CN105205558A - BP neural network model configuration method oriented to building energy consumption prediction - Google Patents
BP neural network model configuration method oriented to building energy consumption prediction Download PDFInfo
- Publication number
- CN105205558A CN105205558A CN201510613729.0A CN201510613729A CN105205558A CN 105205558 A CN105205558 A CN 105205558A CN 201510613729 A CN201510613729 A CN 201510613729A CN 105205558 A CN105205558 A CN 105205558A
- Authority
- CN
- China
- Prior art keywords
- prediction
- network model
- energy consumption
- training
- neural network
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a BP neural network model configuration method oriented to building energy consumption prediction. The method comprises the steps that A, a configurable BP neural network model is constructed; B, the BP neural network model with an energy consumption mode embedded is trained, and a prediction training model is formed; C, building energy consumption data prediction is carried out. According to the BP neural network model configuration method for building energy consumption prediction, complex building energy consumption prediction related influence factors, parameters and algorithms and the building energy consumption actual prediction function are subjected to decoupling separation, configurable construction of any prediction model is achieved, and the development and maintenance cost of a building energy consumption prediction system is reduced, so that an energy consumption prediction system of a large-scale building is easy to achieve, and good application prospects are achieved.
Description
Technical field
The present invention relates to a kind of BP neural network model collocation method towards building energy consumption prediction, belong to building energy consumption electric powder prediction.
Background technology
Along with the sharp increase of Chinese population and improving constantly of people's living standard, China's floor area of building is multiplied, and building energy consumption expense also constantly increases.According to statistics, China's building energy consumption accounts for 1/3 of whole society's total energy consumption.At present, building energy consumption prediction, mainly according to history consumption information and relevant context information, is may predicting by energy situation of future architecture.Building energy consumption prediction can not only help managerial personnel's reasonable arrangement system operation mode, but also can utilize comparing of actual value and predicted value, carries out building energy consumption abnormality detection, thus the safety of guarantee energy is carried out.
In recent years, the research about building energy consumption Forecasting Methodology had occurred that much the method be mainly concerned with had BP neural network algorithm, support vector machine, Time series analysis method etc.But existing method is that object is implemented to carry out mainly with single building, single energy consumption, and result of study all finally provides a good empirical model as optimum prediction model.Also have, although the theoretical method research of building energy consumption prediction is more, but, but there is blank in actual building energy consumption prognoses system, time mainly due to actual building energy consumption prediction, for architectural object scale more, energy consumption wide variety, for different buildings, different classification subitem energy consumption forecast models is then different, single forecast model given by theoretical research is difficult to meet the changeable prognoses system requirement of actual complex, therefore, how to set up a kind of effective, flexibly, configurable forecast model configuration design proposal meets the changeable demand of actual prediction system complex, it is current urgent problem.
Summary of the invention
The object of the invention is the problem being difficult to meet the changeable prognoses system requirement of actual complex in order to overcome existing Individual forecast model.BP neural network model collocation method towards building energy consumption prediction of the present invention, the building energy consumption of complexity prediction Correlative Influence Factors, parameter, algorithm are carried out decoupling zero with building energy consumption actual prediction function and be separated, configurableization realizing any forecast model creates, reduce the development and maintenance cost of building energy consumption prognoses system, make the Energy Consumption Prediction System of extensive building be easy to realize, have a good application prospect.
In order to achieve the above object, the technical solution adopted in the present invention is:
Towards a BP neural network model collocation method for building energy consumption prediction, it is characterized in that: comprise the following steps,
Step (A), creates the BP neural network model of configurableization, comprises and sets up input configuration, output configuration, network structure configuration, and carries out model checking, the preservation of basic network model;
Step (B), training embeds the BP neural network model of power consumption mode, forms prediction training pattern, comprises and predicts that training pattern is selected, prediction input data time is arranged, building energy consumption is predicted, predict the outcome preservation;
Step (C), carries out building energy consumption data prediction, comprises the selection of prediction training pattern, the setting of prediction input data time, building energy consumption prediction, predict the outcome preservation.
The aforesaid BP neural network model collocation method towards building energy consumption prediction, is characterized in that: step (A), creates the BP neural network model of configurableization, comprise following process,
(A1) carry out abstract to the influence factor of energy consumption prediction, define the input storage mode of BP neural network model, analysis mode and Data import mode;
(A2) storage mode, analysis mode and Data import mode that BP neural network model exports is defined;
(A3) the scope screening of BP neural network model is carried out;
(A4) input configuration is set up, screen FR input factor according to the scope of BP neural network model to realize inputting arbitrarily selecting factors, the arbitrarily selection of input number, meanwhile, extract to the related data involved by each input factor the parameter related to arrange;
(A5) output configuration is set up, screen FR output factor according to the scope of BP neural network model to realize exporting arbitrarily selecting factors, exporting number selection arbitrarily, meanwhile, extract to the related data involved by each output factor the parameter related to arrange;
(A6) set up network structure configuration, comprise the configuration of the neuron number to input layer, hidden layer neuron number, output layer neuron number, cycle of training, predetermined period;
(A7) carry out model checking, the BP neural network model configured is verified, comprise input and output configuration and verify with mating of network structure, inputoutput data extracting parameter configuration verification;
(A8) carry out the preservation of basic network model, comprise and preserve network structure configuration information, input configuration information, input configuration information.
The aforesaid BP neural network model collocation method towards building energy consumption prediction, is characterized in that: step (B), training embeds the BP neural network model of power consumption mode, forms prediction training pattern, comprises following process,
(B1) power consumption mode configuration carries out filtration according to the energy consumption model of festivals or holidays, situation of whether going to work to training data to extract training, ensures the accuracy of prediction BP neural network model; Training parameter is arranged, and comprises the setting of weight correction factor, learning rate, training algebraically; Training data extraction time arrange, to training data extract start time and the end time arrange, when carrying out hands-on, then this time range according to select BP neural network model configure input, export carry out corresponding data extraction;
(B2) basic network Model Selection, selection selects the BP neural network model kept in step (A);
(B3) to the basic network model training selected, form prediction training pattern, according to the power consumption mode configured, training parameter, training data extraction time scope, BP neural network algorithm is utilized to choose training to the basic network model selected;
(B4) preservation of training pattern is predicted, related data before training, after training is preserved, related data comprises basic network model ID, training time, training data extraction start time, training data extraction end time, weight correction factor, learning rate, normalization maximal value, normalization minimum value, trains algebraically, input layer to the weight information of hidden layer, and hidden layer is to the weight information of output layer.
The aforesaid BP neural network model collocation method towards building energy consumption prediction, is characterized in that: step (C), carries out building energy consumption data prediction, comprise following process,
(C1) predict that training pattern is selected, the prediction training pattern that step (B) has trained is selected, and to needing the time period predicted to arrange;
(C2) building energy consumption prediction, according to the prediction training pattern selected, utilizes BP neural network algorithm to carry out energy consumption prediction;
(C3) predict the outcome preservation, preserves, be convenient to data display and tracking, comprise the training pattern ID of selection, predicted time, prediction of energy consumption value to the data that building energy consumption dopes.
The invention has the beneficial effects as follows: the BP neural network model collocation method towards building energy consumption prediction of the present invention, the building energy consumption of complexity prediction Correlative Influence Factors, parameter, algorithm are carried out decoupling zero with building energy consumption actual prediction function and be separated, configurableization realizing any forecast model creates, reduce the development and maintenance cost of building energy consumption prognoses system, make the Energy Consumption Prediction System of extensive building be easy to realize, have a good application prospect.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the BP neural network model collocation method towards building energy consumption prediction of the present invention.
Fig. 2 is the process flow diagram forming prediction training pattern of the present invention.
Fig. 3 is the process flow diagram carrying out building energy consumption data prediction of the present invention.
Embodiment
Below in conjunction with Figure of description, the invention will be further described.Following examples only for technical scheme of the present invention is clearly described, and can not limit the scope of the invention with this.
BP neural network model collocation method towards building energy consumption prediction of the present invention, the building energy consumption of complexity prediction Correlative Influence Factors, parameter, algorithm are carried out decoupling zero with building energy consumption actual prediction function and be separated, configurableization realizing any forecast model creates, reduce the development and maintenance cost of building energy consumption prognoses system, the Energy Consumption Prediction System of extensive building is made to be easy to realize, as shown in Figure 1, comprise the following steps
Step (A), creates the BP neural network model of configurableization, comprises and sets up input configuration, output configuration, network structure configuration, and carries out model checking, the preservation of basic network model, and constructive process is as follows,
(A1) carry out abstract to the influence factor of energy consumption prediction, define the input storage mode of BP neural network model, analysis mode and Data import mode;
(A2) storage mode, analysis mode and Data import mode that BP neural network model exports is defined;
(A3) the scope screening of BP neural network model is carried out;
(A4) input configuration is set up, screen FR input factor according to the scope of BP neural network model to realize inputting arbitrarily selecting factors, the arbitrarily selection of input number, meanwhile, extract to the related data involved by each input factor the parameter related to arrange;
(A5) output configuration is set up, screen FR output factor according to the scope of BP neural network model to realize exporting arbitrarily selecting factors, exporting number selection arbitrarily, meanwhile, extract to the related data involved by each output factor the parameter related to arrange;
(A6) set up network structure configuration, comprise the configuration of the neuron number to input layer, hidden layer neuron number, output layer neuron number, cycle of training, predetermined period;
(A7) carry out model checking, the BP neural network model configured is verified, comprise input and output configuration and verify with mating of network structure, inputoutput data extracting parameter configuration verification;
(A8) carry out the preservation of basic network model, comprise and preserve network structure configuration information, input configuration information, input configuration information;
Step (B), training embeds the BP neural network model of power consumption mode, forms prediction training pattern, and comprise and predict that training pattern is selected, prediction input data time is arranged, building energy consumption is predicted, predict the outcome preservation, forming process is as follows,
(B1) power consumption mode configuration carries out filtration according to the energy consumption model of festivals or holidays, situation of whether going to work to training data to extract training, ensures the accuracy of prediction BP neural network model; Training parameter is arranged, and comprises the setting of weight correction factor, learning rate, training algebraically; Training data extraction time arrange, to training data extract start time and the end time arrange, when carrying out hands-on, then this time range according to select BP neural network model configure input, export carry out corresponding data extraction;
(B2) basic network Model Selection, selection selects the BP neural network model kept in step (A);
(B3) to the basic network model training selected, form prediction training pattern, according to the power consumption mode configured, training parameter, training data extraction time scope, BP neural network algorithm is utilized to choose training to the basic network model selected;
Step (C), carries out building energy consumption data prediction, and comprise the selection of prediction training pattern, the setting of prediction input data time, building energy consumption prediction, predict the outcome preservation, forecasting process is as follows,
(C1) predict that training pattern is selected, the prediction training pattern that step (B) has trained is selected, and to needing the time period predicted to arrange;
(C2) building energy consumption prediction, according to the prediction training pattern selected, utilizes BP neural network algorithm to carry out energy consumption prediction;
(C3) predict the outcome preservation, preserves, be convenient to data tracing, comprise the training pattern ID of selection, predicted time, prediction of energy consumption value to the data that building energy consumption dopes.
Above-mentioned step (A)-step (C), the BP neural network model set up, formation prediction training pattern, select prediction training pattern, there is during execution sequencing relation, call relation between model is the relation of one-to-many, namely a BP neural network model may correspond to multiple prediction training pattern, and a prediction training pattern can carry out the prediction of multiple data.
As shown in Figure 2, step (B) training embeds the BP neural network model of power consumption mode, form the process flow diagram of prediction training pattern, basic network model is resolved, carry out data extraction according to the input and output factor combined training data extraction time after resolving, then carry out data prediction successively, export the steps such as data normalization, power consumption mode filtration, BP neural metwork training, training pattern preservation.
As shown in Figure 3, step (C) carries out the process flow diagram of building energy consumption data prediction, according to the prediction training pattern trained, utilizes BP neural network algorithm to predict.
According to the BP neural network model collocation method towards building energy consumption prediction of the present invention, introduce an embodiment,
1) the input factor of BP neural network model is described, the classification of definition BP neural network model input has two-layer, and ground floor is classified, as temperature data, energy consumption data etc. according to factor classification, the second layer Main Basis time itemizes, and input factor describes as shown in table 1:
Table 1
2) output of BP neural network model is described, the classification that definition forecast model exports has two-layer, ground floor is classified according to attribute classification, as single building energy consumption data, garden energy consumption data etc., the second layer Main Basis time itemizes, and output factor describes as shown in table 2:
Table 2
3) input of network structure configuration adopts consolidation form definition, stores and resolve, input building mode is, separate with ' & ' between different inputs, ' | ' is utilized to separate between correlation parameter in input inside, in parameter, Section 1 presentation class, Section 2 represent subitem, first two form the data item needing to extract altogether, 3rd represents to Section 6 the parameter extracted required for data item, and the number of input neuron is the sum of ' & '.
As, be input as & 1|1|110101||| & 2|1|320100A000|01000||, it represents that input parameter has two; First is input as the temperature data of current hour, and the temperature data of needs belongs to the temperature in 110101 cities; Second is input as the energy consumption data of last hour, and it is 320100A000 01000 by an electric energy consumption that concrete data come from building number;
The analytic method of input is corresponding with construction method, and correlation parameter item is extracted in the separation according to relevant character ' & ' and ' | ';
The load mode of input data is: determine according to the classification of input after resolving and subitem information the tables of data Table_A that needs to extract and extract data rows Column_A, relate to Argument List Column_B and Column_C, then extracting data according to the 3 to the 4 given parameter Param1, Param2;
Extraction form is: SelectColumn_AfromTable_AWhereColumn_Blike " Param1 " andColumn_Clike " Param2 ";
4) output of network structure configuration adopts consolidation form definition, stores and resolve, output building mode is: separate with ' & ' between different inputs, export in inside and between correlation parameter, utilize ' | ' to separate, in parameter, Section 1 presentation class, Section 2 represent subitem, first two form the data item needing to extract altogether, and the 3rd represents to Section 6 the parameter extracted required for data item;
As, export as & 1|1|320100A000|01000||, represent that the data building number needing prediction is 320100A000, energy consumption is categorized as the power load of 01000;
The analytic method exported is corresponding with construction method, and correlation parameter item is extracted in the separation according to relevant character ' & ' and ' | ';
Exporting the load mode of data is: determine according to the classification of the output after resolving and subitem information the tables of data Table_A that needs to extract and extract data rows Column_A, relate to Argument List Column_B and Column_C, then extracting data according to the 3 to the 4 given parameter Param1, Param2;
Extraction form is: SelectColumn_AfromTable_AWhereColumn_Blike " Param1 " andColumn_Clike " Param2 ";
5) input layer adopts consolidation form definition to the weight of hidden layer, hidden layer to the weight of output layer, stores and resolve, character ' & ' is utilized to separate between the row of weight matrix and row, the character ' | ' that utilizes between the column and the column in row separates, the number of ' & ' character is total line number of weight matrix, the number of ' | ' character between front and back continuous print two characters ' & ' is total columns of weight matrix
As, input layer is w to the weight matrix of hidden layer:
After building, weight file layout is:
&0.948|0.802|-2.052|2.662|&0.199|0.989|6.614|0.628|&0.576|0.755|0.712|0.650|
The analytic method of weight is corresponding with construction method, and contiguous items is extracted in the separation according to relevant character ' & ' and ' | ';
6) define power consumption mode 5 character bits to represent, front two represents a hour start time, the 3rd, the 4 bit representations hour end time, the 5th for representing whether festivals or holidays, as shown in table 3:
Table 3
Consider, with the forecast model of power consumption modes such as coming off duty festivals or holidays or working, in order to ensure compatibility and the extensibility of forecast model, after data prediction, to carry out data filtering according to energy consumption model.If power consumption mode is 20081, then all data for the setup times section extracted filters, and will be more than or equal to 20 points the time period, and the data being less than the festivals or holidays of 08 retain, and other data are removed.
7), during manual model training, from top to bottom can perform successively according to Fig. 2 building energy consumption forecast model training managing flow process; For automodel training, model training adopts period treatment mechanism, defines adhoc basis model i item cycle of training in basic model saving interface
with training time item next time
adopting the testing mechanism being less than the short time of cycle of training, constantly train, when arriving training time next time, then training, after training, change training time next time is
continue training to detect;
8), during manual energy consumption data prediction test, from top to bottom can perform successively according to Fig. 3 building energy consumption forecast model data prediction treatment scheme; For automatic building energy consumption data prediction, adopt period treatment mechanism, in basic model saving interface, define predetermined period item of adhoc basis model i
with next predicted time item
adopt the testing mechanism being less than the short time of predetermined period, constantly predict, when arriving next predicted time, predict, changing next predicted time after prediction is
continue training to detect.
More than show and describe ultimate principle of the present invention, principal character and advantage.The technician of the industry should understand; the present invention is not restricted to the described embodiments; what describe in above-described embodiment and instructions just illustrates principle of the present invention; without departing from the spirit and scope of the present invention; the present invention also has various changes and modifications, and these changes and improvements all fall in the claimed scope of the invention.Application claims protection domain is defined by appending claims and equivalent thereof.
Claims (4)
1., towards a BP neural network model collocation method for building energy consumption prediction, it is characterized in that: comprise the following steps,
Step (A), creates the BP neural network model of configurableization, comprises and sets up input configuration, output configuration, network structure configuration, and carries out model checking, the preservation of basic network model;
Step (B), training embeds the BP neural network model of power consumption mode, forms prediction training pattern, comprises and predicts that training pattern is selected, prediction input data time is arranged, building energy consumption is predicted, predict the outcome preservation;
Step (C), carries out building energy consumption data prediction, comprises the selection of prediction training pattern, the setting of prediction input data time, building energy consumption prediction, predict the outcome preservation.
2. the BP neural network model collocation method towards building energy consumption prediction according to claim 1, is characterized in that: step (A), creates the BP neural network model of configurableization, comprise following process,
(A1) carry out abstract to the influence factor of energy consumption prediction, define the input storage mode of BP neural network model, analysis mode and Data import mode;
(A2) storage mode, analysis mode and Data import mode that BP neural network model exports is defined;
(A3) the scope screening of BP neural network model is carried out;
(A4) input configuration is set up, screen FR input factor according to the scope of BP neural network model to realize inputting arbitrarily selecting factors, the arbitrarily selection of input number, meanwhile, extract to the related data involved by each input factor the parameter related to arrange;
(A5) output configuration is set up, screen FR output factor according to the scope of BP neural network model to realize exporting arbitrarily selecting factors, exporting number selection arbitrarily, meanwhile, extract to the related data involved by each output factor the parameter related to arrange;
(A6) set up network structure configuration, comprise the configuration of the neuron number to input layer, hidden layer neuron number, output layer neuron number, cycle of training, predetermined period;
(A7) carry out model checking, the BP neural network model configured is verified, comprise input and output configuration and verify with mating of network structure, inputoutput data extracting parameter configuration verification;
(A8) carry out the preservation of basic network model, comprise and preserve network structure configuration information, input configuration information, input configuration information.
3. the BP neural network model collocation method towards building energy consumption prediction according to claim 1, is characterized in that: step (B), and training embeds the BP neural network model of power consumption mode, forms prediction training pattern, comprises following process,
(B1) power consumption mode configuration carries out filtration according to the energy consumption model of festivals or holidays, situation of whether going to work to training data to extract training, ensures the accuracy of prediction BP neural network model; Training parameter is arranged, and comprises the setting of weight correction factor, learning rate, training algebraically; Training data extraction time arrange, to training data extract start time and the end time arrange, when carrying out hands-on, then this time range according to select BP neural network model configure input, export carry out corresponding data extraction;
(B2) basic network Model Selection, selection selects the BP neural network model kept in step (A);
(B3) to the basic network model training selected, form prediction training pattern, according to the power consumption mode configured, training parameter, training data extraction time scope, utilize BP neural network algorithm to choose training to the basic network model selected;
(B4) preservation of training pattern is predicted, related data before training, after training is preserved, related data comprises basic network model ID, training time, training data extraction start time, training data extraction end time, weight correction factor, learning rate, normalization maximal value, normalization minimum value, trains algebraically, input layer to the weight information of hidden layer, and hidden layer is to the weight information of output layer.
4. the BP neural network model collocation method towards building energy consumption prediction according to claim 1, is characterized in that: step (C), carries out building energy consumption data prediction, comprise following process,
(C1) predict that training pattern is selected, the prediction training pattern that step (B) has trained is selected, and to needing the time period predicted to arrange;
(C2) building energy consumption prediction, according to the prediction training pattern selected, utilizes BP neural network algorithm to carry out energy consumption prediction;
(C3) predict the outcome preservation, preserves, be convenient to data display and tracking, comprise the training pattern ID of selection, predicted time, prediction of energy consumption value to the data that building energy consumption dopes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510613729.0A CN105205558B (en) | 2015-09-23 | 2015-09-23 | A kind of BP neural network model configuration method towards building energy consumption prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510613729.0A CN105205558B (en) | 2015-09-23 | 2015-09-23 | A kind of BP neural network model configuration method towards building energy consumption prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105205558A true CN105205558A (en) | 2015-12-30 |
| CN105205558B CN105205558B (en) | 2018-11-06 |
Family
ID=54953227
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510613729.0A Active CN105205558B (en) | 2015-09-23 | 2015-09-23 | A kind of BP neural network model configuration method towards building energy consumption prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105205558B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106199174A (en) * | 2016-07-01 | 2016-12-07 | 广东技术师范学院 | Extruder energy consumption predicting abnormality method based on transfer learning |
| CN106874581A (en) * | 2016-12-30 | 2017-06-20 | 浙江大学 | A kind of energy consumption of air conditioning system in buildings Forecasting Methodology based on BP neural network model |
| CN106910144A (en) * | 2017-01-11 | 2017-06-30 | 华南理工大学 | Based on timesharing carve it is actual with can coefficient heavy construction by when energy consumption on-line prediction method |
| CN106933173A (en) * | 2017-03-06 | 2017-07-07 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | A kind of equipment for dyeing and finishing monitoring system based on CORBA |
| CN107844859A (en) * | 2017-10-31 | 2018-03-27 | 深圳达实智能股份有限公司 | Large medical equipment energy consumption Forecasting Methodology and terminal device based on artificial intelligence |
| CN109033595A (en) * | 2018-07-16 | 2018-12-18 | 清华大学 | Build Green performance sunykatuib analysis new method and device |
| CN109035067A (en) * | 2018-07-19 | 2018-12-18 | 北京泰豪智能工程有限公司 | Building energy consumption processing method and processing device based on RF and ARMA algorithm |
| CN109102103A (en) * | 2018-06-26 | 2018-12-28 | 上海鲁班软件股份有限公司 | A kind of multi-class energy consumption prediction technique based on Recognition with Recurrent Neural Network |
| WO2019114517A1 (en) * | 2017-12-13 | 2019-06-20 | 腾讯科技（深圳）有限公司 | Neural network model deployment method, prediction method, and apparatus |
| CN110569981A (en) * | 2019-07-22 | 2019-12-13 | 国网冀北电力有限公司电力科学研究院 | Method and device for determining fuel consumption of scattered coal |
| CN111752870A (en) * | 2020-05-28 | 2020-10-09 | 苏州浪潮智能科技有限公司 | Compatible method and system of server interface based on neural network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102938092A (en) * | 2012-10-08 | 2013-02-20 | 珠海派诺科技股份有限公司 | Prediction method of building energy consumption in festivals and holidays based on neural network |
| US20130325377A1 (en) * | 2012-05-31 | 2013-12-05 | Johnson Controls Technology Company | Systems and methods for measuring and verifying energy usage in a building |
| CN104239982A (en) * | 2014-10-12 | 2014-12-24 | 刘岩 | Method for predicting energy consumption of buildings during holidays and festivals on basis of time series and neural networks |
-
2015
- 2015-09-23 CN CN201510613729.0A patent/CN105205558B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130325377A1 (en) * | 2012-05-31 | 2013-12-05 | Johnson Controls Technology Company | Systems and methods for measuring and verifying energy usage in a building |
| CN102938092A (en) * | 2012-10-08 | 2013-02-20 | 珠海派诺科技股份有限公司 | Prediction method of building energy consumption in festivals and holidays based on neural network |
| CN104239982A (en) * | 2014-10-12 | 2014-12-24 | 刘岩 | Method for predicting energy consumption of buildings during holidays and festivals on basis of time series and neural networks |
Non-Patent Citations (3)
| Title |
|---|
| 《ENERGY AND BUILDINGS》 * |
| 《中国优秀硕士学位论文全文数据库工程科技II辑》 * |
| 《山东建筑大学学报》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106199174A (en) * | 2016-07-01 | 2016-12-07 | 广东技术师范学院 | Extruder energy consumption predicting abnormality method based on transfer learning |
| CN106874581A (en) * | 2016-12-30 | 2017-06-20 | 浙江大学 | A kind of energy consumption of air conditioning system in buildings Forecasting Methodology based on BP neural network model |
| CN106910144A (en) * | 2017-01-11 | 2017-06-30 | 华南理工大学 | Based on timesharing carve it is actual with can coefficient heavy construction by when energy consumption on-line prediction method |
| CN106910144B (en) * | 2017-01-11 | 2020-04-28 | 华南理工大学 | Large building time-by-time energy consumption online prediction method based on time-division actual energy consumption coefficient |
| CN106933173A (en) * | 2017-03-06 | 2017-07-07 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | A kind of equipment for dyeing and finishing monitoring system based on CORBA |
| CN107844859A (en) * | 2017-10-31 | 2018-03-27 | 深圳达实智能股份有限公司 | Large medical equipment energy consumption Forecasting Methodology and terminal device based on artificial intelligence |
| CN107844859B (en) * | 2017-10-31 | 2020-08-07 | 深圳达实智能股份有限公司 | Large medical equipment energy consumption prediction method based on artificial intelligence and terminal equipment |
| WO2019114517A1 (en) * | 2017-12-13 | 2019-06-20 | 腾讯科技（深圳）有限公司 | Neural network model deployment method, prediction method, and apparatus |
| US12020142B2 (en) | 2017-12-13 | 2024-06-25 | Tencent Technology (Shenzhen) Company Limited | Neural network model deployment method, prediction method and related device |
| CN109102103A (en) * | 2018-06-26 | 2018-12-28 | 上海鲁班软件股份有限公司 | A kind of multi-class energy consumption prediction technique based on Recognition with Recurrent Neural Network |
| CN109033595B (en) * | 2018-07-16 | 2024-03-29 | 清华大学 | New method and device for building green performance simulation analysis |
| CN109033595A (en) * | 2018-07-16 | 2018-12-18 | 清华大学 | Build Green performance sunykatuib analysis new method and device |
| CN109035067A (en) * | 2018-07-19 | 2018-12-18 | 北京泰豪智能工程有限公司 | Building energy consumption processing method and processing device based on RF and ARMA algorithm |
| CN110569981A (en) * | 2019-07-22 | 2019-12-13 | 国网冀北电力有限公司电力科学研究院 | Method and device for determining fuel consumption of scattered coal |
| CN111752870A (en) * | 2020-05-28 | 2020-10-09 | 苏州浪潮智能科技有限公司 | Compatible method and system of server interface based on neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105205558B (en) | 2018-11-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105205558A (en) | BP neural network model configuration method oriented to building energy consumption prediction | |
| CN107895283B (en) | Merchant passenger flow volume big data prediction method based on time series decomposition | |
| CN110929918B (en) | 10kV feeder fault prediction method based on CNN and LightGBM | |
| Yu et al. | Advances and challenges in building engineering and data mining applications for energy-efficient communities | |
| CN107169575A (en) | A kind of modeling and method for visualizing machine learning training pattern | |
| CN104835354B (en) | A kind of control load management system and its appraisal procedure of ATC controller workload | |
| Zhang et al. | Carbon emissions by rural energy in China | |
| CN104166731A (en) | Discovering system for social network overlapped community and method thereof | |
| CN101515409B (en) | Method for confirming airspace capacity through regression analysis on control workload | |
| CN108446794A (en) | One kind being based on multiple convolutional neural networks combination framework deep learning prediction techniques | |
| CN104881735A (en) | System and method of smart power grid big data mining for supporting smart city operation management | |
| CN107563122A (en) | The method of crime prediction of Recognition with Recurrent Neural Network is locally connected based on interleaving time sequence | |
| CN105260835A (en) | Method for modeling, analyzing, and self-optimizing multi-source business big data | |
| CN106199174A (en) | Extruder energy consumption predicting abnormality method based on transfer learning | |
| CN107122860A (en) | Bump danger classes Forecasting Methodology based on grid search and extreme learning machine | |
| CN104182803B (en) | Wind-powered electricity generation data preprocessing method and wind power forecasting method and system | |
| CN105488628A (en) | Electric power big data visualization oriented data mining method | |
| CN107967540A (en) | Student's academic warning system and method | |
| CN110532666A (en) | Urban micro-climate evolution analysis method based on wind environment variation | |
| CN109977747A (en) | A kind of coal rock detection system of comprehensive space-time characteristic | |
| CN104850677A (en) | Deep learning cellular automaton simulation analysis method based on geographic information system | |
| CN111695666A (en) | Wind power ultra-short term conditional probability prediction method based on deep learning | |
| CN110445939A (en) | The prediction technique and device of capacity resource | |
| CN104991506B (en) | Producing line initialization system and its establishing method based on PLC and touch-screen control | |
| CN109902344B (en) | Device and system for predicting structural performance of middle-small span bridge group |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |