CN111256294A  Model predictionbased optimization control method for combined operation of water chilling unit  Google Patents
Model predictionbased optimization control method for combined operation of water chilling unit Download PDFInfo
 Publication number
 CN111256294A CN111256294A CN202010056062.XA CN202010056062A CN111256294A CN 111256294 A CN111256294 A CN 111256294A CN 202010056062 A CN202010056062 A CN 202010056062A CN 111256294 A CN111256294 A CN 111256294A
 Authority
 CN
 China
 Prior art keywords
 water
 water chilling
 unit
 chilling unit
 temperature
 Prior art date
 Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
 Granted
Links
 XLYOFNOQVPJJNPUHFFFAOYSAN water Substances data:image/svg+xml;base64,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 data:image/svg+xml;base64,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 O XLYOFNOQVPJJNPUHFFFAOYSAN 0.000 title claims abstract description 224
 238000005457 optimization Methods 0.000 title claims description 4
 238000005265 energy consumption Methods 0.000 claims abstract description 46
 238000009826 distribution Methods 0.000 claims abstract description 24
 238000005057 refrigeration Methods 0.000 claims abstract description 19
 238000005516 engineering process Methods 0.000 claims abstract description 4
 210000002569 neurons Anatomy 0.000 claims description 33
 239000000498 cooling water Substances 0.000 claims description 21
 238000004364 calculation method Methods 0.000 claims description 8
 238000001816 cooling Methods 0.000 claims description 7
 230000001276 controlling effect Effects 0.000 claims description 4
 230000000875 corresponding Effects 0.000 claims description 4
 238000007710 freezing Methods 0.000 claims description 3
 239000008400 supply water Substances 0.000 claims description 3
 238000010977 unit operation Methods 0.000 claims description 3
 238000005096 rolling process Methods 0.000 claims description 2
 239000000126 substance Substances 0.000 claims description 2
 230000001537 neural Effects 0.000 claims 1
 238000004134 energy conservation Methods 0.000 abstract description 2
 238000004378 air conditioning Methods 0.000 description 3
 230000000694 effects Effects 0.000 description 3
 238000004519 manufacturing process Methods 0.000 description 3
 238000009499 grossing Methods 0.000 description 2
 238000010801 machine learning Methods 0.000 description 2
 230000004048 modification Effects 0.000 description 2
 238000006011 modification reaction Methods 0.000 description 2
 239000008358 core component Substances 0.000 description 1
 238000002790 crossvalidation Methods 0.000 description 1
 230000003247 decreasing Effects 0.000 description 1
 238000010586 diagram Methods 0.000 description 1
 238000011156 evaluation Methods 0.000 description 1
 230000001105 regulatory Effects 0.000 description 1
 230000035945 sensitivity Effects 0.000 description 1
Classifications

 F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
 F24—HEATING; RANGES; VENTILATING
 F24F—AIRCONDITIONING; AIRHUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
 F24F11/00—Control or safety arrangements
 F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
 F24F11/46—Improving electric energy efficiency or saving
 F24F11/47—Responding to energy costs

 G—PHYSICS
 G06—COMPUTING; CALCULATING; COUNTING
 G06N—COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
 G06N3/00—Computer systems based on biological models
 G06N3/02—Computer systems based on biological models using neural network models
 G06N3/04—Architectures, e.g. interconnection topology

 G—PHYSICS
 G06—COMPUTING; CALCULATING; COUNTING
 G06N—COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
 G06N3/00—Computer systems based on biological models
 G06N3/02—Computer systems based on biological models using neural network models
 G06N3/08—Learning methods

 G—PHYSICS
 G06—COMPUTING; CALCULATING; COUNTING
 G06Q—DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
 G06Q10/00—Administration; Management
 G06Q10/04—Forecasting or optimisation, e.g. linear programming, "travelling salesman problem" or "cutting stock problem"
Abstract
The invention provides a GRNNbased water chilling unit energy efficiency model modeling method and a water chilling unit group control method based on cold load prediction and a water chilling unit energy efficiency model, and relates to the field of water chilling unit energy conservation. The energy efficiency model of the water chilling unit is established by utilizing the GRNN technology and the actual operation data of the refrigeration system, so that the energy consumption of the water chilling unit under different operation working conditions can be rapidly and accurately predicted. And (3) optimizing the number of running water chilling units and the load rate distribution in real time by using the cold load prediction data and the water chilling unit energy efficiency model and taking the lowest system energy consumption as a target, thereby effectively improving the running energy efficiency of the refrigeration system.
Description
Technical Field
The invention belongs to the field of energysaving control of a refrigerating system, and particularly relates to an optimal control method for combined operation of a water chilling unit based on model prediction.
Background
The water chilling unit is a core component of the centralized airconditioning system, and the operation energy consumption of the water chilling unit accounts for more than 40% of the total energy consumption of the centralized airconditioning system. For a system in which a plurality of water chilling units of a large building are operated in a combined mode, the water chilling units are different in operation energy efficiency under different load rates, a reasonable water chilling unit group control method is adopted, so that the water chilling units are enabled to operate efficiently as far as possible, the operation efficiency of a refrigeration system is improved to the maximum extent, and the method is an important technical approach for achieving energy saving of a centralized airconditioning system.
The traditional group control of the water chilling units adopts a feedback control method, generally, the cooling load is monitored, and the number of the units is increased or reduced when the actually measured load reaches a set limit value; or the number of the units is increased when the water temperature is lower than a set value by monitoring the outlet water temperature of the chilled water; or by monitoring the flow of the bypass pipe, when the flow of the bypass pipe is more than 1 unit flow, the number of units in operation is reduced. Although the traditional water chiller group control method basically meets the requirement of the cold load of an air conditioner, the operation adjustment is lagged, the change characteristic of the operation energy efficiency of the water chiller along with the load factor and the operation working condition is not considered, and the energysaving operation of a refrigeration system is seriously restricted.
The invention patent document CN110222398A discloses an artificial intelligence control method for a water chilling unit, which comprises the following steps: establishing a cold load prediction model; acquiring historical weather and production data of a water chilling unit within a certain period of time as first training data; training the first training data to the cold load prediction model through a first machine learning algorithm; inputting future weather forecast data and a production plan to the trained cold load prediction model so as to enable the cold load prediction model to output cold load requirements within a preset time; further comprising: establishing a cold machine model; acquiring historical data of the refrigerator collected within a certain time as second training data; training the cold machine model by second training data through a second machine learning algorithm; inputting the cold load demand within the preset time output by the cold load prediction model to the trained refrigerator model, and using the cold load demand as input data to output the cold load demand data after the cold load prediction model and the refrigerator model are combined and optimized; although the established cold load prediction model is trained through historical weather and production data, the technical problem that the future cold load of the water chilling unit cannot be predicted at present is solved, and the cold load of the water chilling unit can be accurately predicted according to the feedback of the terminal working condition without the intervention of field personnel, the energysaving problem is not considered in the invention, and particularly the optimal energysaving mode of the combined operation of the water chilling unit based on model prediction is not considered.
Patent document CN110542178A provides a control method and system for an air conditioner refrigeration machine room with selflearning capability, which can selflearn dynamically adjust input parameters according to the actual working conditions of the air conditioner refrigeration machine room, and find the operating condition parameter set corresponding to the overall COP efficiency maximization of the system, by using a COP efficiency prediction model and obtaining the selected probability value of the operating condition parameter set, thereby achieving the energy saving effect better than that achieved by the conventional control method for a chiller group control system. Although the invention focuses on the situations of prediction and energy conservation, the operation efficiency and the energysaving effect are limited and need to be further improved.
Disclosure of Invention
In order to solve the technical problems in the prior art, the invention provides a water chiller group control method based on model prediction, which is used for carrying out active feedforward control on a refrigeration system with large hysteresis characteristic and seeking the optimal operation efficiency of the refrigeration system.
The specific technical scheme is as follows: on one hand, the invention provides a water chilling unit energy efficiency model modeling method based on GRNN, which is characterized by comprising the following steps:
s1: establishing a training data set according to actual operation data of the refrigeration system, wherein the training data set comprises operation power W of each water chilling unit of the system_{i}Load factor, load factorChilled water supply temperature t_{gi}Cooling water return temperature ct_{hi}；
S2: based on the training data set obtained in step S1 tot_{gi}、ct_{hi}Input parameters, W, for the model_{i}For outputting parameters, use generalized regression godRespectively establishing an energy consumption prediction model of each water chilling unit through a network (GRNN);
s3: inputting the load rate of the water chilling unit based on the energy consumption prediction model of the water chilling unitChilled water supply water temperature set value t_{gsi}And cooling water return water temperature ct_{hi}The running power W of the unit can be obtained through prediction_{i}。
Further, the load factor is the ratio of the refrigerating capacity of the water chilling unit to the rated refrigerating capacity;
further, the energy consumption prediction model of the water chilling unit comprises an input layer, a mode layer, a summation layer and an output layer;
inputting parameters by an input layer and transmitting the input parameters to a mode layer; wherein, the input parameter is the load factor of the water chilling unitChilled water supply temperature t_{gi}And cooling water return water temperature ct_{hi}；
The mode layer is used for outputting n neurons;
the summation layer comprises two types of neurons, and the first type of neuron calculates and sums the n neuron outputs of the mode layer; transfer function S_{D}Comprises the following steps:
the second type of neuron is to perform weighted summation on n neuron outputs of the model layer, wherein the weight is the jth element output by the nth training sample, and the transfer function S_{j}Comprises the following steps:
the output layer has 1 neuron output, namely the running power W of the water chilling unit_{i}Each neuron divides the output of the summation layer to obtain a prediction result;
wherein the content of the first and second substances,the output of the jth neuron corresponds to the predicted result of the jth element.
Further, the establishing of the training data set in step S1 includes the following steps:
s11: recording automatic monitoring data of the refrigerating system, wherein the monitoring data comprises the operating power W of each water chilling unit_{i}Temperature t of chilled water supply_{gi}Return temperature t of chilled water_{h}Cooling water inlet temperature ct_{hi}；
S12: according to the monitoring data and the rated refrigerating capacity Q of the unit_{edi}Calculating the load factor of the unit according to the following formula
Wherein C is the specific heat of water, m_{i}、t_{gsi}、t_{h}Actually measured flow, chilled water supply set temperature and chilled water actually measured return water temperature of the ith water chilling unit; q_{edi}Rated refrigerating capacity of the ith water chilling unit is respectively set;
s13: based on operating power W_{i}Load factor of the unitChilled water supply temperature t_{gi}Cooling water inlet temperature ct_{hi}The data of the GRNN model is established by adopting an SQL database technology.
Further, the operation power W in step S13_{i}Load factor of the unitChilled water supply temperature t_{gi}Cooling water inlet temperature ct_{hi}The data of (a) is at least 1 data for the cooling season.
Furthermore, the energy consumption prediction model of the water chilling unit adopts a rolling training mode, and the operation data needs to be added into a training data set in real time so as to ensure that the energy consumption prediction model of the water chilling unit reflects the current unit operation performance.
On the other hand, the invention provides a water chiller group control method based on cold load prediction and a water chiller energy efficiency model, which is characterized by comprising the following steps of:
s01: determining all combination modes of the running number of the water chilling units and the corresponding gross customized cooling capacity Q according to the types and the number of the water chilling units in the refrigeration system_{ed}(ii) a Customizing the total amount of the cold energy Q under each combination mode of the water chilling unit_{ed}And predicted load Q_{yc}Comparing to obtain a feasible combination mode of the number of the running water chilling unit units;
s02: calculating system energy consumption W under each feasible combination mode of the number of running water chilling units based on cold load predicted value, water chilling unit energy consumption model, chilled water pump and cooling water pump power_{sys}Selecting system energy consumption W_{sys}The lowest combination mode is used as the optimal combination mode of the number of the selected water chilling units;
s03: based on the optimal combination mode of the running numbers of the water chilling units selected in the step S02, the chilled water of the water chilling units is used for supplying water with the set temperature t_{gsi}As a mode for adjusting the load factor of the unit, the system energy consumption W is used_{sys}The lowest is an optimization target, the system water supply temperature is taken as a constraint condition, and the optimal distribution mode of the operation load rate of the water chilling unit is determined through the water chilling unit energy consumption model;
s04: and adjusting the number of the running water chilling units time by time according to the optimal number of the running water chilling units selected in the step S02 and the load rate distribution determined in the step S03, and resetting the set water temperature of each water chilling unit.
Further, in step S01, when the total amount of the customized cooling capacity Q is reached_{ed}And predicted load Q_{yc}Ratio of performanceRelatively, satisfy 2Q_{yc}＞Q_{ed}＞Q_{yc}The combination mode of the number of the running water chilling units under the condition is a feasible combination mode of the number of the running water chilling units.
Further, in step S02, the system power consumption W_{sys}The total running power of all the water chilling units which are planned to run and the matched freezing water pumps and cooling water pumps thereof is calculated; the running power of each water chilling unit is determined according tot_{gsi}And ct_{h}Inputting the energy efficiency model of the water chilling unit to be obtained through prediction; wherein the system load rateThe calculation method is as follows:
further, step S03 specifically includes the following steps:
s031: calculating the load rate of each water chilling unit at different chilled water set temperatures, wherein the calculation method comprises the following steps:
s032: calculating system water supply temperature t 'in different unit load rate distribution modes based on optimal combination mode of running water chiller units'_{g}The calculation method is as follows:
s033: when t'_{g}When the following constraint conditions are met, determining the load factor of the water chilling unit as a feasible distribution mode;
t_{gs}0.5＜t′_{g}＜t_{gs}+0.5
s034: aiming at the feasible distribution mode of the load rate of the water chilling unit,based on the water chilling unit energy consumption model, calculating the total system energy consumption W_{sys}；
S035: system energy consumption W under feasible combination mode of comparing load rates of different water chilling units_{sys}Selecting system energy consumption W_{sys}The lowest optimal distribution mode is used as the load factor of the water chilling units, and the water supply set temperature of each water chilling unit is determined.
Further, step S04 includes realtime unit increase and decrease according to the conventional feedback control, and when an increase and decrease occurs, step S03 is repeated to find the optimal load factor distribution mode and reset the water supply set temperature of each water chilling unit.
Compared with the prior art, the invention has the following beneficial effects:
(1) the water chilling unit energy consumption model established based on GRNN has the advantages of simple modeling and good applicability, truly reflects the actual operation performance of the water chilling unit, and can quickly and effectively judge the energy consumption of the water chilling unit under different operation working conditions;
(2) the model predictionbased water chiller group control method aims at the lowest energy consumption, optimizes the number of running water chillers and the load rate distribution, and can effectively improve the running energy efficiency of the refrigeration system.
Drawings
Fig. 1 is a schematic diagram of an energy efficiency model of a GRNNbased chiller provided by the present invention;
fig. 2 is a flow chart of a water chiller group control method based on model prediction according to the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings.
Referring to fig. 1, the method for modeling the energy efficiency model of a chiller based on GRNN according to the present invention may be implemented as follows:
s1: and establishing a training data set of the GRNN model.
① recording automatic monitoring data of the refrigeration system at 10 min intervals, and the monitoring parameters include the running power W of each water chilling unit_{i}Temperature t of chilled water supply_{gi}Return temperature t of chilled water_{h}Cooling water inlet temperature ct_{hi}。
② calculating unit load factor according to the monitored data and rated refrigerating capacity of the unitAs shown in the following formula:
wherein C is the specific heat of water, m_{i}、t_{gsi}、t_{h}Actually measured flow, chilled water supply set temperature and chilled water actually measured return water temperature of the ith water chilling unit; q_{edi}Respectively the rated refrigerating capacity of the ith water chilling unit.
③ based on W_{i}、t_{gi}And ct_{hi}The data (no less than 1 for cold seasons) is used for establishing a training data set of the GRNN model by adopting an SQL database technology. Each chiller has 1 training data set, and the actual operating data should be added to the data sets in real time.
S2: based on the GRNN principle and the training data set shown in FIG. 1, an energy efficiency model of each water chilling unit is established. When there is an update to the data set, the model should be retrained.
The energy efficiency model of each water chilling unit comprises an input layer, a mode layer, a summation layer and an output layer, and specifically comprises the following steps:
① input layer
The input layer has 3 neurons, i.e. the load factor of the chillerChilled water supply temperature t_{gi}And cooling water return water temperature ct_{hi}Each neuron is a simple distribution unit that passes input parameters to the mode layer.
② mode layer
The pattern layer has N neurons, N being the number of training data set samples. And selecting a Gaussian function as a transfer function form, wherein the output expression of the nth neuron of the mode layer is as follows:
XX_{n}the input of the neuron is a network input vector X and a weight vector X_{n}Euclidean distance of. When the input of the neuron is 0, the output of the neuron is a maximum value of 1. The sensitivity of neurons to input is regulated by the smoothing factor σ. And determining the optimal smoothing factor sigma of the GRNN prediction model by adopting a cross validation method, wherein the evaluation index is the prediction relative error.
③ summation layer
The summation layer includes two types of neurons, the first type of neuron computationally sums the neuron outputs of all mode layers, and the transfer function is:
the second type of neuron is to perform weighted summation on the outputs of all mode layers, and the weight is the output Y of the nth training sample_{n}The jth element of (1), the transfer function is:
④ output layer
The output layer has 1 neuron, i.e. the running power (W) of the water chilling unit_{i}) Each neuron divides the output of the summation layer to obtain a prediction result, and the output of the jth neuron corresponds to the prediction result of the jth element, and the prediction result is as follows:
s3: inputting the load rate of the water chilling unit based on the water chilling unit energy consumption modelChilled water supply water temperature set value t_{gsi}And cooling water return water temperature ct_{hi}The running power W of the unit can be obtained through prediction_{i}。
Referring to fig. 2, the method for controlling a water chiller group based on model prediction provided by the invention comprises the following steps:
s01: and (4) a feasible combination mode of roughly selecting the number of running water chilling units.
① analyzing the composition of the refrigeration system, determining all combination modes of the number of running water chilling units according to the type and number of the water chilling units in the refrigeration system (for example, the type of the water chilling units has 2 (the refrigeration capacity of the large unit is 1000 tons, and the refrigeration capacity of the small unit is 400 tons), the number of each type of units (3 large units and 2 small units), and the number of the running water chilling units can be determined to be 11, such as 1 large unit, 2 large units, 3 large units, 1 large unit +1 small unit, 1 large unit +2 small units, 2 large units +1 small unit, 2 large units +2 small units, 1 small unit, 2 small units, 1 large unit +2 small units, 3 large units +2 small units), and determining the corresponding total customized Q refrigeration capacity_{ed}；
② customizing the total amount of cold Q under each combination mode of the water chilling unit_{ed}And predicted load Q_{yc}Comparing, if the predicted load is 1700 cold tons, when 2Q is satisfied_{yc}＞Q_{ed}＞Q_{yc}Under the condition, the feasible combination modes for determining the number of the running cold water sets are 5: 1 big unit +2 little units, 2 big units +1 little unit, 2 big units +2 little units, 3 big units.
S02: and determining the optimal combination mode of the number of the running water chilling units.
① calculating System energy consumption W under each feasible combination mode of running number of water chilling units_{sys}The system energy consumption comprises the total power of all the water chilling units planned to operate and the matched freezing water pump and cooling water pump. Wherein, the running work of the water chilling unit is based on the energy consumption model of the water chilling unit and is transmitted by the water pumpLoad rate of system entryChilled water supply set temperature t_{gsi}And cooling water return water temperature ct_{h}And predicting to obtain.
② comparing system energy consumption W under feasible combination mode of different water chilling unit operation numbers_{sys}And selecting the optimal combination mode with the lowest energy consumption as the number of the running water chilling units.
S03: and optimizing the distribution mode of the running load rate of the water chilling unit.
① calculating the load factor of each water chilling unit at different chilled water set temperatures (such as 713 deg.C, taking integer), the calculation method is as follows:
② calculating system water supply temperature t 'in different unit load rate distribution modes based on the optimal combination mode of the number of running water chilling units'_{g}The calculation method is as follows:
③ when t'_{g}And when the following constraint conditions are met, determining the load factor of the water chilling unit as a feasible distribution mode.
t_{gs}0.5＜t′_{g}＜t_{gs}+0.5
④ calculating the system energy consumption W based on the chiller energy consumption model aiming at the feasible distribution mode of the chiller load factor_{sys}。
⑤ comparing system energy consumption W under feasible combination mode of different water chilling unit load rates_{sys}And selecting the lowest energy consumption as the optimal distribution mode of the load rate of the water chilling units, and determining the set water supply temperature of each water chilling unit.
S04: group control of water chilling unit
① the number of the running water chilling units is adjusted time by time based on the optimal number of the running water chilling units and the load rate distribution, and the water supply set temperature of each water chilling unit is reset.
②, in order to ensure the safety and stability of the system operation, the unit is increased or decreased in real time according to the traditional feedback control, when the increase or decrease occurs, the step S03 is repeated to find the optimal load rate distribution mode, and the water supply set temperature of each water chilling unit is reset.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. A water chilling unit energy efficiency model modeling method based on GRNN is characterized by comprising the following steps:
s1: establishing a training data set according to actual operation data of the refrigeration system, wherein the training data set comprises operation power W of each water chilling unit of the system_{i}Load factor, load factorChilled water supply temperature t_{gi}Cooling water return temperature ct_{hi}；
S2: based on the training data set obtained in step S1 tot_{gi}、ct_{hi}Input parameters, W, for the model_{i}Adopting a Generalized Regression Neural Network (GRNN) to respectively establish an energy consumption prediction model of each water chilling unit for outputting parameters;
s3: inputting the load rate of the water chilling unit based on the energy consumption prediction model of the water chilling unitChilled water supply water temperature set value t_{gsi}And cooling water return water temperature ct_{hi}The running power W of the unit can be obtained through prediction_{i}。
2. The GRNNbased chiller energy efficiency model modeling method of claim 1, wherein the chiller energy consumption prediction model comprises an input layer, a mode layer, a summation layer and an output layer;
inputting parameters by an input layer and transmitting the input parameters to a mode layer; wherein, the input parameter is the load factor of the water chilling unitChilled water supply temperature t_{gi}And cooling water return water temperature ct_{hi}；
The mode layer is used for outputting n neurons;
the summation layer comprises two types of neurons, and the first type of neuron calculates and sums the n neuron outputs of the mode layer; transfer function S_{D}Comprises the following steps:
the second type of neuron is to perform weighted summation on n neuron outputs of the model layer, wherein the weight is the jth element output by the nth training sample, and the transfer function S_{j}Comprises the following steps:
the output layer has 1 neuron output, namely the running power (W) of the water chilling unit_{i}) Each neuron divides the output of the summation layer to obtain a prediction result
Wherein the content of the first and second substances,the output of the jth neuron corresponds to the predicted result of the jth element.
3. The GRNNbased chiller unit energy efficiency model modeling method according to claim 1, wherein the establishing of the training data set in step S1 specifically includes the steps of:
s11: recording automatic monitoring data of the refrigerating system, wherein the monitoring data comprises the operating power W of each water chilling unit_{i}Temperature t of chilled water supply_{gi}Return temperature t of chilled water_{h}Cooling water inlet temperature ct_{hi}；
S12: according to the monitoring data and the rated refrigerating capacity Q of the unit_{edi}Calculating the load factor of the unit according to the following formula
Wherein C is the specific heat of water, m_{i}、t_{gsi}、t_{h}Actually measured flow, chilled water supply set temperature and chilled water actually measured return water temperature of the ith water chilling unit; q_{edi}Rated refrigerating capacity of the ith water chilling unit is respectively set;
s13: based on operating power W_{i}Load factor of the unitChilled water supply temperature t_{gi}Cooling water inlet temperature ct_{hi}The data of the GRNN model is established by adopting an SQL database technology.
4. The GRNNbased modeling method for the energy efficiency model of the chiller according to claim 3, wherein the step S13 is implementedLine power W_{i}Load factor of the unitChilled water supply temperature t_{gi}Cooling water inlet temperature ct_{hi}The data of (a) is at least 1 data for the cooling season.
5. The GRNNbased chiller unit energy efficiency model modeling method of claim 1, wherein the chiller unit energy consumption prediction model employs a rolling training mode, and the operation data needs to be added to a training data set in real time to ensure that the chiller unit energy consumption prediction model reflects the current unit operation performance.
6. A water chiller group control method based on cold load prediction and a water chiller energy efficiency model is characterized by comprising the following steps:
s01: determining all combination modes of the running number of the water chilling units and the corresponding gross customized cooling capacity Q according to the types and the number of the water chilling units in the refrigeration system_{ed}(ii) a Customizing the total amount of the cold energy Q under each combination mode of the water chilling unit_{ed}And predicted load Q_{yc}Comparing to obtain a feasible combination mode of the number of the running water chilling unit units;
s02: calculating system energy consumption W under each feasible combination mode of the number of running water chilling units based on cold load predicted value, water chilling unit energy consumption model, chilled water pump and cooling water pump power_{sys}Selecting system energy consumption W_{sys}The lowest combination mode is used as the optimal combination mode of the number of the selected water chilling units;
s03: based on the optimal combination mode of the running numbers of the water chilling units selected in the step S02, the chilled water of the water chilling units is used for supplying water with the set temperature t_{gsi}As a mode for adjusting the load factor of the unit, the system energy consumption W is used_{sys}The lowest is an optimization target, the system water supply temperature is taken as a constraint condition, and the optimal distribution mode of the operation load rate of the water chilling unit is determined through the water chilling unit energy consumption model;
s04: and adjusting the number of the running water chilling units time by time according to the optimal number of the running water chilling units selected in the step S02 and the load rate distribution determined in the step S03, and resetting the set water temperature of each water chilling unit.
7. The method for controlling the group of cold water machines based on the cold load prediction and the cold water machine energy efficiency model as claimed in claim 6, wherein in step S01, when the total amount of the customized cooling capacity Q is reached_{ed}And predicted load Q_{yc}When compared, satisfy 2Q_{yc}＞Q_{ed}＞Q_{yc}The combination mode of the number of the running water chilling units under the condition is a feasible combination mode of the number of the running water chilling units.
8. The method for controlling the group of water chiller units based on the cold load prediction and the water chiller unit energy efficiency model as claimed in claim 6, wherein in step S02, the system energy consumption W is_{sys}The total running power of all the water chilling units which are planned to run and the matched freezing water pumps and cooling water pumps thereof is calculated; the running power of each water chilling unit is determined according tot_{gsi}And ct_{h}Inputting the energy efficiency model of the water chilling unit to be obtained through prediction; wherein the system load rateThe calculation method is as follows:
9. the method for controlling the group of water chiller units based on the cold load prediction and the water chiller unit energy efficiency model as claimed in claim 6, wherein the step S03 specifically comprises the following steps:
s031: calculating the load rate of each water chilling unit at different chilled water set temperaturesThe calculation method is as follows:
s032: calculating system water supply temperature t 'in different unit load rate distribution modes based on optimal combination mode of running water chiller units'_{g}The calculation method is as follows:
s033: when t'_{g}When the following constraint conditions are met, determining the load factor of the water chilling unit as a feasible distribution mode;
t_{gs}0.5＜t′_{g}＜t_{gs}+0.5
s034: aiming at the feasible distribution mode of the load rate of the water chilling unit, based on the water chilling unit energy consumption model, the system energy consumption W is calculated_{sys}；
S035: system energy consumption W under feasible combination mode of comparing load rates of different water chilling units_{sys}Selecting system energy consumption W_{sys}The lowest optimal distribution mode is used as the load factor of the water chilling units, and the water supply set temperature of each water chilling unit is determined.
10. The method as claimed in claim 6, wherein the step S04 further comprises performing a unit increase/decrease operation in real time according to a conventional feedback control, and when the increase/decrease operation occurs, repeating the step S03 to find an optimal load factor distribution manner and reset the set water supply temperature of each chiller.
Priority Applications (1)
Application Number  Priority Date  Filing Date  Title 

CN202010056062.XA CN111256294B (en)  20200117  20200117  Model predictionbased optimization control method for combined operation of water chilling unit 
Applications Claiming Priority (1)
Application Number  Priority Date  Filing Date  Title 

CN202010056062.XA CN111256294B (en)  20200117  20200117  Model predictionbased optimization control method for combined operation of water chilling unit 
Publications (2)
Publication Number  Publication Date 

CN111256294A true CN111256294A (en)  20200609 
CN111256294B CN111256294B (en)  20210105 
Family
ID=70948935
Family Applications (1)
Application Number  Title  Priority Date  Filing Date 

CN202010056062.XA Active CN111256294B (en)  20200117  20200117  Model predictionbased optimization control method for combined operation of water chilling unit 
Country Status (1)
Country  Link 

CN (1)  CN111256294B (en) 
Cited By (4)
Publication number  Priority date  Publication date  Assignee  Title 

CN111895629A (en) *  20200722  20201106  格瑞美科技(武汉)有限公司  Universal equipment end implementation method for central air conditioner to participate in demand response 
CN112364999A (en) *  20201019  20210212  深圳市超算科技开发有限公司  Training method and device for water chiller adjustment model and electronic equipment 
CN112526888A (en) *  20210207  20210319  广州大学城能源发展有限公司  Regional concentrated tail end load prediction method 
CN113091234A (en) *  20210408  20210709  贵州汇通华城股份有限公司  Method and system for selecting onoff state of refrigeration host 
Citations (10)
Publication number  Priority date  Publication date  Assignee  Title 

CN101251291A (en) *  20080403  20080827  上海交通大学  Central air conditioning system global optimization energysaving control method and device based on model 
CN201129823Y (en) *  20071121  20081008  厦门立思科技有限公司  Central air conditioner energysaving control device based on artificial neural net technique 
CN101424436A (en) *  20081129  20090506  深圳市奥宇控制系统有限公司  Intelligent optimizing control system and method for central airconditioning 
CN201335488Y (en) *  20081129  20091028  深圳市奥宇控制系统有限公司  Intelligently optimized control device for central air conditioner 
KR20110100895A (en) *  20100305  20110915  삼성물산 주식회사  The building mutual assistance control method which uses an optimization energy management system 
CN102980272A (en) *  20121208  20130320  珠海派诺科技股份有限公司  Air conditioner system energy saving optimization method based on load prediction 
CN103912966A (en) *  20140331  20140709  武汉科技大学  Optimal control method for ground source heat pump refrigerating system 
CN104331737A (en) *  20141121  20150204  国家电网公司  Office building load prediction method based on particle swarm neural network 
CN109945402A (en) *  20190307  20190628  东南大学  A kind of central air conditioning water system powereconomizing method 
CN110107989A (en) *  20190430  20190809  北京工业大学  Smallsized based on chilled water return water temperature optimum set point determines frequency water cooler and becomes temperature control method of water 

2020
 20200117 CN CN202010056062.XA patent/CN111256294B/en active Active
Patent Citations (10)
Publication number  Priority date  Publication date  Assignee  Title 

CN201129823Y (en) *  20071121  20081008  厦门立思科技有限公司  Central air conditioner energysaving control device based on artificial neural net technique 
CN101251291A (en) *  20080403  20080827  上海交通大学  Central air conditioning system global optimization energysaving control method and device based on model 
CN101424436A (en) *  20081129  20090506  深圳市奥宇控制系统有限公司  Intelligent optimizing control system and method for central airconditioning 
CN201335488Y (en) *  20081129  20091028  深圳市奥宇控制系统有限公司  Intelligently optimized control device for central air conditioner 
KR20110100895A (en) *  20100305  20110915  삼성물산 주식회사  The building mutual assistance control method which uses an optimization energy management system 
CN102980272A (en) *  20121208  20130320  珠海派诺科技股份有限公司  Air conditioner system energy saving optimization method based on load prediction 
CN103912966A (en) *  20140331  20140709  武汉科技大学  Optimal control method for ground source heat pump refrigerating system 
CN104331737A (en) *  20141121  20150204  国家电网公司  Office building load prediction method based on particle swarm neural network 
CN109945402A (en) *  20190307  20190628  东南大学  A kind of central air conditioning water system powereconomizing method 
CN110107989A (en) *  20190430  20190809  北京工业大学  Smallsized based on chilled water return water temperature optimum set point determines frequency water cooler and becomes temperature control method of water 
Cited By (5)
Publication number  Priority date  Publication date  Assignee  Title 

CN111895629A (en) *  20200722  20201106  格瑞美科技(武汉)有限公司  Universal equipment end implementation method for central air conditioner to participate in demand response 
CN111895629B (en) *  20200722  20210611  格瑞美科技(武汉)有限公司  Universal equipment end implementation method for central air conditioner to participate in demand response 
CN112364999A (en) *  20201019  20210212  深圳市超算科技开发有限公司  Training method and device for water chiller adjustment model and electronic equipment 
CN112526888A (en) *  20210207  20210319  广州大学城能源发展有限公司  Regional concentrated tail end load prediction method 
CN113091234A (en) *  20210408  20210709  贵州汇通华城股份有限公司  Method and system for selecting onoff state of refrigeration host 
Also Published As
Publication number  Publication date 

CN111256294B (en)  20210105 
Similar Documents
Publication  Publication Date  Title 

CN111256294B (en)  Model predictionbased optimization control method for combined operation of water chilling unit  
CN104134100B (en)  A kind of energysaving management system based on cloud computing  
CN103912966B (en)  A kind of earth source heat pump refrigeration system optimal control method  
CN109959123A (en)  A kind of energysaving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network  
CN105444356A (en)  Intelligent energy efficiency optimizing control system for central air conditioning system and control method of intelligent energy efficiency optimizing control system  
CN106920006A (en)  A kind of subway station air conditioning energy consumption Forecasting Methodology based on ISOA LSSVM  
CN106338127A (en)  Load forecast and control system and method for metro heat ventilation and airconditioning system  
CN108990383B (en)  Predictive control method for air conditioning system of data center  
CN110410942B (en)  Energysaving optimization control method and system for cold and heat source machine room  
CN104833154B (en)  Chilled water loop control method based on fuzzy PID and neural internal model  
CN109945402A (en)  A kind of central air conditioning water system powereconomizing method  
CN101655272A (en)  Energysaving control management system of network central air conditioner and method thereof  
CN103591637A (en)  Centralized heating secondary network operation adjustment method  
CN110392515A (en)  A kind of Cooling and Heat Source equipment room energyconserving control method and system based on historical data  
CN109959122A (en)  A kind of Energy consumption forecast for air conditioning method based on shot and long term memory Recognition with Recurrent Neural Network  
CN107940679A (en)  A kind of group control method based on data center's handpiece Water Chilling Units performance curve  
CN108168030A (en)  A kind of intelligent control method based on refrigeration performance curve  
CN109028446A (en)  A kind of refrigerating plant room control method based on equipment overall performance model  
CN109917646A (en)  A kind of district cooling and heating equipment optimization operating system and method  
CN110567101A (en)  Water chiller highenergyefficiency control method based on support vector machine model  
CN108413567B (en)  Central air conditioner costsaving optimization method and system based on Internet of things  
CN112415924A (en)  Energysaving optimization method and system for air conditioning system  
Zhuang et al.  A decentralized method for energy conservation of an HVAC system  
CN110895029A (en)  Building load prediction method based on temperature of chilled water  
CN111043720B (en)  Lowcost robustness adjustment strategy making method of refrigeration system under load uncertainty 
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 