CN111783947A - Energy consumption prediction method based on LSTM neural network - Google Patents
Energy consumption prediction method based on LSTM neural network Download PDFInfo
- Publication number
- CN111783947A CN111783947A CN202010586975.2A CN202010586975A CN111783947A CN 111783947 A CN111783947 A CN 111783947A CN 202010586975 A CN202010586975 A CN 202010586975A CN 111783947 A CN111783947 A CN 111783947A
- Authority
- CN
- China
- Prior art keywords
- energy consumption
- data
- prediction method
- neural network
- database
- 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
- 238000005265 energy consumption Methods 0.000 title claims abstract description 82
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000004519 manufacturing process Methods 0.000 claims description 15
- 238000013500 data storage Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 230000001502 supplementing effect Effects 0.000 claims description 2
- 238000010200 validation analysis Methods 0.000 claims description 2
- 238000010606 normalization Methods 0.000 abstract 1
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 8
- 230000000694 effects Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 230000005611 electricity Effects 0.000 description 4
- 239000003345 natural gas Substances 0.000 description 4
- 239000013598 vector Substances 0.000 description 4
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000009499 grossing Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000000295 fuel oil Substances 0.000 description 2
- 239000010410 layer Substances 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000015654 memory Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000010422 painting Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 239000002356 single layer Substances 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 239000008399 tap water Substances 0.000 description 1
- 235000020679 tap water Nutrition 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
- G06F16/24564—Applying rules; Deductive queries
- G06F16/24566—Recursive queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- 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
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- 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
-
- 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"
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Remote Sensing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an energy consumption prediction method based on an LSTM neural network, which comprises the following steps: the method comprises the steps of converting total energy consumption data, extracting an energy consumption data set, processing the energy consumption data set through data, processing each energy consumption data in the energy consumption data set in a normalization mode, establishing and training a learning model, and outputting a prediction result according to the trained learning model.
Description
Technical Field
The invention relates to the field of energy consumption prediction, in particular to a method for predicting energy consumption by using an LSTM neural network in the automobile production industry.
Background
The energy consumption prediction is widely applied to steel enterprises, paper enterprises, electric power systems and other aspects. But the application is less in the automobile production industry, while the production stage of automobile raw materials is the link with the largest energy consumption, the automobile assembly stage is followed, and the automobile painting stage is finally carried out. Natural gas and electricity are the most important energy consuming varieties of automobile manufacturing enterprises. In the production process of automobiles, a large amount of tap water, natural gas, high-temperature hot water, steam, electricity, compressed air and the like are required to be consumed. The automobile production enterprise belongs to a high-energy-consumption enterprise, and energy consumption prediction of the automobile production enterprise is helpful for the enterprise to plan energy reserves in advance and save energy consumption usage of related links.
The influence of multidimensional influence factors on the energy consumption is difficult to mine by the traditional energy consumption prediction method. Large-scale data real-time mining and efficient management cannot be achieved. A prediction model cannot be built in time for the production scheduling condition of an automobile enterprise, and the traditional energy consumption prediction algorithm is low in precision and is greatly influenced by human factors. Therefore, how to effectively store and efficiently manage various energy data according to the specific conditions of automobile production enterprises, and further improve the query speed and the prediction precision is a problem to be solved at present.
Disclosure of Invention
The technical problem to be solved by the invention is that the traditional energy consumption prediction method is difficult to mine multidimensional influence factors, cannot predict in time, is low in precision and large in artificial influence factor, and provides an energy consumption prediction method based on an LSTM (Long Short term memory) neural network.
The invention solves the technical problems through the following technical scheme:
an energy consumption prediction method based on an LSTM neural network, the energy consumption prediction method comprising:
converting the total energy consumption data;
extracting an energy consumption data set;
data processing the energy consumption data set;
normalizing each energy consumption data in the energy consumption data set;
establishing and training a learning model;
and outputting a prediction result according to the trained learning model.
Preferably, before converting the total energy consumption data, the method further comprises:
the acquisition system acquires various energy data;
building data storage resources, wherein the data storage resources comprise: the system comprises a distributed HBase database cluster, a weather database, an OSS file database and a factory parameter database;
importing various types of energy data, wherein the importing of various types of energy data comprises: and the various energy data are imported into the HBase database cluster, historical weather conditions are imported into the weather database, a production scheduling plan is imported into the OSS file database, and the types and the used energy parameters of the plant equipment are imported into the plant parameter database.
Preferably, the method further comprises saving the prediction result to an associated database after outputting the prediction result.
Preferably, the data processing comprises: denoising and supplementing missing data.
Further, the noise removal adopts a Savizkey-Golay algorithm, abnormal energy consumption data are removed, and energy consumption data are smoothed to remove noise points.
Further, the supplementary missing data adopts a proximity interpolation method, and the data at two ends nearest to the supplementary missing data is used for estimating the missing point data.
Preferably, the learning model is trained through a training set and a validation set and the error is within an allowable range.
Preferably, the learning models are combined by pre-training and fine-tuning updating, a network structure is initialized during the pre-training, a relatively accurate prediction model is obtained through a large number of iterations, and then the latest data is read and the fine-tuning updating is carried out through a small number of iterations.
On the basis of the common knowledge in the field, the above preferred conditions can be combined randomly to obtain the preferred embodiments of the invention.
The positive progress effects of the invention are as follows: the energy source management system can effectively store and efficiently manage various energy high-frequency energy data, can quickly inquire and process mass historical energy consumption data and weather data, and effectively solves the problems of difficulty in energy consumption data storage and low inquiry speed.
Drawings
FIG. 1 is a flow chart of a method in an embodiment of a method for energy consumption prediction based on an LSTM neural network according to the present invention;
FIG. 2 is a diagram of the prediction effect in an embodiment of the energy consumption prediction method based on the LSTM neural network according to the present invention;
fig. 3 is a diagram of an LSTM unit architecture in an embodiment of the energy consumption prediction method based on the LSTM neural network according to the present invention.
Detailed Description
To facilitate an understanding of the present application, the present application will now be described more fully with reference to the accompanying drawings. Preferred embodiments of the present application are shown in the drawings. This application may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
It will be understood that when an element is referred to as being "connected" to another element, it can be directly connected to the other element and be integral therewith, or intervening elements may also be present. The terms "mounted," "one end," "the other end," and the like are used herein for illustrative purposes only.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the description of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
Fig. 1 is a flowchart of a method in an embodiment of the present invention, in which an acquisition system acquires data information of electricity, fuel oil, and natural gas that are required to be consumed in a large amount in an automobile production industry, creates data Storage resources in consideration of influences of multidimensional factors, establishes a big data HBase database cluster, a relational database MySQL (structured query Language) such as a weather database and an OSS (Object Storage Service) file database such as an OSS file and a plant parameter database, and a connection rule therebetween.
In one example, the input feature vector of the energy consumption prediction method consists of historical reduced energy consumption data, date type, timestamp, maximum air temperature, minimum air temperature, production scheduling plan, plant parameters, all 5-dimensional data, i.e., data
input(t)=[Load(t),DayType(t),t,Tmax,Tmin]
Wherein, Load is the energy consumption value at the point; DayType is a date type, 1 is taken on weekdays, 0.5 is taken on saturday, and 0 is taken on workdays; t is a time value and has a value range of 1 to 96; tmax and Tmin are the maximum temperature and the minimum temperature of the day, respectively.
The output vector is the energy consumption value at the prediction time, namely output (t) ═ Load (t +1), after the input and output characteristic vectors are established, a single-layer LSTM model is established, and the number of network hidden layer neurons is determined to be 12 through testing.
In order to improve the prediction effect and the processing efficiency of the network, a reasonable order is required to be selected. The model order is determined by calculating the autocorrelation coefficient of the energy consumption curve time sequence, and each order of autocorrelation coefficient can reflect the correlation between each time lag state, namely the periodic rule and the time sequence of the sequence can be reflectedxiOf order k autocorrelation coefficient ckThe calculation formula of (2) is as follows:
in the formula (I), the compound is shown in the specification,is a time sequence xiTaking the prediction of the day-ahead energy consumption as an example, the autocorrelation coefficient of the historical energy consumption data set is calculated, and it can be known that the energy consumption curve has significant periodicity, the period length of the energy consumption curve coincides with 96 points in a day, and the correlation coefficient with the predicted day-ahead is the highest. The peak point of the autocorrelation coefficient is selected as the order of the rolling prediction, so 96 is taken as the order of the day-ahead energy consumption prediction.
In one example, various energy data are imported into corresponding databases, the energy data information of electricity, fuel oil and natural gas collected in the collection system is converted into Active MQ intermediate files and then imported into various regions in an H Base database cluster, local historical weather conditions such as temperature, humidity and cloud conditions are imported into a weather database by using a program interface, production scheduling plan data are manually placed into an OSS file database, equipment types and used energy parameters of a factory are imported into a factory parameter database, and total energy consumption data are obtained by converting the historical energy consumption data stored in the HBase database cluster and various energy data input in real time according to energy consumption conversion coefficients.
In one example, the total energy consumption data is extracted to obtain an energy consumption data set, and data in the energy consumption data set is subjected to data processing.
In an alternative example, the Savitazky-Golay algorithm is used to remove abnormal energy consumption data, the data in the energy consumption data set is smoothed to remove noise points, and the savazkey-Golay algorithm is mostly applied to noise smoothing, so as to smooth and calculate the result dispersion with a simplified minimum quadratic convolution. The algorithm is a polynomial of average weight. The polynomial is used to preserve higher values of data to reduce the bias caused by the algorithm, wherein,
in the formula, Y is a series of initial values, Y is a calculation result, CiThe weighting factor, which is the ith number in the smoothing window, can be considered as the measured distance between the original time series value and the polynomial function result, and N is the width of the smoothing window (2m + 1). j is the initial value index. The algorithm window covers 2m +1 neighbor point values, where m is the width of half the filter window.
In an alternative example, the missing data is supplemented by using a proximity interpolation method, the data of two ends nearest to the missing data is used for estimating the missing data, for example, the data of 13 points of energy consumption is missing, and the energy consumption value of 13 points can be estimated by the energy consumption value of 12 points and the energy consumption value of 14 points.
In one example, the energy data in the data-processed energy consumption data set is normalized, wherein the data in the weather database and the OSS file database, such as temperature factors and time values, are normalized to be limited within the range of [0,1],
in the formula, x1(i) For the raw data x (i) normalized data values, xmax、xminThe maximum and minimum values of x (i), respectively.
In one example, the energy consumption value is normalized and then normalized to reduce the adverse effects of large deviations.
In the formula, x2(i) For the normalized data values, mean values, of the element data x (i)Standard deviation of
In one example, a learning model is built and trained on the normalized data, and the learning model effect evaluation index adopts an average absolute percentage error and a root mean square error:
wherein n is the number of predicted points, yi、diAnd respectively obtaining the real value and the predicted value of the energy consumption of the predicted point i.
After each layer of basic structure of the LSTM energy consumption prediction model is built and appropriate parameters are set, reading input data, constructing a training set and writing a program, wherein the updating criterion of the LSTM is as follows:
in the formula, Wi、Wf、WoRespectively updating, forgetting and outputting a weighting matrix; bi、bf、boThe deviations are parameterized for the conversion of the update gate, the forgetting gate and the output gate respectively; sigma is sigmoid function; multiplying corresponding elements of a matrix, adopting a built-in Adam optimizer under a tensoflow packet as an iterative optimization mode, and adopting an Adam method as a self-adaptive learning rate algorithm, compared with other self-adaptive learning rate algorithms, the Adam method has higher convergence speed and more effective learning effect, and can correct the problems existing in other optimization technologies, such as the problems of disappearance of the learning rate, too low convergence, larger fluctuation of a loss function caused by parameter updating of high variance and the like, as shown in FIG. 3, the LSTM unit system structure diagram of the embodiment of the invention has 1 memory unit state and 3 gates, namely, the gates are updatediForgetting doorfAnd output gateoBy using x{1},x{2}… denotes a typical input sequence in an LSTM network, then x is{t}Representing the input characteristics at time t. To achieve long-term storage of important information, a memory unit is set up and maintained throughout the entire period of the LSTM. Activation unit a according to previous time(t-1)And the input x of the current time{t}The specific element of the internal state vector is updated, maintained or forgotten as determined by the 3 gates.
In one example, in order to adapt to actual engineering requirements and accelerate the prediction speed, the energy consumption prediction model adopts a training mode combining pre-training and fine-tuning updating, different iteration times are set for the two training modes, wherein a network structure is initialized during first training, a relatively accurate prediction model is obtained through larger iteration times, then a small amount of new data is introduced into each prediction along with the passing of date, on the basis of loading a historical model, a small batch of iteration is performed by using newly read data to fine-tune model parameters, and the updated model parameters are stored. Finally, calling the model to predict the energy consumption of the required date.
As shown in fig. 2, which is a technical effect diagram of an embodiment of the present invention, the trained and verified learning model is used for energy consumption prediction in the automobile production industry, and the trend of the actual situation can be well simulated for energy consumption prediction, so that various energy high-frequency energy data can be effectively stored and efficiently managed.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that this is by way of example only, and that the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the spirit and scope of the invention, and these changes and modifications are within the scope of the invention.
Claims (8)
1. An energy consumption prediction method based on an LSTM neural network is characterized in that the energy consumption prediction method comprises the following steps:
converting the total energy consumption data;
extracting an energy consumption data set;
data processing the energy consumption data set;
normalizing each energy consumption data in the energy consumption data set;
establishing and training a learning model;
and outputting a prediction result according to the trained learning model.
2. The LSTM neural network-based energy consumption prediction method of claim 1, further comprising, before converting the total energy consumption data:
the acquisition system acquires various energy data;
building data storage resources, wherein the data storage resources comprise: HBase database cluster, weather database, OSS file database and factory parameter database;
importing various types of energy data, wherein the importing of various types of energy data comprises: and the various energy data are imported into the HBase database cluster, historical weather conditions are imported into the weather database, a production scheduling plan is imported into the OSS file database, and the types and the used energy parameters of the plant equipment are imported into the plant parameter database.
3. The LSTM neural network-based energy consumption prediction method of claim 1, further comprising saving the prediction results to an associated database after outputting the prediction results.
4. The LSTM neural network-based energy consumption prediction method of claim 1, wherein the data processing comprises: denoising and supplementing missing data.
5. The energy consumption prediction method based on the LSTM neural network as claimed in claim 4, wherein said denoising employs a Savizkey-Golay algorithm, and removes abnormal energy consumption data to smooth the energy consumption data and remove noise points.
6. The energy consumption prediction method based on the LSTM neural network as claimed in claim 4, wherein said supplementary missing data is obtained by using adjacent interpolation method and using the data of two ends nearest to it to estimate the missing point data.
7. The LSTM neural network-based energy consumption prediction method of claim 1, wherein the learning model is trained with a training set and a validation set and the error is within an allowable range.
8. The energy consumption prediction method according to any one of claims 1 to 7, wherein the learning models are each combined with pre-training and fine-tuning updating, the pre-training initializes the network structure, obtains a relatively accurate prediction model through a larger number of iterations, and then reads the latest data to perform the fine-tuning updating through a smaller number of iterations.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010586975.2A CN111783947A (en) | 2020-06-24 | 2020-06-24 | Energy consumption prediction method based on LSTM neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010586975.2A CN111783947A (en) | 2020-06-24 | 2020-06-24 | Energy consumption prediction method based on LSTM neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111783947A true CN111783947A (en) | 2020-10-16 |
Family
ID=72759717
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010586975.2A Pending CN111783947A (en) | 2020-06-24 | 2020-06-24 | Energy consumption prediction method based on LSTM neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111783947A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112668713A (en) * | 2020-12-21 | 2021-04-16 | 通号智慧城市研究设计院有限公司 | Intelligent energy consumption prediction method and system |
CN112966446A (en) * | 2021-03-24 | 2021-06-15 | 青岛洪锦智慧能源技术有限公司 | Method for predicting cooling water circulation energy consumption in central air-conditioning refrigeration system |
CN113917263A (en) * | 2021-10-09 | 2022-01-11 | 乐创达投资(广东)有限公司 | Abnormal energy consumption secondary monitoring method and system based on prediction data |
WO2023028842A1 (en) * | 2021-08-31 | 2023-03-09 | 西门子股份公司 | Factory operation prediction method and apparatus, and computer-readable storage medium |
CN116245206A (en) * | 2022-11-24 | 2023-06-09 | 华建数创(上海)科技有限公司 | Building energy consumption prediction method based on multi-mode integration model |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019051701A1 (en) * | 2017-09-14 | 2019-03-21 | 深圳传音通讯有限公司 | Photographic terminal, and photographic parameter setting method therefor based on long short-term memory neural network |
CN109685252A (en) * | 2018-11-30 | 2019-04-26 | 西安工程大学 | Building energy consumption prediction technique based on Recognition with Recurrent Neural Network and multi-task learning model |
CN110852515A (en) * | 2019-11-15 | 2020-02-28 | 北京工业大学 | Water quality index prediction method based on mixed long-time and short-time memory neural network |
US20200104200A1 (en) * | 2018-09-27 | 2020-04-02 | Oracle International Corporation | Disk drive failure prediction with neural networks |
CN111178612A (en) * | 2019-12-19 | 2020-05-19 | 绍兴大明电力设计院有限公司 | LSTM load prediction method of grid user based on big data ODPS engine |
-
2020
- 2020-06-24 CN CN202010586975.2A patent/CN111783947A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019051701A1 (en) * | 2017-09-14 | 2019-03-21 | 深圳传音通讯有限公司 | Photographic terminal, and photographic parameter setting method therefor based on long short-term memory neural network |
US20200104200A1 (en) * | 2018-09-27 | 2020-04-02 | Oracle International Corporation | Disk drive failure prediction with neural networks |
CN109685252A (en) * | 2018-11-30 | 2019-04-26 | 西安工程大学 | Building energy consumption prediction technique based on Recognition with Recurrent Neural Network and multi-task learning model |
CN110852515A (en) * | 2019-11-15 | 2020-02-28 | 北京工业大学 | Water quality index prediction method based on mixed long-time and short-time memory neural network |
CN111178612A (en) * | 2019-12-19 | 2020-05-19 | 绍兴大明电力设计院有限公司 | LSTM load prediction method of grid user based on big data ODPS engine |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112668713A (en) * | 2020-12-21 | 2021-04-16 | 通号智慧城市研究设计院有限公司 | Intelligent energy consumption prediction method and system |
CN112966446A (en) * | 2021-03-24 | 2021-06-15 | 青岛洪锦智慧能源技术有限公司 | Method for predicting cooling water circulation energy consumption in central air-conditioning refrigeration system |
CN112966446B (en) * | 2021-03-24 | 2022-09-02 | 青岛洪锦智慧能源技术有限公司 | Method for predicting cooling water circulation energy consumption in central air-conditioning refrigeration system |
WO2023028842A1 (en) * | 2021-08-31 | 2023-03-09 | 西门子股份公司 | Factory operation prediction method and apparatus, and computer-readable storage medium |
CN113917263A (en) * | 2021-10-09 | 2022-01-11 | 乐创达投资(广东)有限公司 | Abnormal energy consumption secondary monitoring method and system based on prediction data |
CN113917263B (en) * | 2021-10-09 | 2022-05-24 | 乐创达投资(广东)有限公司 | Method and system for secondary monitoring of abnormal energy consumption based on prediction data |
CN116245206A (en) * | 2022-11-24 | 2023-06-09 | 华建数创(上海)科技有限公司 | Building energy consumption prediction method based on multi-mode integration model |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111783947A (en) | Energy consumption prediction method based on LSTM neural network | |
CN111967688B (en) | Power load prediction method based on Kalman filter and convolutional neural network | |
CN111027772B (en) | Multi-factor short-term load prediction method based on PCA-DBILSTM | |
CN111353652B (en) | Wind power output short-term interval prediction method | |
CN110163429B (en) | Short-term load prediction method based on similarity day optimization screening | |
CN111079989B (en) | DWT-PCA-LSTM-based water supply amount prediction device for water supply company | |
CN110380444B (en) | Capacity planning method for distributed wind power orderly access to power grid under multiple scenes based on variable structure Copula | |
CN108053082B (en) | Power grid medium and long term load prediction method based on temperature interval decomposition | |
CN103729501A (en) | Short-term power load predicting method based on grey theory | |
CN113107626B (en) | Load prediction method of combined cycle generator set based on multivariable LSTM | |
CN104598985B (en) | Methods of electric load forecasting | |
CN101383023A (en) | Neural network short-term electric load prediction based on sample dynamic organization and temperature compensation | |
CN111027775A (en) | Step hydropower station generating capacity prediction method based on long-term and short-term memory network | |
CN114428803B (en) | Air compression station operation optimization method, system, storage medium and terminal | |
CN105117593A (en) | Wavelet transform and particle swarm optimized grey model-based short-term wind speed forecasting method | |
CN104598986A (en) | Big data based power load prediction method | |
CN116307652A (en) | Artificial intelligent resource allocation method for intelligent power grid | |
CN115860177A (en) | Photovoltaic power generation power prediction method based on combined machine learning model and application thereof | |
CN114218872A (en) | Method for predicting remaining service life based on DBN-LSTM semi-supervised joint model | |
CN112819238A (en) | Short-term wind power prediction method based on chaotic chicken flock optimization algorithm | |
CN112651543A (en) | Daily electric quantity prediction method based on VMD decomposition and LSTM network | |
CN115511657A (en) | Wind power output and photovoltaic output evaluation method based on combined prediction model | |
CN111178612B (en) | LSTM load prediction method of grid user based on big data ODPS engine | |
CN117114160A (en) | Short-term photovoltaic power prediction method | |
CN115759465A (en) | Wind power prediction method based on multi-target collaborative training and NWP implicit correction |
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: 20201016 |
|
RJ01 | Rejection of invention patent application after publication |