CN110264004A - A kind of air-conditioning refrigeration duty dynamic prediction method combined based on PSO-BP with Markov chain - Google Patents
A kind of air-conditioning refrigeration duty dynamic prediction method combined based on PSO-BP with Markov chain Download PDFInfo
- Publication number
- CN110264004A CN110264004A CN201910539184.1A CN201910539184A CN110264004A CN 110264004 A CN110264004 A CN 110264004A CN 201910539184 A CN201910539184 A CN 201910539184A CN 110264004 A CN110264004 A CN 110264004A
- Authority
- CN
- China
- Prior art keywords
- moment
- refrigeration duty
- pso
- air
- solar radiation
- 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
- 238000005057 refrigeration Methods 0.000 title claims abstract description 58
- 238000004378 air conditioning Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 title claims abstract description 24
- 230000005855 radiation Effects 0.000 claims abstract description 27
- 238000013528 artificial neural network Methods 0.000 claims abstract description 22
- 238000012937 correction Methods 0.000 claims abstract description 10
- 238000012546 transfer Methods 0.000 claims abstract description 8
- 238000010219 correlation analysis Methods 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims abstract description 7
- 238000001816 cooling Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000003066 decision tree Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 210000004218 nerve net Anatomy 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 230000008092 positive effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 1
- 240000002853 Nelumbo nucifera Species 0.000 description 1
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
A kind of air-conditioning refrigeration duty dynamic prediction method combined based on PSO-BP with Markov chain, comprising the following steps: step 1, classify to idle call energy data；Step 2,10 input variables such as the T moment outdoor temperature of air-conditioning refrigeration duty, T-1 moment outdoor temperature, T moment solar radiation quantity, T-1 moment solar radiation quantity, T-2 moment solar radiation quantity, T moment relative humidity, the moment outdoor T wind speed, T-1 moment refrigeration duty, T-2 moment refrigeration duty, T-4 moment refrigeration duty and output variable T moment refrigeration duty degree of being associated are analyzed；Step 3, load prediction is carried out using PSO-BP neural network；Step 4, burst error is divided using the prediction result of PSO-BP neural network, constructs Markov probability transfer matrix；Step 5, Markov chain error correction is carried out, final predicted value is obtained.Not only in week and weekend with can situation distinguish, correlation analysis also has been carried out to variable relevant to refrigeration duty, input variable of the variable for selecting the degree of association high as model, and error correction is carried out to built-up pattern, the complexity for reducing redundancy and characteristic model, improves the operation efficiency of algorithm.
Description
Technical field
The invention belongs to Air-conditioning Load Prediction field, in particular to a kind of sky combined based on PSO-BP with Markov chain
Adjust refrigeration duty dynamic prediction method.
Background technique
Accounting of the air conditioning energy consumption in the building energy consumption of market is increasing at present, and huge electricity consumption exacerbates power grid
Pressure has researcher to solve the problems, such as this using ice-storage air-conditioning.The operation of ice-storage air-conditioning needs the wave of Proper Match electricity price
The cooling capacity that cold and ice bank should provide in peak decrease amount first carries out the cooling load at each moment in next day building pre-
It surveys, the division of cooling capacity is then carried out according to prediction result.Therefore the dynamic prediction of refrigeration duty is in the core of ice-storage air-conditioning
Hold.
Mainly support vector machines, statistical regression and neural network etc. that traditional air-conditioning cooling load prediction uses, exist
Due to input data and output data there are matching degree is low and built-up pattern there are process errors, cause prediction result up to not
To desirability.
Summary of the invention
The purpose of the present invention is to provide a kind of air-conditioning refrigeration duty dynamic combined based on PSO-BP with Markov chain is pre-
Survey method, to solve the above problems.
To achieve the above object, the invention adopts the following technical scheme:
A kind of air-conditioning refrigeration duty dynamic prediction method combined based on PSO-BP with Markov chain, comprising the following steps:
Step 1, classify to idle call energy data；
Step 2, to the T moment outdoor temperature of air-conditioning refrigeration duty, T-1 moment outdoor temperature, T moment solar radiation quantity, T-1
Moment solar radiation quantity, T-2 moment solar radiation quantity, T moment relative humidity, the moment outdoor T wind speed, T-1 moment refrigeration duty, T-
10 input variables such as 2 moment refrigeration dutys, T-4 moment refrigeration duty and the analysis of output variable T moment refrigeration duty degree of being associated；
Step 3, load prediction is carried out using PSO-BP neural network；
Step 4, burst error is divided using the prediction result of PSO-BP neural network, building Markov probability shifts square
Battle array；
Step 5, Markov chain error correction is carried out, final predicted value is obtained.
Further, the data classification in step 1 is divided for what is built according to market with energy feature, all foot couple air-conditionings
Workload demand much larger than week in demand, will with can data be divided into data and weekend data in week.
Further, input/output variable is analyzed and is picked using JMP Data Analysis Software degree of being associated in step 2
Except the input variable of the low degree of association.
Further, the load prediction in step 3 obtains T moment outdoor temperature, T-1 moment after correlation analysis
Outdoor temperature, T moment solar radiation quantity, T-1 moment solar radiation quantity, T-2 moment solar radiation quantity, T moment relative humidity, T
Moment outdoor wind speed, T-1 moment refrigeration duty, T-2 moment refrigeration duty and T-24 moment refrigeration duty are as PSO-BP neural network
Input, exports as the predicted load at one day corresponding moment.
Further, error correction is to utilize ready-portioned burst error and Markov probability transfer matrix battle array in step 5
It obtains characteristic value and feature vector, revised result is obtained according to characteristic value and feature vector.
Compared with prior art, the present invention has following technical effect:
It is predicted at present for building cooling load, what most researchers used is the method for data-driven, and main includes branch
Hold vector machine (Support Vector Machine, SVM), statistical regression, decision tree, genetic algorithm, neural network algorithm.
SVM is a kind of common artificial intelligence side, has the processing capacity for converting non-linear relation, but it handles data time-consuming too
Long.Statistical regression is a kind of simple and easy prediction technique, but its predictive ability is lower than SVM, and statistical regression methods is pre-
Device is surveyed to be difficult to select.Decision tree method is a kind of technology that group is splitted data into using dendrogram, it can be readily appreciated that still its prediction knot
Fruit is often and actual result has very big deviation, and cannot handle time series and nonlinear data well.Genetic algorithm
It is the powerful optimization processing algorithm for handling complex model problem, when input data amount is big and complicated, which can basis
Objective function or subjective judgement are suitably solved, but it has that not exclusive result and calculating time are big.Nerve net
Network is widely used in every field as a kind of prediction algorithm, and wherein BP (Back Propagation) neural network is powerful with its
Nonlinear Mapping, the obtained extensive use of self study, extensive, fault-tolerant ability, but its there is also local minimizations and convergence
Slow-footed problem.There is researcher using particle swarm algorithm PSO (Particle Swarm Optimization) to BP nerve net
Network carries out the optimization of initial threshold and weight, improves BP mind convergence rate and precision.But due to input data and output number
According to there are matching degree is low and built-up pattern there are process errors, cause prediction result that desirability is not achieved.
The present invention is added JMP Data Analysis Software and Markov chain, utilizes JMP number on the basis of artificial neural network
The analysis of input data degree of being associated can improve neural network according to analysis software, can be picked using correlation analysis
Lead to neural network since the input/output variable degree of correlation is low except the characteristics of input variable low with the output variable degree of coupling makes up
The weakness that training speed is slow and precision of prediction is low, improves its prediction accuracy, therefore can be applied to the dynamic of building cooling load
State prediction, it is contemplated that Combined model forecast result part can also have some relative errors, therefore using Markov chain to prediction
As a result make further amendment to improve precision.
The present invention is based on the store air-conditioning refrigeration duty dynamic prediction methods that PSO-BP neural network is combined with Markov chain
To market with can feature analyzed, not only in week and weekend with energy situation distinguish, also to related to refrigeration duty
Variable carried out correlation analysis, input variable of the variable for selecting the degree of association high as model, and to built-up pattern into
It has gone error correction, has reduced the complexity of redundancy and characteristic model, improve the operation efficiency of algorithm.Experiments verify that in reality
Market building can be reacted in the application on border in time and accurately uses energy situation, plays to the prediction of market building cooling load
Positive effect.
Detailed description of the invention
Fig. 1 is the flow diagram of the method for the invention.
Fig. 2 is in the week of market described in present example and weekend can load diagram.
Fig. 3 is the fitting effect of this hair name model.
Specific embodiment
Below in conjunction with attached drawing, the present invention is further described:
It please refers to Fig.1 to Fig.3, a kind of air-conditioning refrigeration duty dynamic prediction side combined based on PSO-BP with Markov chain
Method, which comprises the following steps:
Step 1, classify to idle call energy data；
Step 2, to the T moment outdoor temperature of air-conditioning refrigeration duty, T-1 moment outdoor temperature, T moment solar radiation quantity, T-1
Moment solar radiation quantity, T-2 moment solar radiation quantity, T moment relative humidity, the moment outdoor T wind speed, T-1 moment refrigeration duty, T-
10 input variables such as 2 moment refrigeration dutys, T-4 moment refrigeration duty and the analysis of output variable T moment refrigeration duty degree of being associated；
Step 3, load prediction is carried out using PSO-BP neural network；
Step 4, burst error is divided using the prediction result of PSO-BP neural network, building Markov probability shifts square
Battle array；
Step 5, Markov chain error correction is carried out, final predicted value is obtained.
Data classification in step 1 is divided for what is built according to market with energy feature, and the load of all foot couple air-conditionings needs
It asks much larger than the demand in week, data and weekend data in week will be divided into energy data.
Input/output variable is analyzed using JMP Data Analysis Software degree of being associated in step 2 and rejects low association
The input variable of degree.
Load prediction in step 3 obtains T moment outdoor temperature, T-1 moment outdoor temperature, T after correlation analysis
Moment solar radiation quantity, T-1 moment solar radiation quantity, T-2 moment solar radiation quantity, T moment relative humidity, the moment outdoor T wind
Speed, the input of T-1 moment refrigeration duty, T-2 moment refrigeration duty and T-24 moment refrigeration duty as PSO-BP neural network, export and are
The predicted load at one day corresponding moment.
Error correction is to obtain characteristic value using ready-portioned burst error and Markov probability transfer matrix battle array in step 5
And feature vector, revised result is obtained according to characteristic value and feature vector.
A kind of air-conditioning refrigeration duty dynamic prediction method combined based on PSO-BP with Markov chain of the present invention, such as Fig. 1 institute
Show comprising following steps:
Step 1, classify to data；
From Fig. 2 it can be found that in market building week and weekend with can situation different from, weekend is with can be far longer than
Energy is used in week.Therefore, distinguish week in and weekend with energy data be more conform with actual conditions.
Step 2, input variable degree of being associated is analyzed；
Since the more and not all input variable of input variable type all has higher association with output variable
Degree, and the input variable of those low degrees of association not only will increase the complexity of model, lead to the reduction of model running speed, can also make
The precision of prediction of model is lower, therefore filters out the variable of the high degree of association as input.
The common input data of air-conditioning cooling load prediction has T moment outdoor temperature, T-1 moment outdoor temperature, the T moment sun
When amount of radiation, T-1 moment solar radiation quantity, T-2 moment solar radiation quantity, T moment relative humidity, the moment outdoor T wind speed, T-1
Refrigeration duty, T-2 moment refrigeration duty, T-24 moment refrigeration duty are carved, using T moment refrigeration duty as output.
There are part input data and the output data degree of association are lower in view of above-mentioned data, to its degree of being associated point
Analysis, and tested by Pearson correlation conspicuousness, wherein [0,0.3] is regarded as, correlation is weaker, and [0.3,0.5] is recognized
Medium for correlation, [0.5,0.7] thinks that correlation is stronger, and [0.7,1.0] thinks that correlation is very strong.Knot after correlation analysis
Fruit is as shown in table 1.
Table 1 respectively inputs the degree of association of parameter and T moment refrigeration duty
Acquired results are analyzed according to table 1, tests in conjunction with Pearson correlation conspicuousness, the T-2 moment can be rejected too
Three positive amount of radiation, the moment outdoor T wind speed, T-4 moment refrigeration duty input variables.
Step 3, load prediction is carried out using PSO-BP neural network；
By step 1 and step 2, all interior use energy data and weekend energy data have been splitted data into, and have been proposed
The input variable low with the output variable degree of association, therefore can be T moment outdoor temperature with the input variable of PSO-BP neural network,
T-1 moment outdoor temperature, T moment solar radiation quantity, T-1 moment solar radiation quantity, T moment relative humidity, T-1 moment are cold negative
Lotus, T-2 moment refrigeration duty, the input variable of model is T moment refrigeration duty, therefore the neural network model is 7-15-1.
Step 4, burst error is divided using the prediction result of PSO-BP neural network, building Markov probability shifts square
Battle array；
Division burst error is carried out to prediction result using Mean-Variance method, then according to ready-portioned section, to error
Classify, construct Markov probability transfer matrix according to classification results, determines that initial vector state in which is come pre- out
Survey period state in which.
Step 5, Markov chain error correction is carried out, final predicted value is obtained；
The state vector that each period is determined according to the burst error that step 4 determines, according to state transfer vector and generally
Rate transfer matrix solves the predicted value through the revised PSO-BP neural network prediction model of Markov chain.
The present invention carries out real finally by the June in the acquisition world Xi'an Sai Ge shopping center and the truthful data in July
Verifying, shows that the model can accurately predict the situation of change of market refrigeration duty in actual application, subtracts to energy conservation
Row plays very big positive effect.
The training data sample number used for first five ten days in the June in market and 60 days July daily early 8 points to evening
Upper 10 points of input variable, test sample are made using 26,27,28,29 4 days data of July.Final evaluation refers to
It is designated as the mean square deviation of predicted value and actual value.Mean square deviation can detect more except linear trend for mean error
The data pattern not described by model such as periodically air-conditioning refrigeration duty in market is predicted the problem of for this paper, in itself
There are certain periodicity, for example summer gradually warms up with weather, and the utilization rate of air-conditioning increases, and the refrigeration duty of air-conditioning can increase therewith
Greatly, the most hot time has been served as, load can slowly be restored to reduced levels.
Using after training PSO-BP and the model that combines of Markov chain analyzed in test data set, test number
According to fitting effect as shown in figure 3, abscissa is sampled point, ordinate is air-conditioning cold load value, and blue is that true power grid is negative
Charge values, red are the network load value of prediction.
A kind of air-conditioning refrigeration duty dynamic prediction method combined based on PSO-BP with Markov chain of the present invention, in advance
The specific load value measured and true value have a gap, but the variation tendency that the entirety of predicted value curve a raises and reduces
It has been able to accurately reflect true load curve b very much.
It is as shown in table 2 with untreated model comparing result, and wherein model 1 is not differentiate between in week to carry out with weekend data
Prediction, model 2 are to distinguish weekend data in week but do not reject the prediction that the low input data of the degree of association carries out, and model 3 is both
It has distinguished all interior weekend data and has also eliminated the prediction carried out after the low input data of the degree of association, model 4 is by processing and to miss
The prediction carried out after difference amendment.
3 prediction model root-mean-square error of table
Basic principles and main features and advantage of the invention have been shown and described above.The technical staff of the industry should
Understand, the present invention is not limited to the above embodiments, and the above embodiments and description only describe originals of the invention
Reason, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, these changes and improvements
It all fall within the protetion scope of the claimed invention.The claimed scope of the invention is by appended claims and its equivalent circle
It is fixed.
Claims (5)
1. a kind of air-conditioning refrigeration duty dynamic prediction method combined based on PSO-BP with Markov chain, which is characterized in that including
Following steps:
Step 1, classify to idle call energy data；
Step 2, to the T moment outdoor temperature of air-conditioning refrigeration duty, T-1 moment outdoor temperature, T moment solar radiation quantity, T-1 moment
When solar radiation quantity, T-2 moment solar radiation quantity, T moment relative humidity, the moment outdoor T wind speed, T-1 moment refrigeration duty, T-2
Refrigeration duty, 10 input variables of T-4 moment refrigeration duty and output variable T moment refrigeration duty degree of being associated is carved to analyze；
Step 3, load prediction is carried out using PSO-BP neural network；
Step 4, burst error is divided using the prediction result of PSO-BP neural network, constructs Markov probability transfer matrix；
Step 5, Markov chain error correction is carried out, final predicted value is obtained.
2. a kind of air-conditioning refrigeration duty dynamic prediction side combined based on PSO-BP with Markov chain according to claim 1
Method, which is characterized in that the data classification in step 1 is divided for what is built according to market with energy feature, all foot couple air-conditionings
Workload demand will be divided into data and weekend data in week with energy data much larger than the demand in week.
3. a kind of air-conditioning refrigeration duty dynamic prediction side combined based on PSO-BP with Markov chain according to claim 1
Method, which is characterized in that input/output variable is analyzed and rejected using JMP Data Analysis Software degree of being associated in step 2
The input variable of the low degree of association.
4. a kind of air-conditioning refrigeration duty dynamic prediction side combined based on PSO-BP with Markov chain according to claim 1
Method, which is characterized in that the load prediction in step 3 obtains T moment outdoor temperature, T-1 moment outdoor after correlation analysis
Temperature, T moment solar radiation quantity, T-1 moment solar radiation quantity, T-2 moment solar radiation quantity, T moment relative humidity, T moment
The input of outdoor wind speed, T-1 moment refrigeration duty, T-2 moment refrigeration duty and T-24 moment refrigeration duty as PSO-BP neural network,
Output is the predicted load at one day corresponding moment.
5. a kind of air-conditioning refrigeration duty dynamic prediction side combined based on PSO-BP with Markov chain according to claim 1
Method, which is characterized in that error correction is to be obtained using ready-portioned burst error and Markov probability transfer matrix battle array in step 5
Characteristic value and feature vector obtain revised result according to characteristic value and feature vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910539184.1A CN110264004A (en) | 2019-06-20 | 2019-06-20 | A kind of air-conditioning refrigeration duty dynamic prediction method combined based on PSO-BP with Markov chain |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910539184.1A CN110264004A (en) | 2019-06-20 | 2019-06-20 | A kind of air-conditioning refrigeration duty dynamic prediction method combined based on PSO-BP with Markov chain |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110264004A true CN110264004A (en) | 2019-09-20 |
Family
ID=67919966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910539184.1A Pending CN110264004A (en) | 2019-06-20 | 2019-06-20 | A kind of air-conditioning refrigeration duty dynamic prediction method combined based on PSO-BP with Markov chain |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110264004A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111340305A (en) * | 2020-03-10 | 2020-06-26 | 西安建筑科技大学 | Building operation energy consumption prediction method |
| CN111649457A (en) * | 2020-05-13 | 2020-09-11 | 中国科学院广州能源研究所 | Dynamic predictive machine learning type air conditioner energy-saving control method |
| CN111723085A (en) * | 2020-06-28 | 2020-09-29 | 山东建筑大学 | Electric energy consumption data cleaning method and system for air conditioner of public building |
| CN112330077A (en) * | 2021-01-04 | 2021-02-05 | 南方电网数字电网研究院有限公司 | Power load prediction method, power load prediction device, computer equipment and storage medium |
| CN113177675A (en) * | 2021-05-28 | 2021-07-27 | 西安建筑科技大学 | Air conditioner cold load prediction method based on optimization neural network of longicorn group algorithm |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06147598A (en) * | 1992-11-09 | 1994-05-27 | Takasago Thermal Eng Co Ltd | Air conditioning load prediction method |
| CN102779228A (en) * | 2012-06-07 | 2012-11-14 | 华南理工大学 | Method and system for online prediction on cooling load of central air conditioner in marketplace buildings |
| CN104008427A (en) * | 2014-05-16 | 2014-08-27 | 华南理工大学 | Central air conditioner cooling load prediction method based on BP neural network |
-
2019
- 2019-06-20 CN CN201910539184.1A patent/CN110264004A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06147598A (en) * | 1992-11-09 | 1994-05-27 | Takasago Thermal Eng Co Ltd | Air conditioning load prediction method |
| CN102779228A (en) * | 2012-06-07 | 2012-11-14 | 华南理工大学 | Method and system for online prediction on cooling load of central air conditioner in marketplace buildings |
| CN104008427A (en) * | 2014-05-16 | 2014-08-27 | 华南理工大学 | Central air conditioner cooling load prediction method based on BP neural network |
| WO2015172560A1 (en) * | 2014-05-16 | 2015-11-19 | 华南理工大学 | Central air conditioner cooling load prediction method based on bp neural network |
Non-Patent Citations (3)
| Title |
|---|
| 杨熊等: ""基于改进PSO-BP神经网络的冰蓄冷空调冷负荷动态预测模型"", 《土木与环境工程学报(中英文)》 * |
| 王敏华等: "《质量数据分析技术》", 31 May 2016, 中国质检出版社 * |
| 陈娟等: ""基于马尔科夫链对BP神经网络法残差修正模型的短期风电功率预测"", 《电力与能源》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111340305A (en) * | 2020-03-10 | 2020-06-26 | 西安建筑科技大学 | Building operation energy consumption prediction method |
| CN111649457A (en) * | 2020-05-13 | 2020-09-11 | 中国科学院广州能源研究所 | Dynamic predictive machine learning type air conditioner energy-saving control method |
| CN111723085A (en) * | 2020-06-28 | 2020-09-29 | 山东建筑大学 | Electric energy consumption data cleaning method and system for air conditioner of public building |
| CN111723085B (en) * | 2020-06-28 | 2023-06-23 | 山东建筑大学 | Method and system for cleaning electric energy consumption data of public building air conditioner |
| CN112330077A (en) * | 2021-01-04 | 2021-02-05 | 南方电网数字电网研究院有限公司 | Power load prediction method, power load prediction device, computer equipment and storage medium |
| CN113177675A (en) * | 2021-05-28 | 2021-07-27 | 西安建筑科技大学 | Air conditioner cold load prediction method based on optimization neural network of longicorn group algorithm |
| CN113177675B (en) * | 2021-05-28 | 2023-05-23 | 西安建筑科技大学 | Air conditioner cooling load prediction method based on longicorn group algorithm optimization neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110264004A (en) | A kind of air-conditioning refrigeration duty dynamic prediction method combined based on PSO-BP with Markov chain | |
| US11581740B2 (en) | Method, system and storage medium for load dispatch optimization for residential microgrid | |
| CN105184678A (en) | Method for constructing photovoltaic power station generation capacity short-term prediction model based on multiple neural network combinational algorithms | |
| Tian et al. | Predictive model of energy consumption for office building by using improved GWO-BP | |
| CN110490385A (en) | The unified prediction of electric load and thermic load in a kind of integrated energy system | |
| CN109376950A (en) | A kind of polynary Load Forecasting based on BP neural network | |
| CN104408562A (en) | Photovoltaic system generating efficiency comprehensive evaluation method based on BP (back propagation) neural network | |
| CN109636054A (en) | Solar energy power generating amount prediction technique based on classification and error combination prediction | |
| CN107679687A (en) | A kind of photovoltaic output modeling method and Generation System Reliability appraisal procedure | |
| CN113762387B (en) | Multi-element load prediction method for data center station based on hybrid model prediction | |
| CN112149890A (en) | Comprehensive energy load prediction method and system based on user energy label | |
| Li et al. | Photovoltaic array prediction on short-term output power method in centralized power generation system | |
| He et al. | A power forecasting approach for PV plant based on irradiance index and LSTM | |
| CN112288157A (en) | Wind power plant power prediction method based on fuzzy clustering and deep reinforcement learning | |
| CN114997508A (en) | Greenhouse electricity utilization optimization method and system based on multi-energy complementation | |
| CN117272850B (en) | Elastic space analysis method for safe operation scheduling of power distribution network | |
| CN112801343A (en) | Energy storage system capacity planning method considering multi-meteorological-scene adaptive cost | |
| CN109149644B (en) | Light-storage integrated online strategy matching and collaborative optimization method based on big data analysis | |
| Chen et al. | Accounting information disclosure and financial crisis beforehand warning based on the artificial neural network | |
| Zhang et al. | Prediction of Photovoltaic Power Generation Based on D-vine Copula Model in Typical Climates | |
| CN110212520A (en) | A kind of power predicating method based on convolutional neural networks | |
| Xin | Photovoltaic Power Prediction Based on RBF Neural Network Optimized by Gray Wolf Algorithm | |
| CN112365025B (en) | Renewable energy market long-term heat storage method supporting vector machine width learning | |
| CN114548804A (en) | Day-ahead power prediction method and system based on improved lifting tree | |
| Li et al. | A C-DCGAN-based method for generating extreme risk scenarios of high percentage new energy systems |
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: 20190920 |
|
| RJ01 | Rejection of invention patent application after publication |