of refrigeration system under load uncertainty
Technical Field
The invention belongs to the field of system control, and particularly relates to a low-cost robustness adjustment strategy making method for a refrigeration system under load uncertainty.
Background
The refrigerating unit is a main energy-using device in a building system, and in order to maintain the indoor comfortable temperature under the condition of hot outdoor climate, the refrigerating system is required to consume certain electric energy to provide enough cold energy for the air conditioning system. However, since the cooling load is fluctuating, there is usually a problem that the system is not matched with the cooling load, which causes the system to operate at a lower efficiency and a large amount of energy is wasted. Adaptation is a method of implementing cost-effective maintenance and operation measures in a building to achieve design intent and optimal operation of the system. The energy efficiency of the system can be improved by utilizing the adjustment, and the waste of energy sources is reduced.
At present, a plurality of system adjustment methods are used for solving the problem of mismatching of system operation and load, and many of the methods are combined with the adjustment of a system design method, and equipment needs to be redesigned and modified to better meet the load requirement, however, the adjustment strategy has large investment and long recovery period, and is difficult to be accepted by owners. Therefore, a low-cost adaptation method is required. In addition, the current adaptation method uses the load point prediction result as the basis of adaptation, the influence of load uncertainty is not considered, and the situation that the load is not matched with the cooling capacity is easy to occur.
Therefore, in the case of load uncertainty, describing the uncertainty thereof, and adapting the refrigeration system by a low-cost method is a technical problem which needs to be solved urgently at present.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a low-cost robustness adjustment strategy making method of a refrigeration system under load uncertainty.
The method utilizes a quantile regression neural network model to establish prediction on load uncertainty, establishes an optimized low-cost adaptation method for the system operation condition in a prediction interval, and finally determines a final adaptation strategy by utilizing a multi-objective decision method of an entropy weight method.
In order to achieve the purpose, the invention adopts the following technical scheme: a method for making a low-cost robustness adjustment strategy of a refrigeration system under load uncertainty comprises the following steps:
establishing a quantile regression neural network model by using the original data to obtain a probability distribution prediction result of the load under the load uncertainty;
dividing system working conditions in a prediction interval, establishing a model reflecting system characteristics, establishing a low-cost adjustment method, and determining adjustment moderate control parameters by using an optimization algorithm;
an entropy weight method is adopted as a multi-objective decision method for the adaptation methods under different working conditions to determine the adaptation strategy with energy conservation and robustness.
Further, the quantile regression neural network model input variables include: outdoor air temperature, outdoor air humidity, solar radiation intensity, indoor personnel number, and output variable is probability distribution of load value.
Furthermore, the quantile regression neural network model comprises an input layer, a hidden layer and an output layer, wherein the number of nodes of the input layer is the same as that of input variables of load prediction, the number of nodes of the output layer is 1, and only one hidden layer is included.
Further, the quantile regression neural network model adopts marbles loss as the loss function of the neural network model, and the expression is as follows:
for an input variable x, the value of the loss function ρ at the corresponding quantile ττ(x) Comprises the following steps:
Figure BDA0002240354820000021
further, the quantile regression neural network model is represented as follows:
for independent variable
Figure BDA0002240354820000022
The dependent variable Y (τ) corresponding to the τ quantile can be expressed as:
Figure BDA0002240354820000023
wherein W (τ) ═ Wji(τ), j is 1, 2, … m, i is 1, 2, … n } is a weight coefficient matrix from the input layer to the hidden layer, and V (τ) is { V { (V) }j(τ), j ═ 1, 2, … m } is the hidden-layer-to-output-layer weight coefficient vector. n is the number of nodes of the hidden layer, and m is the number of nodes of the output layer. g1(. and g)2The excitation equations between the input layer and the hidden layer and between the hidden layer and the output layer, respectively.
Furthermore, the system working condition division takes the partial load rate of the unit as the basis for dividing different working conditions, the partial load rate of the unit is divided into one working condition every time the partial load rate of the unit is reduced by 5% from 100%, and the corresponding unit cooling capacity is taken as the cooling capacity of the divided working condition operation.
Furthermore, the system model comprises a refrigerating unit model, a variable frequency water pump model and a chilled water circulating pipe network model.
Further, the adjustable parameter of the refrigerating unit model is chilled water supply temperature, and the specific unit model can be expressed as:
for chilled water supply temperature Tchw，oAnd COP of the partial load rate PLR unit can be expressed as:
COP＝β0+β1PLR+β2Tchw，o+β3PLR2+β4PLRTchw，o
when the cooling capacity of the unit is Q, the energy consumption of the unit is EchillerCan be expressed as:
Figure BDA0002240354820000031
wherein, beta0-β4Are coefficients obtained by the fitting.
Further, in the variable frequency water pump model, the lift and energy consumption of the water pump are related to the frequency and flow of the motor, which can be specifically expressed as:
the frequency of the motor is n (Hz), and the frequency is n under the rated working conditionm(Hz), a flow rate M, a head H (MPa) of the water pump:
Figure BDA0002240354820000032
the power is as follows:
Figure BDA0002240354820000033
wherein a is0-a3，b1-b3Is the correlation coefficient.
Furthermore, the model of the chilled water circulation pipe network is a simplified model for chilled water circulation and is suitable for different end forms. In the model, the pressure difference of water supply and return of the pipe network can be kept at a certain level by controlling the valve on the main pipe, and the change of the opening of the valve can cause the change of the resistance coefficient of the pipe network, so that the change of the working state point influences the flow value. Therefore, the method for controlling the resistance of the pipe network can change the flow of the system under the condition of not changing the frequency of the water pump electrode. The specific pipe network pressure differential setting can be calculated by the following equation:
PD＝(Scon+SPD)M2
Figure BDA0002240354820000034
wherein PD is the pressure difference, Spump，Schiller，Ss，Sr，SunitRespectively the resistance coefficient of the pump, the resistance coefficient of the cold machine, the resistance coefficient of the water supply main pipe, the resistance coefficient of the water return main pipe, the resistance coefficient of the branch pipe of the most unfavorable loop, SiThe resistance coefficient of the main pipe section connected with the branch pipe,
Figure BDA0002240354820000035
is the ratio of the flow of each branch pipe to the flow of the main pipe.
Further, the low-cost adaptation method comprises the control of the supply water temperature of the chilled water and the control of the flow rate of the chilled water. The control of the chilled water supply temperature is controlled by a set water supply temperature set value. The flow of the chilled water is controlled by the frequency conversion control of the frequency conversion water pump and the pressure difference control of the pipe network.
Further, in the optimization algorithm, the optimization target is that the energy consumption of the unit is minimum, which may be specifically expressed as:
Figure BDA0002240354820000036
wherein N is1Number of units to be started, N2The number of pumps that are turned on.
Further, the optimization algorithm is used for variable chilled water supply water temperature Tchw，oThe constraint condition of (2) is that the supply water temperature of the chilled water is set between 5 ℃ and 9 ℃.
Furthermore, the low-cost adjustment method realizes the operation of the unit under the uncertainty of the load only by adjusting the supply water temperature of the chilled water of the unit.
Further, the two evaluation criteria of the multi-objective decision method for the adaptation method are respectively: the energy consumption of the adaptation method and the robustness of the adaptation method to load uncertainty. The robustness of the adaptation method is represented by the cumulative probability density of the load satisfied by the adaptation method.
Further, the entropy weight method needs to normalize evaluation indexes under different adaptation methods, and a formula for normalizing energy consumption of the adaptation methods is as follows:
Figure RE-GDA0002406650390000041
the formula for the robustness normalization of the adaptation method is:
Figure RE-GDA0002406650390000042
wherein, XiIs the ith evaluation index, xijIs an index XiJ-th data of (1), YijAs data xijNormalized result of (2), max (X)i) And min (X)i) Are respectively an index XiMaximum and minimum values of.
Furthermore, the entropy weight method can calculate the information entropy E of different indexes according to the standardized result of different index dataiThe specific calculation formula is as follows:
Figure RE-GDA0002406650390000043
Figure BDA0002240354820000045
where m is the number of data of an index, and if pijWhen 0, then
Figure BDA0002240354820000046
Furthermore, the information entropy E of different indexes in the entropy weight methodiThe weight W of each index can be calculatedi：
Figure BDA0002240354820000047
Furthermore, the score of different indexes in the entropy weight method is an index weight WiNormalized index value Y of the adaptation methodijThe product of (a). And the comprehensive scores of the different adaptation methods are the sum of the index scores, and the result with the largest score is the final robustness adaptation strategy.
Advantageous effects
(1) The load uncertainty is described based on the quantile regression neural network model, and is represented by the probability distribution of the load. Meanwhile, the probability distribution result of the load can be used as a parameter for evaluating the robustness of the adaptation method.
(2) The invention adopts a low-cost adjustment method to adjust the refrigeration system, compared with the method of replacing equipment and redesigning the system, the system adjustment realized by only adjusting the control parameters of the equipment has more economical efficiency and can be widely adopted.
Drawings
FIG. 1 is a flow chart of a method for developing a low-cost robustness tuning strategy for a refrigeration system under load uncertainty in accordance with the present invention;
FIG. 2 is a diagram illustrating the load prediction results of the quantile regression neural network in an embodiment of the present invention;
FIG. 3 is a graph illustrating the predicted load probability distribution of the quantile regression neural network in an embodiment of the present invention;
FIG. 4 is a flow chart of a method for optimizing the low-cost adaptation method according to the present invention;
FIG. 5 shows original values of evaluation indexes under different conditions in an embodiment of the present invention;
FIG. 6 is a multi-objective decision result of different working condition adaptation methods according to an embodiment of the present invention.
Detailed Description
In order that the objects, technical solutions and advantages of the present invention will become more apparent, the present invention will be further described in detail with reference to the accompanying drawings in conjunction with the following specific embodiments.
As shown in fig. 1, the present embodiment provides a method for making a low-cost robust adaptation strategy of a refrigeration system under load uncertainty, comprising the following steps:
step 1: and predicting the probability distribution of the load by adopting a quantile regression neural network model.
In the embodiment, the load prediction input data comprises outdoor air temperature, outdoor air humidity, solar radiation intensity and the number of people in a building, the load condition of 5 days is predicted through a quantile regression neural network obtained through training, the result of a prediction interval is shown in fig. 2, and the load distribution condition in one time interval is shown in fig. 3.
Step 2: and dividing different working conditions in the prediction interval, and obtaining a low-cost adjustment method of each working condition by using an optimization algorithm.
In the embodiment, the cooling capacity of the unit at full load is 1500kW, and the part load rate interval is 5% as the basis for dividing the system working conditions, so that the part load rates corresponding to the divided working conditions are 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95% and 100% respectively. The corresponding cooling capacity is 825kW, 900kW, 975kW, 1050kW, 1125kW, 1200kW, 1275kW, 1350kW, 1425kW, 1500 kW.
After different working conditions are obtained, an optimal low-cost control method under the corresponding cooling capacity is calculated by using an optimization algorithm, and the objective function of the optimization algorithm is as follows:
Figure BDA0002240354820000061
the energy consumption of the unit and the energy consumption of the pump are determined by the established refrigerating unit model, the frequency conversion water pump model and the chilled water circulating pipe network model, and the control variables are chilled water supply temperature, frequency of a motor of the frequency conversion pump and set pressure difference. The process of optimization is shown in fig. 4.
The energy consumption of the pump can be calculated through a variable-frequency water pump model, and specifically comprises the following steps:
Figure BDA0002240354820000062
the system circulation flow is determined through the working state point jointly determined by the water pump model and the pipe network model, and the method specifically comprises the following steps:
PD＝(Scon+SPD)M2
Figure BDA0002240354820000063
the working state point of the circulating pipe network and the system flow can be solved by combining the two models.
The energy consumption of the unit can be calculated through a refrigerating unit model, and specifically comprises the following steps:
COP＝β0+β1PLR+β2Tchw，o+β3PLR2+β4PLRTchw，o
Figure BDA0002240354820000064
the final optimization results in an optimized low-cost adaptation method under different working conditions, namely a corresponding chilled water temperature set point, a water pump motor frequency set point and a pressure difference set point.
And step 3: the adaptation strategy with robustness and low energy consumption can be realized simultaneously by utilizing the multi-objective decision method to carry out decision determination on the adaptation methods under different working conditions.
The prediction result and the adaptation strategy at a moment are taken as examples. In this example, the load prediction interval is [834.87kW, 1303.96kW ]. The interval comprises 6 working conditions which are 900kW, 975kW, 1050kW, 1125kW, 1200kW and 1275kW respectively. The energy consumption and the robustness of the 6 working condition adapting methods are used as evaluation indexes of multi-target decision, wherein the robustness of the adapting methods is used as an index for evaluating the robustness through corresponding values of the robustness of the adapting methods in the cumulative probability density function of the load prediction result corresponding to the cooling capacity. The evaluation indexes of these 6 conditions are shown in fig. 5. Firstly, the two evaluation indexes are normalized, and the formula for normalizing the energy consumption of the adaptation method is as follows:
Figure RE-GDA0002406650390000065
the formula for the robustness normalization of the adaptation method is:
Figure RE-GDA0002406650390000066
wherein, XiIs the ith evaluation index, xijIs an index XiJ-th data of (1), YijAs data xijNormalized result of (2), max (X)i) And min (X)i) Are respectively an index XiMaximum and minimum values of.
Obtaining the standardized results of different index data, and calculating the information entropy E of different indexesiThe specific calculation formula is as follows:
Figure RE-GDA0002406650390000071
Figure BDA0002240354820000073
where m is the number of data of an index, and if pijWhen 0, then
Figure BDA0002240354820000075
Obtaining information entropy E of different indexesiThe weight W of each index can be calculatedi：
Figure BDA0002240354820000074
The scores of different indexes are index weights WiNormalized index value Y of the adaptation methodijThe product of (a). The comprehensive scores of the different adaptation methods are the sum of index scores of the different adaptation methods, and the result with the largest score is the final robustness adaptation strategy. The scores of the indexes and the final scores of the 6 working conditions are shown in FIG. 6. The adaptation method corresponding to the working condition with the cooling capacity of 1200kW can be determined as the final adaptation strategy at the moment through the decision result.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention and are not intended to limit the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (15)

1. The method for making the low-cost robustness adjustment strategy of the refrigeration system under the condition of load uncertainty is characterized by comprising the following steps of:
establishing a quantile regression neural network model by using the original data to obtain a probability distribution prediction result of the load under the load uncertainty;
dividing system working conditions in a prediction interval, establishing a model reflecting system characteristics, establishing a low-cost adjustment method, and determining adjustment moderate control parameters by using an optimization algorithm;
determining an adaptation strategy with energy conservation and robustness by adopting an entropy weight method as a multi-objective decision method for adaptation methods under different working conditions;
the quantile regression neural network model has the input variables of: outdoor air temperature, outdoor air humidity, solar radiation intensity, indoor personnel number and probability distribution with output variable as load value;
the quantile regression neural network model comprises an input layer, a hidden layer and an output layer, wherein the number of nodes of the input layer is the same as that of load prediction input variables, the number of nodes of the output layer is 1, and only one hidden layer is included;
the quantile regression neural network model adopts marble loss as a loss function of the neural network model, and the expression is as follows:
for an input variable x, the value of the loss function ρ at the corresponding quantile ττ(x) Comprises the following steps:
Figure FDA0002993118780000011
the quantile regression neural network model is represented as follows:
for argument X ═ Xi，i＝1…m，x∈RkThat corresponds to the dependent variable Y (τ) at quantile τ being expressed as:
Figure FDA0002993118780000012
wherein W (τ) ═ Wji(τ), j is 1, 2, … m, i is 1, 2, … n } is a weight coefficient matrix from the input layer to the hidden layer, and V (τ) is { V { (V) }j(τ), j ═ 1, 2, … m } is the weight coefficient vector from hidden layer to output layer;
n is the number of nodes of the hidden layer, and m is the number of nodes of the output layer;
g1(. and g)2The excitation equations between the input layer and the hidden layer and between the hidden layer and the output layer, respectively.
2. The method for making the low-cost robustness adaptation strategy of the refrigeration system under the load uncertainty according to claim 1, wherein the system working conditions are divided, a partial load rate of a unit is used as a basis for dividing different working conditions, the partial load rate of the unit is divided into one working condition every time when the partial load rate of the unit is reduced from 100%, and the corresponding unit cooling capacity is used as the cooling capacity for operation under the divided working conditions.
3. The method for formulating a low-cost robustness adaptation strategy for a refrigeration system under load uncertainty as recited in claim 1, wherein the system model comprises a refrigeration unit model, a variable frequency water pump model, and a chilled water circulation pipeline network model.
4. The method for formulating the low-cost robustness adaptation strategy of the refrigeration system under the uncertainty of the load as recited in claim 3, wherein the adjustable parameter of the refrigeration unit model is chilled water supply temperature, and the specific unit model can be expressed as:
for chilled water supply temperature Tchw，oAnd COP of the partial load rate PLR unit can be expressed as:
COP＝β0+β1PLR+β2Tchw，o+β3PLR2+β4PLRTchw，o
when the cooling capacity of the unit is Q, the energy consumption of the unit is EchillerCan be expressed as:
Figure FDA0002993118780000021
wherein, beta0-β4Are coefficients obtained by the fitting.
5. The method for formulating the low-cost robustness adaptation strategy of the refrigeration system under the uncertainty of the load as recited in claim 3, wherein the variable-frequency water pump model, the head and the energy consumption of the water pump are related to the frequency and the flow rate of the motor, and can be specifically expressed as:
the frequency of the motor is n (Hz), and the frequency is n under the rated working conditionm(Hz), a flow rate M, a head H (MPa) of the water pump:
Figure FDA0002993118780000022
the power is as follows:
Figure FDA0002993118780000023
wherein, a0-a3，b1-b3Is the correlation coefficient.
6. The method for formulating a low-cost robustness adaptation strategy for a refrigeration system under load uncertainty as recited in claim 3, wherein the chilled water circulation pipeline network model is a simplified model for chilled water circulation, and is applicable to different terminal forms; in the model, the pressure difference of water supply and return of a pipe network can be kept at a certain level by controlling a valve on a main pipe, and the change of the opening of the valve can cause the change of the resistance coefficient of the pipe network, so that the change of a working state point influences a flow value;
the method for controlling the resistance of the pipe network can change the flow of the system under the condition of not changing the frequency of the water pump electrode, and the specific pressure difference setting of the pipe network can be calculated by the following formula:
PD＝(Scon+SPD)M2
Figure FDA0002993118780000024
wherein PD is the pressure difference, Spump，Schiller，Ss，Sr，SunitRespectively the resistance coefficient of the pump, the resistance coefficient of the cold machine, the resistance coefficient of the water supply main pipe, the resistance coefficient of the water return main pipe, the resistance coefficient of the branch pipe of the most unfavorable loop, SiThe resistance coefficient of the main pipe section connected with the branch pipe,
Figure FDA0002993118780000025
is the ratio of the flow of each branch pipe to the flow of the main pipe.
7. The method for making a low-cost robustness adaptation strategy for a refrigeration system under load uncertainty according to claim 1, wherein the low-cost adaptation method comprises the following steps of controlling chilled water supply water temperature and chilled water flow:
controlling the supply water temperature of the chilled water through a set supply water temperature set value;
the flow of the chilled water is controlled by the frequency conversion control of the frequency conversion water pump and the pressure difference control of the pipe network.
8. The method for formulating the low-cost robustness adaptation strategy of the refrigeration system under the uncertainty of the load as recited in claim 1, wherein the optimization algorithm aims to minimize the unit energy consumption, and can be specifically expressed as:
Figure FDA0002993118780000031
wherein N1 is the number of units turned on, N2The number of pumps that are turned on.
9. The method of claim 8, wherein the optimization algorithm is applied to variable chilled water supply temperature Tchw，oThe constraint condition of (2) is that the supply water temperature of the chilled water is set between 5 ℃ and 9 ℃.
10. The method for formulating the low-cost robustness adaptation strategy of the refrigeration system under the load uncertainty as recited in claim 1, wherein the low-cost adaptation method is used for realizing the operation of the unit under the load uncertainty only by adjusting the supply water temperature of the chilled water of the unit.
11. The method for making the low-cost robustness adaptation strategy of the refrigeration system under the load uncertainty according to claim 1, wherein the two evaluation criteria of the adaptation method are respectively: the energy consumption of the adaptation method and the robustness of the adaptation method to load uncertainty;
the robustness of the adaptation method is represented by the cumulative probability density of the load that the adaptation method satisfies.
12. The method for formulating the low-cost robustness adaptation strategy of the refrigeration system under the uncertainty of the load as recited in claim 1, wherein the entropy weight method is required to standardize evaluation indexes under different adaptation methods, and the formula for standardizing the energy consumption of the adaptation method is as follows:
Figure FDA0002993118780000032
the formula for the robustness normalization of the adaptation method is:
Figure FDA0002993118780000033
wherein, XiIs the ith evaluation index, xijIs an index XiJ-th data of (1), YijAs data xijNormalized result of (2), max (X)i) And min (X)i) Are respectively an index XiMaximum and minimum values of.
13. The method for formulating a low-cost robustness adaptation strategy for a refrigeration system under load uncertainty as recited in claim 12, wherein the entropy weight method is used for calculating the information entropy E of different indexes according to the standardized results of different index dataiThe specific calculation formula is as follows:
Figure FDA0002993118780000041
Figure FDA0002993118780000042
where m is the number of data of an index, and if pijWhen 0, then
Figure FDA0002993118780000044
14. The method for formulating a low-cost robustness adaptation strategy for a refrigeration system under load uncertainty as recited in claim 12, wherein the entropy weight method is information entropy E of different indexesiThe weight W of each index can be calculatedi：
Figure FDA0002993118780000043
15. The method for formulating a low-cost robustness adaptation strategy for a refrigeration system under load uncertainty as recited in claim 12 wherein the entropy weight method, the scoring of different indices is an index weight WiNormalized index value Y of the adaptation methodijThe product of (a); and the comprehensive scores of the different adaptation methods are the sum of the indexes with the highest score, and the result with the highest score is the final robustness adaptation strategy.
CN201910997972.5A 2019-10-21 2019-10-21 Low-cost robustness adjustment strategy making method of refrigeration system under load uncertainty Active CN111043720B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910997972.5A CN111043720B (en) 2019-10-21 2019-10-21 Low-cost robustness adjustment strategy making method of refrigeration system under load uncertainty

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910997972.5A CN111043720B (en) 2019-10-21 2019-10-21 Low-cost robustness adjustment strategy making method of refrigeration system under load uncertainty

Publications (2)

Publication Number Publication Date
CN111043720A CN111043720A (en) 2020-04-21
CN111043720B true CN111043720B (en) 2021-05-14

Family

ID=70232323

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910997972.5A Active CN111043720B (en) 2019-10-21 2019-10-21 Low-cost robustness adjustment strategy making method of refrigeration system under load uncertainty

Country Status (1)

Country Link
CN (1) CN111043720B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113094873B (en) * 2021-03-15 2023-04-07 中联重科股份有限公司 Stable working condition identification method, working condition evaluation method and concrete pumping equipment

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103277875A (en) * 2013-06-21 2013-09-04 上海能誉科技发展有限公司 Energy-saving control system for refrigeration plant room
CN103912966A (en) * 2014-03-31 2014-07-09 武汉科技大学 Optimal control method for ground source heat pump refrigerating system
CN106529021A (en) * 2016-11-09 2017-03-22 东南大学 Air conditioning system simulation method based on feature recognition
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
CN109708258A (en) * 2018-12-20 2019-05-03 南京达实能源技术有限公司 A kind of temperature of ice house feedforward-Fuzzy control system and control method based on load dynamic change

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103277875A (en) * 2013-06-21 2013-09-04 上海能誉科技发展有限公司 Energy-saving control system for refrigeration plant room
CN103912966A (en) * 2014-03-31 2014-07-09 武汉科技大学 Optimal control method for ground source heat pump refrigerating system
CN106529021A (en) * 2016-11-09 2017-03-22 东南大学 Air conditioning system simulation method based on feature recognition
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
CN109708258A (en) * 2018-12-20 2019-05-03 南京达实能源技术有限公司 A kind of temperature of ice house feedforward-Fuzzy control system and control method based on load dynamic change

Also Published As

Publication number Publication date
CN111043720A (en) 2020-04-21

Similar Documents

Publication Publication Date Title
CN111256294B (en) Model prediction-based optimization control method for combined operation of water chilling unit
CN110410942B (en) Energy-saving optimization control method and system for cold and heat source machine room
CN110392515B (en) Historical data-based cold and heat source machine room energy-saving control method and system
CN110288164B (en) Predictive control method for building air-conditioning refrigeration station system
CN109945402A (en) A kind of central air conditioning water system power-economizing 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
CN107192097A (en) Control method and device for air-conditioning chilled water system
CN113757788B (en) Station-load linked two-network balance online dynamic intelligent regulation and control method and system
CN108990383A (en) A kind of data center's air-conditioning system forecast Control Algorithm
CN112415924A (en) Energy-saving optimization method and system for air conditioning system
CN113739365A (en) Central air-conditioning cold station group control energy-saving control method, device, equipment and storage medium
CN116266253A (en) Optimization control method, system and computer readable storage medium for air conditioner parameters
CN113701321B (en) Energy-saving frequency conversion control method for central air-conditioning water pump
CN110895029A (en) Building load prediction method based on temperature of chilled water
CN116398994B (en) Water chilling unit group control optimization method based on load prediction
CN114154677A (en) Air conditioner operation load model construction and prediction method, device, equipment and medium
CN111043720B (en) Low-cost robustness adjustment strategy making method of refrigeration system under load uncertainty
CN117628659A (en) Predictive regulation and control method for multi-system cooperative operation of heating ventilation air conditioner
CN102679649B (en) Energy conservation control device and energy conservation control method of refrigerating system
CN117366799A (en) Control method of building cold source system
CN115630331A (en) Air-conditioning refrigeration station global energy-saving optimization and regulation and control method based on mathematical physical model
CN114358604A (en) Urban centralized heating system time-sharing and zone-dividing regulation and control method considering building heat storage characteristics
CN202648280U (en) Energy-saving control device of refrigerating system
CN110836518A (en) System basic knowledge based global optimization control method for self-learning air conditioning system
CN112146247B (en) Control method and system for parallel connection of multiple same-model variable-frequency freezing water pumps

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