CN109959123A - A kind of energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network - Google Patents
A kind of energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network Download PDFInfo
- Publication number
- CN109959123A CN109959123A CN201910182274.XA CN201910182274A CN109959123A CN 109959123 A CN109959123 A CN 109959123A CN 201910182274 A CN201910182274 A CN 201910182274A CN 109959123 A CN109959123 A CN 109959123A
- Authority
- CN
- China
- Prior art keywords
- air
- cooling supply
- conditioning
- water temperature
- 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.)
- Pending
Links
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
- F24F11/30—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
-
- 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
- F24F11/63—Electronic processing
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Air Conditioning Control Device (AREA)
Abstract
A kind of energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network, comprising the following steps: step 1, establish Energy consumption forecast for air conditioning assessment models;Step 2, Optimal Parameters are determined;Step 3, cooling supply water temperature, cooling supply backwater temperature difference are encoded with genetic algorithm, it is in a certain range, random to generate cooling supply water temperature, cooling supply backwater temperature difference according to this coding, obtain several genomes at initial population;Step 4, the other parameters of current working and chromosome parameter are decoded to and inputted LSTM-RNN air-conditioning forecast assessment model, carry out chromosome assessment, calculate fitness function, and intersected to more excellent chromosome, made a variation;Obtained optimal chromosome decoding i.e. optimized parameter;Step 5, by the other parameters under optimized parameter combination current working, air-conditioning power consumption after input prediction assessment models are optimized.Invention improves forecast assessment accuracy rate, achievees the effect that preferably to optimize energy consumption.
Description
Technical field
The present invention relates to a kind of energy-saving method for air conditioner that Recognition with Recurrent Neural Network is remembered based on genetic algorithm and shot and long term.
Background technique
In China, the energy consumption of building increases year by year, has accounted for 40% of global energy requirements or so.Meanwhile air-conditioning and
Heating system accounts for about the half of building total energy consumption, and proportion is continuously increased in recent years.According to statistics, China's public building
Energy saving compliance rate is less than 10%.So doing certain adjustment for air-conditioning system, it can accomplish that the maximum of energy-saving potential excavates.It is existing
Generation building is usually combined with various technologies, accomplishes a degree of building energy saving.
Building automation system (BAS) is to be integrated with technology of Internet of things, the system of the technologies such as control technology, network technology.
It by the monitoring of various equipment Comprehensive Automations to building (group) and management, for owner and user provide safety, comfortably,
The work and living environment of convenient and efficient, and whole system and one of the various equipment is made to be in optimal working condition, thus
The economy of guarantee system operation and modernization, informationization and the intelligence of management.At the same time, in BAS system, largely
Air-conditioning data such as temperature, humidity, flow, power etc. are all recorded in the database.But these data seldom effectively by with
In air-conditioning analysis, modeling, optimization.Air-conditioning system is analyzed by mass data, models, optimize, it can be preferably pre-
Air conditioning energy consumption is surveyed, reflect air conditioning condition in building and carries out automatic management in real time, automatic management and energy conservation is realized, improves simultaneously
The comfort of personnel in building.
Summary of the invention
Assessment modeling, Optimization Steps in order to overcome the shortcomings of existing air conditioning energy consumption optimization method are many and diverse, and the present invention provides
The pretty good energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network of a kind of relatively simple and effect.
The technical solution adopted by the present invention to solve the technical problems is:
A kind of energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network, comprising the following steps:
Step 1, Energy consumption forecast for air conditioning assessment models are established, the Water cooled air conditioners project data normalizing provided using the favourable opposition energy
After change, the input of Recognition with Recurrent Neural Network, the air-conditioning total energy consumption air-conditioning total energy under current working are remembered as LSTM-RNN shot and long term
Consumption is used as neural network prediction target, after carrying out network training, obtains final Energy consumption forecast for air conditioning assessment models;
Step 2, Optimal Parameters are determined, in the case where air conditioner refrigerating amount is constant, reduce power consumption to greatest extent, are improved
cop;Cooling supply water temperature is set, cooling supply backwater temperature difference is Optimal Parameters;
Step 3, initialization of population is encoded cooling supply water temperature, cooling supply backwater temperature difference with genetic algorithm, according to
This coding, it is in a certain range, random to generate cooling supply water temperature, cooling supply backwater temperature difference, obtain several genomes at
Initial population;
Step 4, parameter optimization, by the other parameters of current working and chromosome parameter decoding and input step 3 obtains
LSTM-RNN air-conditioning forecast assessment model, carry out chromosome assessment, calculate fitness function, and more excellent chromosome is handed over
Fork, variation;Repetitive operation step 4, until being optimized to setting degree or being genetic to setting algebra, obtained optimal chromosome decoding
The optimized parameter exactly obtained;
Step 5, the decoding of obtained optimal chromosome is combined into the other parameters under current working, input prediction assesses mould
Type, air-conditioning power consumption after being optimized.
Further, the data in the step 1 are as follows:
(1) input variable: wherein the input quantity includes outdoor temperature, outside humidity, cooling supply water temperature, cooling backwater
Temperature, freezing supply water temperature, freezing return water temperature, general pipeline flow, month label and time tag;
(2) predictive variable: predictive variable is the air conditioning energy consumption under current working;
(3) data normalization: the sample data of input variable is normalized by following formula
Wherein x' is the data after normalization, and x is input data, and μ is the mean value of data, and σ is the standard deviation of data.
Further, the LSTM-RNN shot and long term in the step 1 remembers Recognition with Recurrent Neural Network prediction model network structure
It is as follows:
(1) indicate the parameter of memory historical information length in Timesteps:RNN, value represent that RNN can utilize when
Between sequence length, use timesteps for 1 with more the model of versatility;
(2) activation primitive: selecting tanh as activation primitive in RNN, it can the continuous real value of input " compressed " to-
Between 1 and 1, if it is very big negative, then output is exactly -1;If it is very big positive number, output is exactly 1;
(3) neuron selects: shot and long term memory (long-short term memory, LSTM) network is one in RNN
The specific neuron of kind, controls retaining for information in hidden layer by the structure of 3 doors, can filter out weight in information flow
To keep result more accurate with unessential information;
(4) loss function: being modified mean square error (mean squared error, MSE), when calculating predicted value
predictedtWith actual value observertDifference square after, then divided by actual value observert, obtain square-error for reality
The ratio of actual value, then be averaged;So that model suitably focuses on the adjustment of big error rate data, but can be very good to exclude from
The interference of group's point, some particular values can also be looked after, and obtained result is average, can more preferably accomplish universality;
(5) other parameters: the hidden layer that model uses 10 layers of LSTM node to constitute, every layer has 50 nodes, adaptive ladder
Descent method is spent, exercise wheel number is 5000 wheels, batch_size 30, learning rate 0.00005.
Further, the step 2, in 3,4, genetic algorithm parameter setting and optimization structure are as follows:
(1) chromosome coding: use binary coding as the coding mode of genetic algorithm.In the ginseng that air-conditioning operates normally
Air-conditioning parameter is set in number range;
(2) it optimization object function: in the case where operating normally and meeting the precondition of refrigeration demand, needs to reach central air-conditioning
Total equipment operation energy consumption is minimum, can according to objective function and constraint condition, using penalty penalty, by constraint condition into
The problem of row is converted accordingly, and Prescribed Properties are become unconfined condition, forces the solution of objective function in constraint condition;
Constraint condition:
(TCooling backwater temperature-TWet-bulb temperature)max≥TCooling backwater temperature-TWet-bulb temperature
≥(TCooling backwater temperature-TWet-bulb temperature)min
(TCooling supply water temperature-TWet-bulb temperature)max≥TCold supply and return water temperature-TWet-bulb temperature
≥(TCooling supply water temperature-TWet-bulb temperature)min
Penalty:
Penalty=[max { 0, (2- (TCooling supply water temperature-TCooling supply backwater temperature difference–TWet-bulb temperature))}]2
+[max{0,(TCooling supply water temperature-TCooling supply backwater temperature difference-TWet-bulb temperature-6)}]2
+[max{0,(4-(TCooling supply water temperature-TWet-bulb temperature))}]2
+[max{0,(TCooling supply water temperature-TWet-bulb temperature-8)}]2
Obtain objective function:
F=PPredict power consumption+penalty
(3) fitness function: fitness function can assess all chromosome, and obtained fitness value is to measure currently
The embodiment of the superiority and inferiority degree of chromosome;
Fitness=-F
(4) other conditions: being evolved using 50 individual populations, initializes gene by the way of random roulette
Population primary is selected, using the interleaved mode of single point crossing, crossover probability 80%, 10% gene mutation rate terminates item
Part is the evolution by 100 generations, selects the highest gene of fitness value as final optimal gene.
Technical concept of the invention are as follows: in the part air-conditioning project data that favourable opposition energy science and technology company provides, and the external world
On the basis of environmental data, certain data prediction is carried out, Feature Selection obtains more higher with the air conditioning energy consumption degree of association
Characteristic is trained characteristic and energy consumption data by specific algorithm, generates energy consumption prediction model, further according to
Air-conditioning data and environmental data data, and parameter optimization is carried out using genetic algorithm, in the case where refrigerating capacity is constant, reduce empty
Power Regulation consumption, improves air-conditioning cop.
Beneficial effects of the present invention are mainly manifested in: when handling air-conditioning data, in statistics
The methods of related coefficient excludes some extraneous features, and increases some correlated characteristics according to information such as times;On this basis, make
With LSTM-RNN come training pattern, model training process is simplified to a certain extent, improves forecast assessment accuracy rate.It uses
Genetic algorithm, relatively simple optimization air-conditioning parameter achieve the effect that preferably to optimize energy consumption.
Detailed description of the invention
Fig. 1 is the stream of the energy-saving method for air conditioner of the present invention that Recognition with Recurrent Neural Network is remembered based on genetic algorithm and shot and long term
Cheng Tu;
Fig. 2 is LSTM-RNN basic block diagram belonging to the present invention;
Fig. 3 is LSTM unit basic block diagram.
Fig. 4 is genetic algorithm basic flow chart
Specific embodiment
The invention will be further described below in conjunction with the accompanying drawings.
Referring to Fig.1~Fig. 4, a kind of energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network,
It the described method comprises the following steps:
Step 1, Energy consumption forecast for air conditioning assessment models are established, the Water cooled air conditioners project data normalizing provided using the favourable opposition energy
After change, the input of Recognition with Recurrent Neural Network, the air-conditioning total energy consumption air-conditioning total energy under current working are remembered as LSTM-RNN shot and long term
Consumption is used as neural network prediction target, after carrying out network training, obtains final Energy consumption forecast for air conditioning assessment models;
As shown in Figure 1, the present invention is using LSTM-RNN algorithm, the Water cooled air conditioners that preprepared has been pre-processed
Input training set of the air conditioning energy consumption data as network under data and current working, sets loss function, makes loss gradient
Decline reaches after certain fitting precision or after being recycled to certain number, deconditioning obtains air-conditioning energy to be fitted air conditioning energy consumption
Consume prediction model
Data used are as follows:
(1) input variable: wherein the input quantity includes at least outdoor temperature, outside humidity, cooling supply water temperature, cooling
Return water temperature, freezing supply water temperature, freezing return water temperature, general pipeline flow, month label and time tag;
(2) predictive variable: predictive variable is the air conditioning energy consumption under current working;
(3) data normalization: the sample data of input variable is normalized by following formula
Wherein x' is the data after normalization, and x is input data, and μ is the mean value of data, and σ is the standard deviation of data.
It is as follows that LSTM-RNN shot and long term remembers Recognition with Recurrent Neural Network prediction model network parameter:
(1) Timesteps:Timesteps, that is, time step indicates the parameter of memory historical information length, value in RNN
The length of time series that RNN can be utilized is represented, due to needing the model with more versatility, directly adopt Timesteps is
1。
(2) activation primitive: selecting tanh as activation primitive in RNN, it can the continuous real value of input " compressed " to-
Between 1 and 1.Particularly, if it is very big negative, then output is exactly -1;If it is very big positive number, output is exactly
1。
(3) neuron selects: shot and long term memory (long-short term memory, LSTM) network is one in RNN
The specific neuron of kind, controls retaining for information in hidden layer by the structure of 3 doors, can filter out weight in information flow
To keep result more accurate with unessential information.As shown in figure 3, LSTM cellular construction is as follows:
It in time t, inputs as Xt, the preceding input of hidden layer is ht-1 generation, may be not too important in table last data
Information.The preceding input of unit is that Ct-1 represents possible important information in last data.These three inputs pass through input gate
It, the processing for forgeing door ft and out gate ot, form unit output state Ct, and hidden layer exports ht and final output Yt.
Input gate:
Forget door:
Out gate:
Unit input:
Unit output:
Hidden layer output:
ht=ot*tanh(Ct)
WhereinBy xtIt is connected to the weight matrix of three doors and unit input,It is by ht-1It is connected to the weight matrix of three doors and the input of unit unit, bi, bf, bi, bCIt is three doors
With the shift term of unit input, σ represents sigmoid functionTanh is exactly hyperbolic tangent function
(4) loss function: mean square error (mean squared error, MSE) is modified.When calculating predicted value
predictedtWith actual value observertDifference square after, then divided by actual value observert, obtain square-error for
The ratio of actual value, then be averaged.So that model suitably focuses on the adjustment of big error rate data, and can be very good to exclude
The interference of outlier, some particular values can also be looked after, and obtained result is average, can more preferably accomplish universality.
(5) other parameters: the hidden layer that model uses 10 layers of LSTM node to constitute, every layer has 50 nodes.Adaptive ladder
Descent method is spent, exercise wheel number is 5000 wheels.Batch_size is 30, learning rate 0.00005.Table 1 is training set and test set
As a result:
Data set | Training set error rate | Test set error rate |
Project 7 | 0.00852 | 0.01103 |
Project 8 | 0.00892 | 0.01633 |
Project 9 | 0.00824 | 0.01386 |
Table 1
Above-mentioned air-conditioning prediction model evaluation criterion is mean error ME (Mean Error), and formula is as follows:
Mean error refers in equal precision measurement, the arithmetic mean of instantaneous value of the random error of measured all measured values.
Step 2, Optimal Parameters are determined, in the case where making air conditioner refrigerating amount constant, power consumption is reduced to greatest extent, mentions
High cop;Cooling supply water temperature is set in numerous parameters, cooling supply backwater temperature difference is Optimal Parameters.
Step 3, initialization of population is encoded cooling supply water temperature, cooling supply backwater temperature difference with genetic algorithm.According to
This coding, it is random to generate cooling supply water temperature, cooling supply backwater temperature difference, obtain several genomes at initial population;
(1) chromosome coding: use binary coding as the coding mode of genetic algorithm, cooling supply water temperature code area
Between for outdoor temperature+11 and outdoor temperature -4, then set interval is 4 He of maximum value that data-oriented is concentrated to cooling supply backwater temperature difference
Minimum value 1, it is ensured that obtained optimized parameter can operate normally in air-conditioning equipment.Encoding precision is 0.01.
(2) it optimization object function: in the case where operating normally and meeting the precondition of refrigeration demand, needs to reach central air-conditioning
Total equipment operation energy consumption is minimum.Can according to objective function and constraint condition, using penalty penalty, by constraint condition into
The problem of row is converted accordingly, and Prescribed Properties are become unconfined condition, forces the solution of objective function in constraint condition.For
Ensure that system operates normally, using the maximin value of data set as constraint condition.In data set, cooling backwater temperature
It is largely fallen between 2 to 6 with the difference of wet-bulb temperature, cooling supply water temperature and wet-bulb temperature difference section are between 4 to 8.
Constraint condition:
(TCooling backwater temperature-TWet-bulb temperature)max≥TCooling backwater temperature-TWet-bulb temperature
≥(TCooling backwater temperature-TWet-bulb temperature)min
(TCooling supply water temperature-TWet-bulb temperature)max≥TCold supply and return water temperature-TWet-bulb temperature
≥(TCooling supply water temperature-TWet-bulb temperature)min
Penalty:
Penalty=[max { 0, (2- (TCooling supply water temperature-TCooling supply backwater temperature difference–TWet-bulb temperature))}]2
+[max{0,(TCooling supply water temperature-TCooling supply backwater temperature difference-TWet-bulb temperature-6)}]2
+[max{0,(4-(TCooling supply water temperature-TWet-bulb temperature))}]2
+[max{0,(TCooling supply water temperature-TWet-bulb temperature-8)}]2
Obtain objective function:
F=PPredict power consumption+penalty
(3) fitness function: fitness function can assess all chromosome, and obtained fitness value is to measure currently
The embodiment of the superiority and inferiority degree of chromosome.The fitness value of this method largely derives from the forecast assessment mould that RNN is fitted
Type predicts the power consumption of entire air-conditioning under current working by input parameter.And the power consumption the low just represents under current refrigeration capacity, it is empty
Adjusting body energy consumption is minimum, cop highest.And operating condition is adjusted according to the principle of the big temperature difference of small flow, so needing to be added certain
Penalty as adjustment, that is, prediction power consumption it is lower, cooling it is bigger for return water, fitness value is higher.So fitness
Function formula is as follows: fitness=-F
(4) other conditions: this method is evolved using 50 individual populations, and initialization gene uses random roulette
Mode select population primary.Using the interleaved mode of single point crossing, crossover probability 80%, 10% gene mutation rate.
Termination condition is the evolution by 100 generations, selects the highest gene of fitness value as final optimal gene.
Step 4, parameter optimization, by the other parameters of current working and chromosome parameter decoding and input step 3 obtains
LSTM-RNN air-conditioning forecast assessment model, carry out chromosome assessment, calculate fitness function, and more excellent chromosome is handed over
Fork, variation.Repetitive operation step 4, until being optimized to setting degree or being genetic to setting algebra, obtained optimal chromosome decoding
The optimized parameter exactly obtained.
Step 5, the decoding of obtained optimal chromosome is combined into the other parameters under current working, input prediction assesses mould
Type, air-conditioning power consumption after being optimized.
Table 2,3,4 is the energy-saving effect table of three projects day, and table 2 is before project air-conditioning system on 7 July 23 optimizes
Energy consumption comparison table afterwards.
Table 2
Table 3 is 8 August of project air-conditioning system optimization front and back energy consumption comparison table on the 23rd.
Table 3
Table 4 is that project air-conditioning system on 9 July 27 optimizes front and back energy consumption comparison table.
Table 4
Table 5 is the energy conservation front and back final table of comparisons of result.
Project number | Former total power consumption | Total power consumption after energy conservation | Amount of energy saving | Energy saving ratio |
7 | 13,742.41 | 11,344.12 | 2,398.29 | 0.17 |
8 | 14550.59 | 10338.74 | 4211.85 | 0.29 |
9 | 23851.02 | 17101.29 | 6749.73 | 0.28 |
Table 5
Final evaluation criterion of the present invention is a project one day total energy saving ratio, formula are as follows:
Wherein ESR is total fractional energy savings, P0tFor the original power of current working, PetFor the power after optimization under current working.
Claims (4)
1. a kind of energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network, which is characterized in that described
Method the following steps are included:
Step 1, Energy consumption forecast for air conditioning assessment models are established, are normalized using the Water cooled air conditioners project data that the favourable opposition energy provides
Afterwards, the input of Recognition with Recurrent Neural Network, the air-conditioning total energy consumption air-conditioning total energy consumption under current working are remembered as LSTM-RNN shot and long term
Final Energy consumption forecast for air conditioning assessment models are obtained after carrying out network training as neural network prediction target;
Step 2, Optimal Parameters are determined, in the case where air conditioner refrigerating amount is constant, power consumption is reduced to greatest extent, improves cop;If
Set cooling supply water temperature, cooling supply backwater temperature difference is Optimal Parameters;
Step 3, initialization of population is encoded cooling supply water temperature, cooling supply backwater temperature difference with genetic algorithm, according to this volume
Code, it is in a certain range, random to generate cooling supply water temperature, cooling supply backwater temperature difference, obtain several genomes at it is initial
Population.
Step 4, the other parameters of current working and chromosome parameter are decoded what simultaneously input step 3 obtained by parameter optimization
LSTM-RNN air-conditioning forecast assessment model carries out chromosome assessment, calculates fitness function, and hand over more excellent chromosome
Fork, variation;Repetitive operation step 4, until being optimized to setting degree or being genetic to setting algebra, obtained optimal chromosome decoding
The optimized parameter exactly obtained;
Step 5, the decoding of obtained optimal chromosome is combined into the other parameters under current working, input prediction assessment models obtain
Air-conditioning power consumption after to optimization.
2. a kind of air conditioner energy saving side based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network according to claim 1
Method, it is characterised in that: the data in the step 1 are as follows:
(1) input variable: wherein the input quantity includes outdoor temperature, outside humidity, cooling supply water temperature, cools back water temperature
Degree, freezing supply water temperature, freezing return water temperature, general pipeline flow, month label and time tag;
(2) predictive variable: predictive variable is the air conditioning energy consumption under current working;
(3) data normalization: the sample data of input variable is normalized by following formula
Wherein x' is the data after normalization, and x is input data, and μ is the mean value of data, and σ is the standard deviation of data.
3. a kind of air-conditioning section based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network according to claim 1 or 2
Energy method, it is characterised in that: in the step 1, LSTM-RNN shot and long term remembers Recognition with Recurrent Neural Network prediction model network structure
It is as follows:
(1) indicate that the parameter of memory historical information length, value represent the time sequence that RNN can be utilized in Timesteps:RNN
Column length uses timesteps for 1 with more the model of versatility;
(2) activation primitive: selecting tanh as activation primitive in RNN, it can be the continuous real value of input " compressed " to -1 and 1
Between, if it is very big negative, then output is exactly -1;If it is very big positive number, output is exactly 1;
(3) neuron selects: shot and long term memory LSTM network is the neuron of one of RNN, passes through 3 doors in hidden layer
Structure control retaining for information, can filter out important with unessential information in information flow;
(4) loss function: being modified mean square error MSE, as calculating predicted value predictedtAnd actual value
observertDifference square after, then divided by actual value observert, show that square-error for the ratio of actual value, then carries out
It is average;
(5) other parameters: the hidden layer that model uses 10 layers of LSTM node to constitute, every layer has 50 nodes;Under self-adaption gradient
Drop method, exercise wheel number are 5000 wheels.Batch_size is 30, learning rate 0.00005.
4. a kind of air-conditioning section based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network according to claim 1 or 2
It can method, it is characterised in that: the step 2, in 3,4, genetic algorithm parameter setting and optimization structure are as follows:
(1) chromosome coding: use binary coding as the coding mode of genetic algorithm, in the parameter model that air-conditioning operates normally
Enclose interior setting air-conditioning parameter;
(2) it optimization object function: in the case where operating normally and meeting the precondition of refrigeration demand, needs to reach central air-conditioning and always sets
Standby operation energy consumption is minimum;, using penalty penalty, constraint condition can be subjected to phase according to objective function and constraint condition
The conversion answered, forces the solution of objective function in constraint condition at the problem of Prescribed Properties are become unconfined condition;
Constraint condition:
(TCooling backwater temperature-TWet-bulb temperature)max≥TCooling backwater temperature-TWet-bulb temperature
≥(TCooling backwater temperature-TWet-bulb temperature)min
(TCooling supply water temperature-TWet-bulb temperature)max≥TCold supply and return water temperature-TWet-bulb temperature
≥(TCooling supply water temperature-TWet-bulb temperature)min
Penalty:
Penalty=[max { 0, (2- (TCooling supply water temperature-TCooling supply backwater temperature difference-TWet-bulb temperature))}]2
+ [max { 0, (TCooling supply water temperature-TCooling supply backwater temperature difference-TWet-bulb temperature-6)}]2
+ [max { 0, (4- (TCooling supply water temperature-TWet-bulb temperature))}]2
+ [max { 0, (TCooling supply water temperature-TWet-bulb temperature-8)}]2
Obtain objective function:
F=PPredict power consumption+penalty
(3) fitness function: fitness function can assess all chromosome, and obtained fitness value is to measure current dyeing
The embodiment of the superiority and inferiority degree of body, fitness function formula are as follows:
Fitness=-F
(4) other conditions: being evolved using 50 individual populations, and initialization gene is selected by the way of random roulette
Population primary is selected, using the interleaved mode of single point crossing, crossover probability 80%, 10% gene mutation rate, termination condition is
By the evolution in 100 generations, select the highest gene of fitness value as final optimal gene.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910182274.XA CN109959123A (en) | 2019-03-11 | 2019-03-11 | A kind of energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910182274.XA CN109959123A (en) | 2019-03-11 | 2019-03-11 | A kind of energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109959123A true CN109959123A (en) | 2019-07-02 |
Family
ID=67024215
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910182274.XA Pending CN109959123A (en) | 2019-03-11 | 2019-03-11 | A kind of energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109959123A (en) |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110542178A (en) * | 2019-09-02 | 2019-12-06 | 上海德昂科技有限公司 | Air conditioner and refrigerator room control method and system with self-learning capability |
CN110726216A (en) * | 2019-10-29 | 2020-01-24 | 珠海格力电器股份有限公司 | Air conditioner, control method, device and system thereof, storage medium and processor |
CN110805997A (en) * | 2019-11-14 | 2020-02-18 | 中金新源(天津)科技有限公司 | Energy-saving control method for central air-conditioning system |
CN111126673A (en) * | 2019-12-02 | 2020-05-08 | 珠海格力电器股份有限公司 | Equipment energy consumption prediction method and device and equipment controller |
CN111445065A (en) * | 2020-03-23 | 2020-07-24 | 清华大学 | Energy consumption optimization method and system for refrigeration group control of data center |
CN111523939A (en) * | 2020-04-23 | 2020-08-11 | 腾讯科技(深圳)有限公司 | Promotion content delivery method and device, storage medium and electronic equipment |
CN111538233A (en) * | 2020-05-06 | 2020-08-14 | 上海雁文智能科技有限公司 | Central air conditioner artificial intelligence control method based on energy consumption reward |
CN111780350A (en) * | 2020-05-29 | 2020-10-16 | 浙江工业大学 | Air conditioner energy saving method based on fusion genetic algorithm |
CN112215412A (en) * | 2020-09-27 | 2021-01-12 | 中国农业大学 | Dissolved oxygen prediction method and device |
CN112344521A (en) * | 2020-09-25 | 2021-02-09 | 国网浙江省电力有限公司双创中心 | Energy efficiency improving method for heating and refrigerating system |
CN112413809A (en) * | 2020-11-25 | 2021-02-26 | 珠海格力电器股份有限公司 | Method, device and system for evaluating operation of cold station of central air conditioner |
CN112415924A (en) * | 2020-10-30 | 2021-02-26 | 南京华盾电力信息安全测评有限公司 | Energy-saving optimization method and system for air conditioning system |
CN112729411A (en) * | 2021-01-14 | 2021-04-30 | 金陵科技学院 | Distributed drug warehouse environment monitoring method based on GA-RNN |
CN112883642A (en) * | 2021-02-08 | 2021-06-01 | 上海叠腾网络科技有限公司 | Energy station regional cooling equipment system parameter optimization method, system and medium |
CN112963946A (en) * | 2021-02-26 | 2021-06-15 | 南京邮电大学 | Heating, ventilating and air conditioning system control method and device for shared office area |
CN113283156A (en) * | 2021-03-29 | 2021-08-20 | 北京建筑大学 | Subway station air conditioning system energy-saving control method based on deep reinforcement learning |
CN113432247A (en) * | 2021-05-20 | 2021-09-24 | 中南大学 | Water chilling unit energy consumption prediction method and system based on graph neural network and storage medium |
CN113780405A (en) * | 2021-09-07 | 2021-12-10 | 科希曼电器有限公司 | Air conditioner parameter regression optimization method based on deep neural network |
CN113806996A (en) * | 2020-06-15 | 2021-12-17 | 天津工业大学 | Grate cooler optimization method based on time memory modeling |
CN113947705A (en) * | 2021-10-25 | 2022-01-18 | 四川大学 | Target detection method based on evolutionary neural network under constraint condition |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106765959A (en) * | 2016-12-27 | 2017-05-31 | 武汉虹信技术服务有限责任公司 | Heat-air conditioner energy-saving control method based on genetic algorithm and depth B P neural network algorithms |
US20180252427A1 (en) * | 2017-03-03 | 2018-09-06 | Andreas Hieke | Methods of reducing consumption of energy and other resources associated with operating buildings |
CN109084415A (en) * | 2018-07-26 | 2018-12-25 | 杭州哲达节能科技有限公司 | Central air-conditioning operating parameter optimization method based on artificial neural network and genetic algorithms |
CN109243172A (en) * | 2018-07-25 | 2019-01-18 | 华南理工大学 | Traffic flow forecasting method based on genetic algorithm optimization LSTM neural network |
WO2019017555A1 (en) * | 2017-07-19 | 2019-01-24 | 제주대학교 산학협력단 | System and method for optimizing building energy on basis of indoor environment parameter prediction and dynamic user setup |
-
2019
- 2019-03-11 CN CN201910182274.XA patent/CN109959123A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106765959A (en) * | 2016-12-27 | 2017-05-31 | 武汉虹信技术服务有限责任公司 | Heat-air conditioner energy-saving control method based on genetic algorithm and depth B P neural network algorithms |
US20180252427A1 (en) * | 2017-03-03 | 2018-09-06 | Andreas Hieke | Methods of reducing consumption of energy and other resources associated with operating buildings |
WO2019017555A1 (en) * | 2017-07-19 | 2019-01-24 | 제주대학교 산학협력단 | System and method for optimizing building energy on basis of indoor environment parameter prediction and dynamic user setup |
CN109243172A (en) * | 2018-07-25 | 2019-01-18 | 华南理工大学 | Traffic flow forecasting method based on genetic algorithm optimization LSTM neural network |
CN109084415A (en) * | 2018-07-26 | 2018-12-25 | 杭州哲达节能科技有限公司 | Central air-conditioning operating parameter optimization method based on artificial neural network and genetic algorithms |
Non-Patent Citations (2)
Title |
---|
廖文强等: "基于长短期记忆神经网络的暖通空调系统能耗预测", 《制冷技术》 * |
闫军威等: "商场过渡季节中央空调冷源系统运行参数优化研究", 《制冷与空调》 * |
Cited By (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110542178B (en) * | 2019-09-02 | 2020-06-02 | 上海德昂科技有限公司 | Air conditioner and refrigerator room control method and system with self-learning capability |
CN110542178A (en) * | 2019-09-02 | 2019-12-06 | 上海德昂科技有限公司 | Air conditioner and refrigerator room control method and system with self-learning capability |
CN110726216B (en) * | 2019-10-29 | 2020-10-02 | 珠海格力电器股份有限公司 | Air conditioner, control method, device and system thereof, storage medium and processor |
CN110726216A (en) * | 2019-10-29 | 2020-01-24 | 珠海格力电器股份有限公司 | Air conditioner, control method, device and system thereof, storage medium and processor |
CN110805997A (en) * | 2019-11-14 | 2020-02-18 | 中金新源(天津)科技有限公司 | Energy-saving control method for central air-conditioning system |
CN111126673A (en) * | 2019-12-02 | 2020-05-08 | 珠海格力电器股份有限公司 | Equipment energy consumption prediction method and device and equipment controller |
CN111445065A (en) * | 2020-03-23 | 2020-07-24 | 清华大学 | Energy consumption optimization method and system for refrigeration group control of data center |
CN111523939A (en) * | 2020-04-23 | 2020-08-11 | 腾讯科技(深圳)有限公司 | Promotion content delivery method and device, storage medium and electronic equipment |
CN111523939B (en) * | 2020-04-23 | 2023-11-28 | 腾讯科技(深圳)有限公司 | Popularization content delivery method and device, storage medium and electronic equipment |
CN111538233A (en) * | 2020-05-06 | 2020-08-14 | 上海雁文智能科技有限公司 | Central air conditioner artificial intelligence control method based on energy consumption reward |
CN111780350A (en) * | 2020-05-29 | 2020-10-16 | 浙江工业大学 | Air conditioner energy saving method based on fusion genetic algorithm |
CN113806996B (en) * | 2020-06-15 | 2023-10-17 | 天津工业大学 | Grate cooler optimization method based on time memory modeling |
CN113806996A (en) * | 2020-06-15 | 2021-12-17 | 天津工业大学 | Grate cooler optimization method based on time memory modeling |
CN112344521A (en) * | 2020-09-25 | 2021-02-09 | 国网浙江省电力有限公司双创中心 | Energy efficiency improving method for heating and refrigerating system |
CN112215412A (en) * | 2020-09-27 | 2021-01-12 | 中国农业大学 | Dissolved oxygen prediction method and device |
CN112215412B (en) * | 2020-09-27 | 2023-12-22 | 中国农业大学 | Dissolved oxygen prediction method and device |
CN112415924A (en) * | 2020-10-30 | 2021-02-26 | 南京华盾电力信息安全测评有限公司 | Energy-saving optimization method and system for air conditioning system |
CN112413809B (en) * | 2020-11-25 | 2021-11-30 | 珠海格力电器股份有限公司 | Method, device and system for evaluating operation of cold station of central air conditioner |
CN112413809A (en) * | 2020-11-25 | 2021-02-26 | 珠海格力电器股份有限公司 | Method, device and system for evaluating operation of cold station of central air conditioner |
CN112729411A (en) * | 2021-01-14 | 2021-04-30 | 金陵科技学院 | Distributed drug warehouse environment monitoring method based on GA-RNN |
CN112729411B (en) * | 2021-01-14 | 2022-09-13 | 金陵科技学院 | Distributed drug warehouse environment monitoring method based on GA-RNN |
CN112883642A (en) * | 2021-02-08 | 2021-06-01 | 上海叠腾网络科技有限公司 | Energy station regional cooling equipment system parameter optimization method, system and medium |
CN112883642B (en) * | 2021-02-08 | 2023-10-31 | 上海叠腾网络科技有限公司 | Energy station area cooling equipment system parameter optimization method, system and medium |
CN112963946A (en) * | 2021-02-26 | 2021-06-15 | 南京邮电大学 | Heating, ventilating and air conditioning system control method and device for shared office area |
CN112963946B (en) * | 2021-02-26 | 2022-06-17 | 南京邮电大学 | Heating, ventilating and air conditioning system control method and device for shared office area |
CN113283156B (en) * | 2021-03-29 | 2023-09-15 | 北京建筑大学 | Energy-saving control method for subway station air conditioning system based on deep reinforcement learning |
CN113283156A (en) * | 2021-03-29 | 2021-08-20 | 北京建筑大学 | Subway station air conditioning system energy-saving control method based on deep reinforcement learning |
CN113432247B (en) * | 2021-05-20 | 2022-04-26 | 中南大学 | Water chilling unit energy consumption prediction method and system based on graph neural network and storage medium |
CN113432247A (en) * | 2021-05-20 | 2021-09-24 | 中南大学 | Water chilling unit energy consumption prediction method and system based on graph neural network and storage medium |
CN113780405A (en) * | 2021-09-07 | 2021-12-10 | 科希曼电器有限公司 | Air conditioner parameter regression optimization method based on deep neural network |
CN113947705B (en) * | 2021-10-25 | 2022-08-12 | 四川大学 | Target detection method based on evolutionary neural network under constraint condition |
CN113947705A (en) * | 2021-10-25 | 2022-01-18 | 四川大学 | Target detection method based on evolutionary neural network under constraint condition |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109959123A (en) | A kind of energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network | |
CN111256294B (en) | Model prediction-based optimization control method for combined operation of water chilling unit | |
CN109959122A (en) | A kind of Energy consumption forecast for air conditioning method based on shot and long term memory Recognition with Recurrent Neural Network | |
CN112232980B (en) | Regulation and control method for heat pump unit of regional energy heat supply system | |
CN106920006B (en) | Subway station air conditioning system energy consumption prediction method based on ISOA-LSSVM | |
CN111353656B (en) | Steel enterprise oxygen load prediction method based on production plan | |
CN104566868B (en) | A kind of central air conditioning system and its control method | |
CN109146121A (en) | The power predicating method stopped in the case of limited production based on PSO-BP model | |
CN112415924A (en) | Energy-saving optimization method and system for air conditioning system | |
CN113326651A (en) | Refrigerating station load and energy efficiency ratio dynamic modeling method based on T-S fuzzy model | |
CN103912966A (en) | Optimal control method for ground source heat pump refrigerating system | |
Song et al. | An indoor temperature prediction framework based on hierarchical attention gated recurrent unit model for energy efficient buildings | |
Wang et al. | Multi-criteria comprehensive study on predictive algorithm of heating energy consumption of district heating station based on timeseries processing | |
Deng et al. | Optimal control of chilled water system with ensemble learning and cloud edge terminal implementation | |
CN114186469A (en) | Air conditioning system energy efficiency prediction method and air conditioning system | |
CN111473480A (en) | Central air conditioner energy-saving control method based on decision tree classification | |
CN114611823B (en) | Optimized dispatching method and system for electricity-cold-heat-gas multi-energy-demand typical park | |
CN111291465A (en) | Ship air conditioner thermal comfort temperature prediction method based on genetic algorithm and neural network | |
CN115730228A (en) | Central air conditioner energy consumption analysis method based on BP neural network | |
Jia et al. | Optimization design of industrial water supply pump station considering the influence of atmospheric temperature on operation cost | |
Yang et al. | BP-GA mixed algorithms for short-term load forecasting | |
CN204853838U (en) | Central air conditioning control system | |
Lin et al. | Optimizing for Large Time Delay Systems by BP Neural Network and Evolutionary Algorithm Improving. | |
CN116045461B (en) | Energy-saving control method and device for air-cooled air conditioner based on water supply and return temperature adjustment | |
CN117366810B (en) | Air conditioning system control method and device |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190702 |
|
RJ01 | Rejection of invention patent application after publication |