CN117094651A - Integrated energy storage management system and energy management method - Google Patents
Integrated energy storage management system and energy management method Download PDFInfo
- Publication number
- CN117094651A CN117094651A CN202310468867.9A CN202310468867A CN117094651A CN 117094651 A CN117094651 A CN 117094651A CN 202310468867 A CN202310468867 A CN 202310468867A CN 117094651 A CN117094651 A CN 117094651A
- Authority
- CN
- China
- Prior art keywords
- energy consumption
- time sequence
- vector
- feature vector
- energy
- 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
- 238000007726 management method Methods 0.000 title claims abstract description 79
- 238000004146 energy storage Methods 0.000 title claims abstract description 70
- 239000013598 vector Substances 0.000 claims description 415
- 238000005265 energy consumption Methods 0.000 claims description 332
- 230000008859 change Effects 0.000 claims description 143
- 238000000605 extraction Methods 0.000 claims description 73
- 238000009826 distribution Methods 0.000 claims description 37
- 230000004927 fusion Effects 0.000 claims description 26
- 239000011159 matrix material Substances 0.000 claims description 25
- 238000012544 monitoring process Methods 0.000 claims description 19
- 238000005430 electron energy loss spectroscopy Methods 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 11
- 238000012512 characterization method Methods 0.000 claims description 10
- 230000006870 function Effects 0.000 description 45
- 238000010586 diagram Methods 0.000 description 14
- 238000004590 computer program Methods 0.000 description 9
- 230000002596 correlated effect Effects 0.000 description 9
- 230000014509 gene expression Effects 0.000 description 8
- 238000013473 artificial intelligence Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 238000000034 method Methods 0.000 description 6
- 230000002159 abnormal effect Effects 0.000 description 5
- 239000002699 waste material Substances 0.000 description 5
- 230000001419 dependent effect Effects 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 2
- 230000000875 corresponding effect Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000036962 time dependent Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- 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
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the field of intelligent management, and particularly discloses an integrated energy storage management system and an energy management method.
Description
Technical Field
The application relates to the field of intelligent management, and more particularly, to an integrated energy storage management system and an energy management method.
Background
As the international energy market is affected by the factor of unreliability, the energy price is high-order to oscillate; the power supply is tension due to abnormal climate change in multiple superimposed countries, so that power alarms are triggered, and a power limiting mode is opened in a dispute mode. If a user wants to realize peak and valley shifting of electric energy by using a mode of low charging and high discharging of the peak Gu Jiacha energy storage system, the traditional energy storage management cannot meet the requirements of the user because the traditional energy storage management is not connected with the Internet.
The IESS integrated energy storage management unit can integrally manage data of some energy storage devices including PCS, BMS, ammeter, battery cell and the like, and can realize cloud management of the data. However, the IESS is more often used to monitor the energy usage of a building or facility, and cannot analyze and determine whether the energy usage of the building as a whole or the facility is normal. Thus, an optimized integrated energy storage management scheme is desired.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides an integrated energy storage management system and an energy management method, which can more accurately judge whether the whole building or the energy use condition of the building is normal or not by adopting an artificial intelligence technology based on deep learning to extract hidden association information between the energy use condition and the energy consumption change condition of the facility.
According to one aspect of the present application, there is provided an integrated energy storage management system comprising: an energy consumption monitoring unit for acquiring energy consumption values of the monitored facility acquired by the IESS unit for a plurality of days within a predetermined time span; the data structuring unit is used for arranging the energy consumption values of the monitored facility for a plurality of days in a preset time span into energy-consuming time sequence input vectors according to a time dimension; the change information characterization unit is used for calculating the difference value between the energy consumption values of two positions of each vector in the energy consumption time sequence input vector to obtain an energy consumption change time sequence input vector; the time sequence feature extraction unit is used for enabling the energy-consumption time sequence input vector to pass through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer to obtain an energy-consumption related time sequence feature vector; the change feature extraction unit is used for enabling the energy consumption change time sequence input vector to pass through a multi-scale neighborhood feature extraction module comprising a first convolution layer and a second convolution layer to obtain an energy consumption change associated time sequence feature vector; the feature fusion unit is used for fusing the energy consumption related time sequence feature vector and the energy consumption change related time sequence feature vector to obtain a classification feature vector; and the monitoring result generating unit is used for passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the energy consumption of the monitored facility is normal or not.
In the above integrated energy storage management system, the time sequence feature extraction unit includes: and the full-connection coding subunit is used for performing full-connection coding on the energy-consumption time sequence input vector by using a full-connection layer of the time sequence coder according to the following formula to extract high-dimensional implicit characteristics of characteristic values of all positions in the energy-consumption time sequence input vector, wherein the formula is as follows:wherein->Is the energy-consuming time-ordered input vector, +.>Is the output vector, +.>Is a weight matrix, < >>Is a bias vector, ++>Representing matrix multiplication; a one-dimensional convolution coding subunit, configured to perform one-dimensional convolution coding on the energy-saving time sequence input vector by using a one-dimensional convolution layer of the time sequence encoder according to the following formula to extract high-dimensional implicit correlation features between feature values of each position in the energy-saving time sequence input vector, where the formula is:wherein (1)>For convolution kernel +.>Width in direction, ++>Is a convolution kernel parameter vector,For a local vector matrix operating with a convolution kernel function, < ->For the size of the convolution kernel +.>Representing the energy-consuming time sequence input vector, +.>Representing one-dimensional convolutional encoding of the energy-time ordered input vector.
In the above integrated energy storage management system, the multi-scale neighborhood feature extraction module includes: the device comprises a first convolution layer, a second convolution layer parallel to the first convolution layer and a multi-scale feature fusion layer connected with the first convolution layer and the second convolution layer, wherein the first convolution layer uses a one-dimensional convolution kernel with a first length, and the second convolution layer uses a one-dimensional convolution kernel with a second length.
In the above integrated energy storage management system, the change feature extraction unit includes: a first neighborhood scale feature extraction subunit, configured to input the energy consumption change timing sequence input vector to a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first neighborhood scale energy consumption change associated timing sequence feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; second neighborhood scale featureThe sign extraction subunit is used for inputting the energy consumption change time sequence input vector into a second convolution layer of the multi-scale neighborhood feature extraction module to obtain a second neighborhood scale energy consumption change related time sequence feature vector, wherein the second convolution layer is provided with a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and the multi-scale cascading subunit is used for cascading the first neighborhood scale energy consumption change related time sequence feature vector and the second neighborhood scale energy consumption change related time sequence feature vector to obtain the energy consumption change related time sequence feature vector. Wherein the first neighborhood scale feature extraction subunit is configured to: performing one-dimensional convolution coding on the energy consumption change time sequence input vector by using a first convolution layer of the multi-scale neighborhood feature extraction module according to the following one-dimensional convolution formula to obtain a first neighborhood scale energy consumption change associated time sequence feature vector; wherein, the formula is: Wherein->For the first convolution kernel at->Width in direction, ++>For the first convolution kernel parameter vector, +.>For a local vector matrix operating with a convolution kernel function, < ->For the size of the first convolution kernel, +.>Representing the energy consumption variation time sequence input vector, < >>Representing the pair ofPerforming one-dimensional convolution coding on the energy consumption change time sequence input vector; and, the second neighborhood scale feature extraction subunit is configured to: performing one-dimensional convolution coding on the energy consumption change time sequence input vector by using a second convolution layer of the multi-scale neighborhood feature extraction module according to the following one-dimensional convolution formula to obtain a second neighborhood scale energy consumption change associated time sequence feature vector; wherein, the formula is:wherein->For the second convolution kernel>Width in direction, ++>For a second convolution kernel parameter vector, +.>For a local vector matrix operating with a convolution kernel function, < ->For the size of the second convolution kernel, +.>Representing the energy consumption variation time sequence input vector, < >>Representing one-dimensional convolutional encoding of the energy consumption variation time sequence input vector.
In the above integrated energy storage management system, the feature fusion unit includes: an information compensation loss subunit for calculating a probability distribution shift information compensation loss function value for the energy consumption-related timing feature vector and the energy consumption variation-related timing feature vector; the weighting subunit is used for weighting the energy consumption related time sequence feature vector and the energy consumption change related time sequence feature vector based on the probability distribution shift information compensation loss function value so as to obtain a weighted energy consumption related time sequence feature vector and a weighted energy consumption change related time sequence feature vector; and a fusion subunit, configured to fuse the weighted energy consumption related time sequence feature vector and the weighted energy consumption variation related time sequence feature vector to obtain the classification feature vector.
In the above integrated energy storage management system, the information compensation loss subunit is configured to: calculating probability distribution shift information compensation loss function values for the energy consumption-related timing feature vector and the energy consumption variation-related timing feature vector in the following formula; wherein, the formula is:wherein->And->The energy consumption related timing characteristic vector and the energy consumption variation related timing characteristic vector, +.>Andcompensating for shift superparameter, and +.>Weighted superparameter->Represents a logarithmic function value based on 2, < +.>Representation ofFunction (F)>Representation->Function (F)>The probability distribution shift information is shown to compensate for the loss function value.
In the above integrated energy storage management system, the fusion subunit is configured to: fusing the weighted energy consumption related time sequence feature vector and the weighted energy consumption change related time sequence feature vector by using the following cascade formula to obtain the classification feature vector; wherein, the formula is:wherein->Representing the weighted energy consumption related time sequence feature vector, < >>Representing the weighted energy consumption change associated time sequence feature vector +.>Representing a function of the cascade of functions,representing the classification feature vector.
In the above integrated energy storage management system, the monitoring result generating unit is configured to: processing the classification feature vector using the classifier to obtain a classification result with the following formula:wherein->To->Is a weight matrix>To->For the bias vector +.>Is a classification feature vector.
According to another aspect of the present application, there is provided an integrated energy storage management method, including: acquiring energy consumption values of the monitored facility acquired by the IESS unit for a plurality of days within a preset time span; arranging the energy consumption values of the monitored facilities for a plurality of days in a preset time span into energy-saving time sequence input vectors according to a time dimension; calculating the difference between the energy consumption values of two positions of each vector in the energy consumption time sequence input vector to obtain an energy consumption change time sequence input vector; the energy consumption time sequence input vector passes through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer to obtain an energy consumption related time sequence feature vector; the energy consumption change time sequence input vector passes through a multi-scale neighborhood feature extraction module comprising a first convolution layer and a second convolution layer to obtain an energy consumption change associated time sequence feature vector; fusing the energy consumption related time sequence feature vector and the energy consumption change related time sequence feature vector to obtain a classification feature vector; and passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the energy consumption of the monitored facility is normal or not.
According to still another aspect of the present application, there is provided an electronic apparatus including: a processor; and a memory having stored therein computer program instructions that, when executed by the processor, cause the processor to perform the integrated energy storage management method as described above.
According to yet another aspect of the present application, there is provided a computer readable medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform an integrated energy storage management method as described above.
Compared with the prior art, the integrated energy storage management system and the energy management method provided by the application have the advantages that the hidden association information between the energy use condition and the energy consumption change condition of the facility is dug by adopting the artificial intelligence technology based on deep learning, so that whether the energy use condition of the whole building or the facility is normal or not can be judged more accurately, and the problems of energy waste and the like in time can be found by further prompting the energy consumption investigation of the facility or the building.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
Fig. 1 is a block diagram of an integrated energy storage management system according to an embodiment of the present application.
Fig. 2 is a system architecture diagram of an integrated energy storage management system according to an embodiment of the present application.
Fig. 3 is a block diagram of a timing feature extraction unit in the integrated energy storage management system according to an embodiment of the present application.
Fig. 4 is a block diagram of a change feature extraction unit in an integrated energy storage management system according to an embodiment of the present application.
Fig. 5 is a block diagram of a feature fusion unit in an integrated energy storage management system according to an embodiment of the present application.
Fig. 6 is a flowchart of an integrated energy storage management method according to an embodiment of the present application.
Fig. 7 is a block diagram of an electronic device according to an embodiment of the application.
Detailed Description
Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Summary of the application: aiming at the technical problems, the technical conception of the application is as follows: the energy use condition of the building or the facility is monitored through the IESS, and whether the energy use condition of the whole building or the facility is normal or not is judged by combining deep learning and artificial intelligence technology.
Specifically, in the technical solution of the present application, first, the energy consumption values of the monitored facility collected by the IESS unit are acquired for a plurality of days within a predetermined time span. By obtaining the energy consumption values of the monitored facility over a predetermined time span for a plurality of days, time series data of the energy usage of the facility can be obtained.
For further processing and analysis, in the technical scheme of the application, the energy consumption values of the monitored facilities for a plurality of days in a preset time span are arranged into energy-consuming time sequence input vectors according to a time dimension. Wherein the energy-time sequence input vector integrates the energy consumption values of the monitored facility collected by the IESS unit for a plurality of days within a predetermined time span, that is, each characteristic value corresponds to the energy consumption value of the monitored facility on a certain day.
Then, the difference between the energy consumption values of two positions of each vector in the energy consumption time sequence input vector is calculated to obtain an energy consumption change time sequence input vector. Here, by calculating the difference between two positions of each vector, the energy consumption change situation between adjacent time points can be obtained, so that the deviation existing due to the influence of factors such as monitoring equipment and environment on the energy consumption data is eliminated, and the energy consumption change situation is reflected more accurately.
And then, the energy consumption time sequence input vector passes through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer to obtain an energy consumption related time sequence characteristic vector. Here, the timing encoder may convert the time-series data into a timing feature vector having a higher level of abstract feature. That is, the energy-consuming time sequence input vector is passed through a sequence encoder comprising a one-dimensional convolution layer and a fully-connected layer, and the relevant features of the utility energy use condition in the time dimension can be extracted. Specifically, the one-dimensional convolution layer can effectively capture local time characteristics of energy use conditions, and the full-connection layer can integrate the local time characteristics to obtain global time characteristics. By the method, time-related characteristics of the energy use condition of the facility can be effectively extracted, and subsequent classification processing is facilitated.
And then, the energy consumption change time sequence input vector passes through a multi-scale neighborhood feature extraction module comprising a first convolution layer and a second convolution layer to obtain an energy consumption change associated time sequence feature vector. It should be understood that the time sequence feature vector associated with the energy consumption change reflects the change condition of the energy use condition of the facility, and features with different scales can be extracted from the time sequence data by adopting the multi-scale neighborhood feature extraction module, so that the accuracy of analyzing and judging the energy use condition of the facility is further improved.
In the technical scheme of the application, in order to comprehensively consider the energy use condition and the energy consumption change condition of the facility, the accuracy of analyzing and judging the energy use condition of the facility is improved, and the energy consumption related time sequence feature vector and the energy consumption change related time sequence feature vector are fused to obtain the classification feature vector. That is, the energy consumption-related time series feature vector reflecting the time-dependent feature of the utility energy use condition and the energy consumption change-related time series feature vector reflecting the change condition of the utility energy use condition are fused together, so that the classification feature vector has more excellent information characterization capability.
After the classification feature vector is obtained, the classification feature vector is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the energy consumption of the monitored facility is normal or not. The classifier can classify the energy use condition of the facilities according to the classification feature vector, namely, the energy consumption of the monitored facilities is normal and the energy consumption of the monitored facilities is abnormal. Therefore, based on the classification result, whether the energy use condition of the facility accords with the expectations can be timely judged, and if the energy waste or abnormal use phenomenon exists, corresponding measures can be taken for adjustment and optimization. Meanwhile, the classification result can also be used as feedback information to help a user optimize an energy management strategy of the facility, so that the energy utilization efficiency is further improved, and the energy consumption cost is reduced.
Here, when the energy consumption related time sequence feature vector and the energy consumption variation related time sequence feature vector are fused to obtain the classification feature vector, since the energy consumption related time sequence feature vector and the energy consumption variation related time sequence feature vector respectively express time sequence related features of an energy consumption absolute value and an energy consumption variation value, probability distribution of the time sequence related feature vector and the energy consumption variation related time sequence feature vector relative to class labels of a classifier can be different, and therefore when the fused classification feature vector passes through the classifier, characteristic distribution of the energy consumption related time sequence feature vector and the energy consumption variation related time sequence feature vector backwards propagates through the classifier in a parameter space of a model, degradation problems of respective characteristic probability distribution expressions caused by probability distribution shifting can be encountered, and characteristic expression effects of the classification feature vector are affected.
Based on this, the applicant of the present application introduced a time-series feature vector associated with the energy consumptionTime sequence characteristic vector correlated with the energy consumption change>The probability distribution shift information compensation loss function of (2) is expressed as:wherein->And->Compensating for shift superparameter, and +.>Is a weighted superparameter.
Here, a timing feature vector is correlated from the energy consumption based on a Softmax function Time sequence characteristic vector correlated with the energy consumption change>The respective derived class probability values themselves follow probability distributions for the respective feature distributions, using the probability distribution shift information to compensate for a loss function for the energy consumption-dependent time-series feature vector +.>Time sequence characteristic vector correlated with the energy consumption change>Information compensation is performed by shifting the probability distribution of the feature representation of (2), and cross information entropy caused by compensation is maximized by a bool function, so that the feature distribution of the classification feature vector after fusion can restore the energy consumption related time sequence feature vector +_x before fusion to the maximum extent>Time sequence characteristic vector correlated with the energy consumption change>The feature probability distribution expression information of the classification feature vector is improved, so that the feature expression effect of the classification feature vector is improved, and the accuracy of a classification result obtained by the classifier is improved.
Based on this, the present application proposes an integrated energy storage management system comprising: an energy consumption monitoring unit for acquiring energy consumption values of the monitored facility acquired by the IESS unit for a plurality of days within a predetermined time span; the data structuring unit is used for arranging the energy consumption values of the monitored facility for a plurality of days in a preset time span into energy-consuming time sequence input vectors according to a time dimension; the change information characterization unit is used for calculating the difference value between the energy consumption values of two positions of each vector in the energy consumption time sequence input vector to obtain an energy consumption change time sequence input vector; the time sequence feature extraction unit is used for enabling the energy-consumption time sequence input vector to pass through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer to obtain an energy-consumption related time sequence feature vector; the change feature extraction unit is used for enabling the energy consumption change time sequence input vector to pass through a multi-scale neighborhood feature extraction module comprising a first convolution layer and a second convolution layer to obtain an energy consumption change associated time sequence feature vector; the feature fusion unit is used for fusing the energy consumption related time sequence feature vector and the energy consumption change related time sequence feature vector to obtain a classification feature vector; and the monitoring result generating unit is used for passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the energy consumption of the monitored facility is normal or not.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Exemplary System: fig. 1 is a block diagram of an integrated energy storage management system according to an embodiment of the present application. As shown in fig. 1, an integrated energy storage management system 300 according to an embodiment of the present application includes: an energy consumption monitoring unit 310; a data structuring unit 320; a change information characterization unit 330; a timing feature extraction unit 340; a change feature extraction unit 350; a feature fusion unit 360; and a monitoring result generation unit 370.
Wherein the energy consumption monitoring unit 310 is configured to obtain energy consumption values of the monitored facility acquired by the IESS unit for a plurality of days within a predetermined time span; the data structuring unit 320 is configured to arrange energy consumption values of the monitored facility for a plurality of days within a predetermined time span according to a time dimension into energy-consuming time sequence input vectors; the change information characterization unit 330 is configured to calculate a difference between energy consumption values of two positions of each vector in the energy consumption time sequence input vector to obtain an energy consumption change time sequence input vector; the timing characteristic extraction unit 340 is configured to pass the energy-consuming timing input vector through a timing encoder including a one-dimensional convolution layer and a full connection layer to obtain an energy-consuming associated timing characteristic vector; the change feature extraction unit 350 is configured to pass the energy consumption change timing input vector through a multi-scale neighborhood feature extraction module including a first convolution layer and a second convolution layer to obtain an energy consumption change associated timing feature vector; the feature fusion unit 360 is configured to fuse the energy consumption related time sequence feature vector and the energy consumption variation related time sequence feature vector to obtain a classification feature vector; and the monitoring result generating unit 370 is configured to pass the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether the energy consumption of the monitored facility is normal.
Fig. 2 is a system architecture diagram of an integrated energy storage management system according to an embodiment of the present application. As shown in fig. 2, in the network architecture, first, the energy consumption value of the monitored facility acquired by the IESS unit for a plurality of days within a predetermined time span is acquired by the energy consumption monitoring unit 310; next, the data structuring unit 320 arranges the energy consumption values of the monitored facility acquired by the energy consumption monitoring unit 310 for a plurality of days within a predetermined time span into energy-consuming time sequence input vectors according to a time dimension; the change information characterization unit 330 calculates a difference between the energy consumption values of two positions of each vector in the energy consumption time sequence input vector obtained by the data structuring unit 320 to obtain an energy consumption change time sequence input vector; then, the timing feature extraction unit 340 passes the energy-consuming timing input vector obtained by the data structuring unit 320 through a timing encoder including a one-dimensional convolution layer and a full connection layer to obtain an energy-consuming associated timing feature vector; the change feature extraction unit 350 passes the energy consumption change time sequence input vector calculated by the change information characterization unit 330 through a multi-scale neighborhood feature extraction module comprising a first convolution layer and a second convolution layer to obtain an energy consumption change associated time sequence feature vector; the feature fusion unit 360 fuses the energy consumption related time sequence feature vector obtained by the time sequence feature extraction unit 340 and the energy consumption variation related time sequence feature vector obtained by the variation feature extraction unit 350 to obtain a classification feature vector; further, the monitoring result generating unit 370 passes the classification feature vector obtained by the feature fusion unit 360 through a classifier to obtain a classification result indicating whether the energy consumption of the monitored facility is normal.
Specifically, during operation of the integrated energy storage management system 300, the energy consumption monitoring unit 310 is configured to obtain energy consumption values of the monitored facility collected by the IESS unit for a plurality of days within a predetermined time span. It should be understood that the IESS is generally used to monitor the energy usage of a building or facility, so in the technical solution of the present application, the energy usage of the building or facility is monitored by the IESS, and the deep learning and artificial intelligence technologies are combined to determine whether the energy usage of the whole building or facility is normal. First, energy consumption values of the monitored facility collected by the IESS unit are obtained for a plurality of days within a predetermined time span. By obtaining the energy consumption values of the monitored facility over a predetermined time span for a plurality of days, time series data of the energy usage of the facility can be obtained.
Specifically, during the operation of the integrated energy storage management system 300, the data structuring unit 320 is configured to arrange the energy consumption values of the monitored facility for a plurality of days within a predetermined time span into energy-consuming time sequence input vectors according to a time dimension. That is, in the technical solution of the present application, the energy consumption values of the monitored facility for a plurality of days within a predetermined time span are arranged as energy-time-series input vectors according to a time dimension. Wherein the energy-time sequence input vector integrates the energy consumption values of the monitored facility collected by the IESS unit for a plurality of days within a predetermined time span, that is, each characteristic value corresponds to the energy consumption value of the monitored facility on a certain day.
Specifically, during the operation of the integrated energy storage management system 300, the change information characterization unit 330 is configured to calculate a difference between the energy consumption values of two positions of each vector in the energy consumption time sequence input vector to obtain the energy consumption change time sequence input vector. That is, the difference between the energy consumption values of two positions per vector in the energy consumption time sequence input vector is calculated to obtain an energy consumption change time sequence input vector. Here, by calculating the difference between two positions of each vector, the energy consumption change situation between adjacent time points can be obtained, so that the deviation existing due to the influence of factors such as monitoring equipment and environment on the energy consumption data is eliminated, and the energy consumption change situation is reflected more accurately.
Specifically, during operation of the integrated energy storage management system 300, the timing feature extraction unit 340 is configured to pass the energy-consuming time sequence input vector through a timing encoder including a one-dimensional convolution layer and a full connection layer to obtain an energy-consuming related timing feature vector. In the technical scheme of the application, the energy consumption time sequence input vector is passed through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer to obtain an energy consumption related time sequence feature vector. Here, the timing encoder may convert the time-series data into a timing feature vector having a higher level of abstract feature. That is, the energy-consuming time sequence input vector is passed through a sequence encoder comprising a one-dimensional convolution layer and a fully-connected layer, and the relevant features of the utility energy use condition in the time dimension can be extracted. Specifically, the one-dimensional convolution layer can effectively capture local time characteristics of energy use conditions, and the full-connection layer can integrate the local time characteristics to obtain global time characteristics. By the method, time-related characteristics of the energy use condition of the facility can be effectively extracted, and subsequent classification processing is facilitated.
Fig. 3 is a block diagram of a timing feature extraction unit in the integrated energy storage management system according to an embodiment of the present application. As shown in fig. 3, the timing characteristic extraction unit 340 includes: a full-connection encoding subunit 341, configured to perform full-connection encoding on the time-consuming time sequence input vector by using a full-connection layer of the time-sequence encoder according to the following formula to extract high-dimensional implicit features of feature values of each position in the time-consuming time sequence input vector, where the formula is:wherein->Is the energy-consuming time-ordered input vector, +.>Is the output vector, +.>Is a weight matrix, < >>Is a bias vector, ++>Representing matrix multiplication; a one-dimensional convolution encoding subunit 342, configured to perform one-dimensional convolution encoding on the time-consuming time-sequential input vector by using a one-dimensional convolution layer of the time-sequential encoder to extract high-dimensional implicit correlation features between feature values of each position in the time-consuming time-sequential input vector, where the formula is: />Wherein (1)>For convolution kernel +.>Width in direction, ++>For convolution kernel parameter vector, ">For a local vector matrix operating with a convolution kernel function, < ->For the size of the convolution kernel +.>Representing the energy-consuming time sequence input vector, +. >Representing one-dimensional convolutional encoding of the energy-time ordered input vector.
Specifically, during operation of the integrated energy storage management system 300, the change feature extraction unit 350 is configured to pass the energy consumption change time sequence input vector through a multi-scale neighborhood feature extraction module including a first convolution layer and a second convolution layer to obtain an energy consumption change associated time sequence feature vector. After the energy consumption change time sequence input vector is obtained, the energy consumption change time sequence input vector is further processed through a multi-scale neighborhood feature extraction module comprising a first convolution layer and a second convolution layer to obtain an energy consumption change associated time sequence feature vector. It should be understood that the time sequence feature vector associated with the energy consumption change reflects the change condition of the energy use condition of the facility, and features with different scales can be extracted from the time sequence data by adopting the multi-scale neighborhood feature extraction module, so that the accuracy of analyzing and judging the energy use condition of the facility is further improved. Wherein, the multiscale neighborhood feature extraction module comprises: the device comprises a first convolution layer, a second convolution layer parallel to the first convolution layer and a multi-scale feature fusion layer connected with the first convolution layer and the second convolution layer, wherein the first convolution layer uses a one-dimensional convolution kernel with a first length, and the second convolution layer uses a one-dimensional convolution kernel with a second length.
Fig. 4 is a block diagram of a change feature extraction unit in an integrated energy storage management system according to an embodiment of the present application. As shown in fig. 4, the change feature extraction unit 350 includes: a first neighborhood scale feature extraction subunit 351, configured to input the energy consumption variation timing sequence input vector to a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first neighborhood scale energy consumption variation associated timing sequence feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a second neighborhood scale feature extraction subunit 352 configured to input the energy consumption variation timing input vector into a second convolution layer of the multi-scale neighborhood feature extraction module to obtain a second neighborhood scale energy consumption variation associated timing feature vector, where the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length; and a multi-scale cascading subunit 353, configured to cascade the first neighborhood scale energy consumption variation related timing feature vector and the second neighborhood scale energy consumption variation related timing feature vector to obtain the energy consumption variation related timing feature vector. Wherein, the first neighborhood scale feature extraction subunit 351 is configured to: using the multi-scale neighbor The first convolution layer of the domain feature extraction module carries out one-dimensional convolution coding on the energy consumption change time sequence input vector according to the following one-dimensional convolution formula so as to obtain a first neighborhood scale energy consumption change associated time sequence feature vector; wherein, the formula is:wherein->For the first convolution kernel at->Width in direction, ++>For the first convolution kernel parameter vector, +.>For a local vector matrix operating with a convolution kernel function, < ->For the size of the first convolution kernel, +.>Representing the energy consumption variation time sequence input vector, < >>Representing one-dimensional convolution encoding of the energy consumption change time sequence input vector; and, the second neighborhood scale feature extraction subunit 352 is configured to: performing one-dimensional convolution coding on the energy consumption change time sequence input vector by using a second convolution layer of the multi-scale neighborhood feature extraction module according to the following one-dimensional convolution formula to obtain a second neighborhood scale energy consumption change associated time sequence feature vector; wherein, the formula is: />Wherein->For the second rollThe accumulation of heart is->Width in direction, ++>For a second convolution kernel parameter vector, +.>For a local vector matrix operating with a convolution kernel function, < ->For the size of the second convolution kernel, +. >Representing the energy consumption variation time sequence input vector, < >>Representing one-dimensional convolutional encoding of the energy consumption variation time sequence input vector.
Specifically, during the operation of the integrated energy storage management system 300, the feature fusion unit 360 is configured to fuse the energy consumption related time sequence feature vector and the energy consumption variation related time sequence feature vector to obtain a classification feature vector. In the technical scheme of the application, in order to comprehensively consider the energy use condition and the energy consumption change condition of the facility, the accuracy of analyzing and judging the energy use condition of the facility is improved, and the energy consumption related time sequence feature vector and the energy consumption change related time sequence feature vector are fused to obtain the classification feature vector. That is, the energy consumption-related time series feature vector reflecting the time-dependent feature of the utility energy use condition and the energy consumption change-related time series feature vector reflecting the change condition of the utility energy use condition are fused together, so that the classification feature vector has more excellent information characterization capability. Here, when the classification feature vector is obtained by fusing the energy consumption-related timing feature vector and the energy consumption variation-related timing feature vector, the energy consumption-related timing feature vector and the energy consumption variation-related timing feature vector are separated The time sequence associated features of the absolute value of energy consumption and the change value of energy consumption are expressed respectively, so that the probability distribution of class labels of the class labels relative to the classifier is different, and therefore, when the fused classification feature vector passes through the classifier, the characteristic distribution of each energy consumption associated time sequence feature vector and each energy consumption change associated time sequence feature vector backwards propagates through the classifier in a parameter space of a model, and the degradation problem of the characteristic probability distribution expression caused by the shift of the probability distribution is also encountered, so that the characteristic expression effect of the classification feature vector is affected. Based on this, the applicant of the present application introduced a time-series feature vector associated with the energy consumptionTime sequence characteristic vector correlated with the energy consumption change>The probability distribution shift information compensation loss function of (2) is expressed as: />Wherein->And->The energy consumption related timing characteristic vector and the energy consumption variation related timing characteristic vector, +.>And->Compensating for shift superparameter, and +.>Weighted superparameter->Represents a logarithmic function value based on 2, < +.>Representation->The function of the function is that,representation->Function (F)>Representing the probability distribution shift information compensation loss function value. Here, from the energy consumption-dependent timing feature vector +_based on Softmax function >Time sequence characteristic vector correlated with the energy consumption change>The respective derived class probability values themselves follow probability distributions for the respective feature distributions, using the probability distribution shift information to compensate for a loss function for the energy consumption-dependent time-series feature vector +.>Time sequence characteristic vector correlated with the energy consumption change>Information compensation is performed by shifting the probability distribution of the feature representation of (2), and cross information entropy caused by compensation is maximized by a bool function, so that the feature distribution of the classification feature vector after fusion can restore the energy consumption related time sequence feature vector +_x before fusion to the maximum extent>Time sequence characteristic vector correlated with the energy consumption change>Feature probability score of (2)And distributing the expression information, so that the feature expression effect of the classification feature vector is improved, and the accuracy of the classification result obtained by the classifier is improved.
Fig. 5 is a block diagram of a feature fusion unit in an integrated energy storage management system according to an embodiment of the present application. As shown in fig. 5, the feature fusion unit 360 includes: an information compensation loss subunit 361 for calculating a probability distribution shift information compensation loss function value for the energy consumption-related timing feature vector and the energy consumption variation-related timing feature vector; a weighting subunit 362, configured to weight the energy consumption related timing feature vector and the energy consumption variation related timing feature vector based on the probability distribution shift information compensation loss function value to obtain a weighted energy consumption related timing feature vector and a weighted energy consumption variation related timing feature vector; and a fusion subunit 363, configured to fuse the weighted energy consumption related time sequence feature vector and the weighted energy consumption variation related time sequence feature vector to obtain the classification feature vector. The fusing subunit 363 includes fusing the weighted energy consumption related timing feature vector and the weighted energy consumption variation related timing feature vector in a cascade formula to obtain the classification feature vector; wherein, the formula is: Wherein->Representing the weighted energy consumption related time sequence feature vector, < >>Representing the weighted energy consumption change associated time sequence feature vector +.>Representing a cascade function->Representing the classification feature vector.
In particular, in the integrationIn the operation process of the energy storage management system 300, the monitoring result generating unit 370 is configured to pass the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether the energy consumption of the monitored facility is normal. That is, after the classification feature vector is obtained, it is further passed through a classifier to obtain a classification result indicating whether or not the energy consumption of the monitored facility is normal, specifically, the classification feature vector is processed using the classifier in the following formula to obtain a classification result, wherein the formula is:wherein->To->Is a weight matrix>To->For the bias vector +.>Is a classification feature vector. Specifically, the classifier includes a plurality of fully connected layers and a Softmax layer cascaded with a last fully connected layer of the plurality of fully connected layers. In the classification processing of the classifier, multiple full-connection encoding is carried out on the classification feature vectors by using multiple full-connection layers of the classifier to obtain encoded classification feature vectors; further, the encoded classification feature vector is input to a Softmax layer of the classifier, i.e., the encoded classification feature vector is classified using the Softmax classification function to obtain a classification label. The classifier can classify the energy use condition of the facilities according to the classification feature vector, namely, the energy consumption of the monitored facilities is normal and the energy consumption of the monitored facilities is abnormal. Thus, the facility can be timely judged based on the classification result If the energy use condition meets the expectations, if the energy waste or abnormal use phenomenon exists, corresponding measures can be taken to adjust and optimize. Meanwhile, the classification result can also be used as feedback information to help a user optimize an energy management strategy of the facility, so that the energy utilization efficiency is further improved, and the energy consumption cost is reduced.
In summary, the integrated energy storage management system 300 according to the embodiment of the present application is illustrated, which uses the artificial intelligence technology based on deep learning to dig out the implicit association information between the energy usage situation and the energy consumption change situation of the facility, so as to more accurately determine whether the energy usage situation of the whole building or the facility is normal, and in this way, further prompt the energy consumption of the facility or the building to be checked, and timely find out the problems of energy waste, etc.
As described above, the integrated energy storage management system according to the embodiment of the present application may be implemented in various terminal devices. In one example, the integrated energy storage management system 300 according to an embodiment of the present application may be integrated into the terminal device as one software module and/or hardware module. For example, the integrated energy storage management system 300 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the integrated energy storage management system 300 may also be one of a plurality of hardware modules of the terminal device.
Alternatively, in another example, the integrated energy storage management system 300 and the terminal device may be separate devices, and the integrated energy storage management system 300 may be connected to the terminal device through a wired and/or wireless network and transmit the interactive information in a agreed data format.
An exemplary method is: fig. 6 is a flowchart of an integrated energy storage management method according to an embodiment of the present application. As shown in fig. 6, the integrated energy storage management method according to the embodiment of the application includes the steps of: s110, acquiring energy consumption values of monitored facilities acquired by the IESS unit for a plurality of days within a preset time span; s120, arranging the energy consumption values of the monitored facilities for a plurality of days in a preset time span into energy-consuming time sequence input vectors according to a time dimension; s130, calculating the difference value between the energy consumption values of two positions of each vector in the energy consumption time sequence input vector to obtain an energy consumption change time sequence input vector; s140, the energy consumption time sequence input vector passes through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer to obtain an energy consumption related time sequence feature vector; s150, the energy consumption change time sequence input vector passes through a multi-scale neighborhood feature extraction module comprising a first convolution layer and a second convolution layer to obtain an energy consumption change associated time sequence feature vector; s160, fusing the energy consumption related time sequence feature vector and the energy consumption change related time sequence feature vector to obtain a classification feature vector; and S170, passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the energy consumption of the monitored facility is normal or not.
In one example, in the above integrated energy storage management method, the step S140 includes: and performing full-connection coding on the energy-time sequence input vector by using a full-connection layer of the time sequence encoder according to the following formula to extract high-dimensional implicit features of feature values of all positions in the energy-time sequence input vector, wherein the formula is as follows:wherein->Is the energy-consuming time-ordered input vector, +.>Is the output vector, +.>Is a weight matrix, < >>Is a bias vector, ++>Representing matrix multiplication; one-dimensional convolution layer using the timing encoder is paired with the following formulaThe energy-time sequence input vector is subjected to one-dimensional convolution coding to extract high-dimensional implicit correlation features among feature values of each position in the energy-time sequence input vector, wherein the formula is as follows: />Wherein (1)>Is convolution kernel inWidth in direction, ++>For convolution kernel parameter vector, ">For a local vector matrix operating with a convolution kernel function, < ->For the size of the convolution kernel +.>Representing the energy-consuming time sequence input vector, +.>Representing one-dimensional convolutional encoding of the energy-time ordered input vector.
In one example, in the above integrated energy storage management method, the step S150 includes: inputting the energy consumption change time sequence input vector into a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first neighborhood scale energy consumption change associated time sequence feature vector, wherein the first convolution layer is provided with a first one-dimensional convolution kernel with a first length; inputting the energy consumption change time sequence input vector into a second convolution layer of the multi-scale neighborhood feature extraction module to obtain a second neighborhood scale energy consumption change related time sequence feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length The second length; and cascading the first neighborhood scale energy consumption change related time sequence feature vector and the second neighborhood scale energy consumption change related time sequence feature vector to obtain the energy consumption change related time sequence feature vector. Wherein, the multiscale neighborhood feature extraction module comprises: the device comprises a first convolution layer, a second convolution layer parallel to the first convolution layer and a multi-scale feature fusion layer connected with the first convolution layer and the second convolution layer, wherein the first convolution layer uses a one-dimensional convolution kernel with a first length, and the second convolution layer uses a one-dimensional convolution kernel with a second length. More specifically, inputting the energy consumption change time sequence input vector into a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first neighborhood scale energy consumption change associated time sequence feature vector, including: performing one-dimensional convolution coding on the energy consumption change time sequence input vector by using a first convolution layer of the multi-scale neighborhood feature extraction module according to the following one-dimensional convolution formula to obtain a first neighborhood scale energy consumption change associated time sequence feature vector; wherein, the formula is:wherein- >For the first convolution kernel at->Width in direction, ++>For the first convolution kernel parameter vector, +.>For a local vector matrix operating with a convolution kernel function, < ->For the size of the first convolution kernel, +.>Representing the energy consumptionVarying the timing input vector, ">Representing one-dimensional convolution encoding of the energy consumption change time sequence input vector; and the second neighborhood scale feature extraction subunit is configured to: performing one-dimensional convolution coding on the energy consumption change time sequence input vector by using a second convolution layer of the multi-scale neighborhood feature extraction module according to the following one-dimensional convolution formula to obtain a second neighborhood scale energy consumption change associated time sequence feature vector; wherein, the formula is:wherein->For the second convolution kernel>Width in direction, ++>For a second convolution kernel parameter vector, +.>For a local vector matrix operating with a convolution kernel function, < ->For the size of the second convolution kernel, +.>Representing the energy consumption variation time sequence input vector, < >>Representing one-dimensional convolutional encoding of the energy consumption variation time sequence input vector.
In one example, in the above integrated energy storage management method, the step S160 includes: calculating a probability distribution shift information compensation loss function for the energy consumption-related timing feature vector and the energy consumption variation-related timing feature vector A value; weighting the energy consumption related time sequence feature vector and the energy consumption change related time sequence feature vector based on the probability distribution shift information compensation loss function value to obtain a weighted energy consumption related time sequence feature vector and a weighted energy consumption change related time sequence feature vector; and fusing the weighted energy consumption related time sequence feature vector and the weighted energy consumption change related time sequence feature vector to obtain the classification feature vector. Wherein calculating a probability distribution shift information compensation loss function value for the energy consumption-related timing feature vector and the energy consumption variation-related timing feature vector includes: calculating probability distribution shift information compensation loss function values for the energy consumption-related timing feature vector and the energy consumption variation-related timing feature vector in the following formula; wherein, the formula is:wherein->And->The energy consumption related timing characteristic vector and the energy consumption variation related timing characteristic vector, +.>And->Compensating for shift superparameter, and +.>Weighted superparameter->Represents a logarithmic function value based on 2, < +.>Representation->Function (F)>Representation ofFunction (F)>Representing the probability distribution shift information compensation loss function value. More specifically, fusing the weighted energy consumption-related timing feature vector and the weighted energy consumption variation-related timing feature vector to obtain the classification feature vector includes: fusing the weighted energy consumption related time sequence feature vector and the weighted energy consumption change related time sequence feature vector by using the following cascade formula to obtain the classification feature vector; wherein, the formula is: / >Wherein->Representing the weighted energy consumption related time sequence feature vector, < >>Representing the weighted energy consumption change associated time sequence feature vector +.>Representing a cascade function->Representing the classification feature vector.
In one example, in the above integrated energy storage management method, the step S170 includes: processing the classification feature vector using the classifier to obtain a classification result with the following formula:wherein->To->Is a weight matrix>To->For the bias vector +.>Is a classification feature vector.
In summary, the integrated energy storage management method according to the embodiment of the application is explained, and by adopting the artificial intelligence technology based on deep learning to dig out the implicit association information between the energy use condition and the energy consumption change condition of the facility, whether the energy use condition of the whole building or the facility is normal or not can be judged more accurately, and by adopting the mode, the energy utilization of the facility or the building can be further prompted to be checked, and the problems of energy waste and the like can be discovered in time.
Exemplary electronic device: next, an electronic device according to an embodiment of the present application is described with reference to fig. 7.
Fig. 7 illustrates a block diagram of an electronic device according to an embodiment of the application.
As shown in fig. 7, the electronic device 10 includes one or more processors 11 and a memory 12.
The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.
Memory 12 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. One or more computer program instructions may be stored on the computer readable storage medium that can be executed by the processor 11 to implement the functions in the integrated energy storage management system of the various embodiments of the present application described above and/or other desired functions. Various contents such as a generated energy multi-scale feature vector may also be stored in the computer-readable storage medium.
In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown).
The input means 13 may comprise, for example, a keyboard, a mouse, etc.
The output device 14 may output various information including the classification result and the like to the outside. The output means 14 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.
Of course, only some of the components of the electronic device 10 that are relevant to the present application are shown in fig. 7 for simplicity, components such as buses, input/output interfaces, etc. are omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.
Exemplary computer program product and computer readable storage Medium: in addition to the methods and apparatus described above, embodiments of the application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform steps in the functions of the integrated energy storage management method according to various embodiments of the application described in the "exemplary systems" section of this specification.
The computer program product may write program code for performing operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present application may also be a computer-readable storage medium, on which computer program instructions are stored, which when being executed by a processor, cause the processor to perform steps in the functions of the integrated energy storage management method according to various embodiments of the present application described in the "exemplary systems" section above in this specification.
The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not intended to be limiting, and these advantages, benefits, effects, etc. are not to be considered as essential to the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not necessarily limited to practice with the above described specific details.
The block diagrams of the devices, apparatuses, devices, systems referred to in the present application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (10)
1. An integrated energy storage management system, comprising: an energy consumption monitoring unit for acquiring energy consumption values of the monitored facility acquired by the IESS unit for a plurality of days within a predetermined time span; the data structuring unit is used for arranging the energy consumption values of the monitored facility for a plurality of days in a preset time span into energy-consuming time sequence input vectors according to a time dimension; the change information characterization unit is used for calculating the difference value between the energy consumption values of two positions of each vector in the energy consumption time sequence input vector to obtain an energy consumption change time sequence input vector; the time sequence feature extraction unit is used for enabling the energy-consumption time sequence input vector to pass through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer to obtain an energy-consumption related time sequence feature vector; the change feature extraction unit is used for enabling the energy consumption change time sequence input vector to pass through a multi-scale neighborhood feature extraction module comprising a first convolution layer and a second convolution layer to obtain an energy consumption change associated time sequence feature vector; the feature fusion unit is used for fusing the energy consumption related time sequence feature vector and the energy consumption change related time sequence feature vector to obtain a classification feature vector; and the monitoring result generating unit is used for passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the energy consumption of the monitored facility is normal or not.
2. The integrated energy storage management system of claim 1, wherein the timing feature extraction unit comprises: and the full-connection coding subunit is used for performing full-connection coding on the energy-consumption time sequence input vector by using a full-connection layer of the time sequence coder according to the following formula to extract high-dimensional implicit characteristics of characteristic values of all positions in the energy-consumption time sequence input vector, wherein the formula is as follows:wherein->Is the energy-consuming time-ordered input vector, +.>Is the output vector, +.>Is a weight matrix, < >>Is a bias vector, ++>Representing matrix multiplication; a one-dimensional convolution coding subunit, configured to perform one-dimensional convolution coding on the energy-saving time sequence input vector by using a one-dimensional convolution layer of the time sequence encoder according to the following formula to extract high-dimensional implicit correlation features between feature values of each position in the energy-saving time sequence input vector, where the formula is: />Wherein (1)>For convolution kernel +.>Width in direction, ++>For convolution kernel parameter vector, ">For a local vector matrix operating with a convolution kernel function, < ->For the size of the convolution kernel +.>Representing the energy-consuming time sequence input vector, +.>Representing one-dimensional convolutional encoding of the energy-time ordered input vector.
3. The integrated energy storage management system of claim 2, wherein the multi-scale neighborhood feature extraction module comprises: the device comprises a first convolution layer, a second convolution layer parallel to the first convolution layer and a multi-scale feature fusion layer connected with the first convolution layer and the second convolution layer, wherein the first convolution layer uses a one-dimensional convolution kernel with a first length, and the second convolution layer uses a one-dimensional convolution kernel with a second length.
4. The integrated energy storage management system of claim 3, wherein the change feature extraction unit comprises: a first neighborhood scale feature extraction subunit, configured to input the energy consumption change timing sequence input vector to a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first neighborhood scale energy consumption change associated timing sequence feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a second neighborhood scale feature extraction subunit, configured to input the energy consumption variation timing sequence input vector to a second convolution layer of the multi-scale neighborhood feature extraction module to obtain a second neighborhood scale energy consumption variation associated timing sequence feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; the multi-scale cascading subunit is configured to cascade the first neighborhood scale energy consumption variation related time sequence feature vector and the second neighborhood scale energy consumption variation related time sequence feature vector to obtain the energy consumption variation related time sequence feature vector, where the first neighborhood scale feature extraction subunit is configured to: performing one-dimensional convolution coding on the energy consumption change time sequence input vector by using a first convolution layer of the multi-scale neighborhood feature extraction module according to the following one-dimensional convolution formula to obtain a first neighborhood scale energy consumption change associated time sequence feature vector; wherein, the formula is: Wherein->For the first convolution kernel at->Width in direction, ++>For the first convolution kernel parameter vector, +.>For a local vector matrix operating with a convolution kernel function, < ->For the size of the first convolution kernel, +.>Representing the energy consumption variation time sequence input vector, < >>Representing one-dimensional convolution encoding of the energy consumption change time sequence input vector; and the second neighborhood scale feature extraction subunit is configured to: performing one-dimensional convolution coding on the energy consumption change time sequence input vector by using a second convolution layer of the multi-scale neighborhood feature extraction module according to the following one-dimensional convolution formula to obtain a second neighborhood scale energy consumption change associated time sequence feature vector; wherein, the formula is:wherein->For the second convolution kernel>Width in direction, ++>For a second convolution kernel parameter vector, +.>For a local vector matrix operating with a convolution kernel function, < ->For the size of the second convolution kernel, +.>Representing the energy consumption variation time sequence input vector, < >>Representing one-dimensional convolutional encoding of the energy consumption variation time sequence input vector.
5. The integrated energy storage management system of claim 4, wherein the feature fusion unit comprises: an information compensation loss subunit for calculating a probability distribution shift information compensation loss function value for the energy consumption-related timing feature vector and the energy consumption variation-related timing feature vector; the weighting subunit is used for weighting the energy consumption related time sequence feature vector and the energy consumption change related time sequence feature vector based on the probability distribution shift information compensation loss function value so as to obtain a weighted energy consumption related time sequence feature vector and a weighted energy consumption change related time sequence feature vector; and a fusion subunit, configured to fuse the weighted energy consumption related time sequence feature vector and the weighted energy consumption variation related time sequence feature vector to obtain the classification feature vector.
6. The integrated energy storage management system of claim 5, wherein the information compensation loss subunit is configured to: calculating probability distribution shift information compensation loss function values for the energy consumption-related timing feature vector and the energy consumption variation-related timing feature vector in the following formula; wherein, the formula is:wherein->And->The energy consumption related timing characteristic vector and the energy consumption variation related timing characteristic vector, +.>And->Compensating for shift superparameter, and +.>Weighted superparameter->Represents a logarithmic function value based on 2, < +.>Representation->Function (F)>Representation ofFunction (F)>Representing the probability distribution shift information compensation loss function value.
7. The integrated energy storage management system of claim 6, wherein the fusion subunit is configured to: fusing the weighted energy consumption related time sequence feature vector and the weighted energy consumption change related time sequence feature vector by using the following cascade formula to obtain the classification feature vector; which is a kind ofThe formula is as follows:wherein->Representing the weighted energy consumption related time sequence feature vector, < >>Representing the weighted energy consumption change associated timing feature vector, Representing a cascade function->Representing the classification feature vector.
8. The integrated energy storage management system of claim 7, wherein the monitoring result generation unit is configured to: processing the classification feature vector using the classifier to obtain a classification result with the following formula:wherein->To->Is a weight matrix>To->For the bias vector +.>Is a classification feature vector.
9. An integrated energy storage management method, comprising: acquiring energy consumption values of the monitored facility acquired by the IESS unit for a plurality of days within a preset time span; arranging the energy consumption values of the monitored facilities for a plurality of days in a preset time span into energy-saving time sequence input vectors according to a time dimension; calculating the difference between the energy consumption values of two positions of each vector in the energy consumption time sequence input vector to obtain an energy consumption change time sequence input vector; the energy consumption time sequence input vector passes through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer to obtain an energy consumption related time sequence feature vector; the energy consumption change time sequence input vector passes through a multi-scale neighborhood feature extraction module comprising a first convolution layer and a second convolution layer to obtain an energy consumption change associated time sequence feature vector; fusing the energy consumption related time sequence feature vector and the energy consumption change related time sequence feature vector to obtain a classification feature vector; and passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the energy consumption of the monitored facility is normal or not.
10. The integrated energy storage management method of claim 9, wherein passing the classification feature vector through a classifier to obtain a classification result, the classification result being used to indicate whether energy consumption of the monitored facility is normal, comprises: processing the classification feature vector using the classifier to obtain a classification result with the following formula:wherein->To->Is a weight matrix>To->For the bias vector +.>Is a classification feature vector.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310468867.9A CN117094651A (en) | 2023-04-27 | 2023-04-27 | Integrated energy storage management system and energy management method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310468867.9A CN117094651A (en) | 2023-04-27 | 2023-04-27 | Integrated energy storage management system and energy management method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117094651A true CN117094651A (en) | 2023-11-21 |
Family
ID=88775853
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310468867.9A Pending CN117094651A (en) | 2023-04-27 | 2023-04-27 | Integrated energy storage management system and energy management method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117094651A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117892252A (en) * | 2024-03-18 | 2024-04-16 | 浙江威星电子系统软件股份有限公司 | Intelligent park operation management platform based on big data |
CN118071385A (en) * | 2024-04-18 | 2024-05-24 | 湖北华中电力科技开发有限责任公司 | Marketing business fusion method and system oriented to all-Internet-of-things architecture |
-
2023
- 2023-04-27 CN CN202310468867.9A patent/CN117094651A/en active Pending
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117892252A (en) * | 2024-03-18 | 2024-04-16 | 浙江威星电子系统软件股份有限公司 | Intelligent park operation management platform based on big data |
CN117892252B (en) * | 2024-03-18 | 2024-07-19 | 浙江威星电子系统软件股份有限公司 | Intelligent park operation management platform based on big data |
CN118071385A (en) * | 2024-04-18 | 2024-05-24 | 湖北华中电力科技开发有限责任公司 | Marketing business fusion method and system oriented to all-Internet-of-things architecture |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhai et al. | MLP neural network based gas classification system on Zynq SoC | |
CN117094651A (en) | Integrated energy storage management system and energy management method | |
US20160239592A1 (en) | Data-driven battery aging model using statistical analysis and artificial intelligence | |
Wei et al. | Learning the health index of complex systems using dynamic conditional variational autoencoders | |
CN115759658A (en) | Enterprise energy consumption data management system suitable for smart city | |
Guo et al. | A hybrid prognosis scheme for rolling bearings based on a novel health indicator and nonlinear Wiener process | |
Hou et al. | Stock trend prediction with multi-granularity data: A contrastive learning approach with adaptive fusion | |
He et al. | Feature extraction of analogue circuit fault signals via cross‐wavelet transform and variational Bayesian matrix factorisation | |
KR102150170B1 (en) | Method and System for Power Load Forecasting based on Pattern Tagging | |
CN110502751A (en) | Bulk power grid operation situation cognitive method, terminal device and storage medium | |
CN117041017A (en) | Intelligent operation and maintenance management method and system for data center | |
Raghavendra et al. | Artificial humming bird with data science enabled stability prediction model for smart grids | |
CN116759053A (en) | Medical system prevention and control method and system based on Internet of things system | |
CN118151020B (en) | Method and system for detecting safety performance of battery | |
CN115324843A (en) | Wind generating set fault diagnosis system and method based on monitoring data | |
Baur et al. | Explainability and interpretability in electric load forecasting using machine learning techniques–A review | |
Liu et al. | State of health estimation of lithium-ion batteries based on multi-feature extraction and temporal convolutional network | |
Song et al. | Dynamic thermal line rating model of conductor based on prediction of meteorological parameters | |
Chen et al. | Variational encoding based on factorized temporal-channel fusion and feature fusion for interpretable remaining useful life prediction | |
CN116785967B (en) | Automatic batching system for electronic grade lithium hexafluorophosphate preparation | |
CN116526670A (en) | Information fusion method for power big data visualization | |
CN117011609A (en) | Automatic tracking system and method for textile processing progress | |
CN115018212B (en) | Power generation water consumption prediction analysis method and system and cloud platform | |
Yu et al. | Interpretable data‐driven contingency classification for real‐time corrective security‐constrained economic dispatch | |
Xu et al. | High-Dimensional Feature Selection for Power System Congestion Event Prognosis With Enhanced Evolutionary Computation |
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 |