CN116595896A - Big data-based energy management system - Google Patents
Big data-based energy management system Download PDFInfo
- Publication number
- CN116595896A CN116595896A CN202310868139.7A CN202310868139A CN116595896A CN 116595896 A CN116595896 A CN 116595896A CN 202310868139 A CN202310868139 A CN 202310868139A CN 116595896 A CN116595896 A CN 116595896A
- Authority
- CN
- China
- Prior art keywords
- energy management
- representing
- matrix
- column
- policy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000011159 matrix material Substances 0.000 claims abstract description 82
- 238000011156 evaluation Methods 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 10
- 238000005457 optimization Methods 0.000 claims abstract description 9
- 238000013210 evaluation model Methods 0.000 claims abstract description 8
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 claims description 32
- 238000012546 transfer Methods 0.000 claims description 24
- 238000010438 heat treatment Methods 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 18
- 239000002918 waste heat Substances 0.000 claims description 18
- 239000003345 natural gas Substances 0.000 claims description 16
- 238000000034 method Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 239000002699 waste material Substances 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 206010027175 memory impairment Diseases 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Resources & Organizations (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Medical Informatics (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Computer Hardware Design (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Entrepreneurship & Innovation (AREA)
- Databases & Information Systems (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Algebra (AREA)
- Geometry (AREA)
Abstract
The invention relates to the technical field of big data, and discloses an energy management system based on big data, which comprises the following components: the strategy pre-generation module is used for establishing an energy management strategy matrix; a policy generator for generating a plurality of initial energy management policy matrices conforming to constraints of an energy management policy; a policy set generation module that generates a first policy set based on an initial energy management policy matrix; the first processing module is used for calculating the first strategy set to generate a second strategy set; the second processing module is used for inputting the second energy management strategy matrixes in the second strategy set into the energy optimization evaluation model, outputting evaluation scores of each second energy management strategy matrix, and selecting the first K second energy management strategy matrixes as final energy management strategy matrixes according to the descending order of the evaluation scores.
Description
Technical Field
The invention relates to the technical field of big data, in particular to an energy management system based on big data.
Background
Limited clean energy such as solar energy and recovered energy cannot meet the heat supply requirement of a heat supply system, so that the requirement of the heat supply system is comprehensively regulated and controlled by combining the traditional energy, the complementary management strategy of multiple types of multiple energy is complex, and the traditional interference supply management strategy can operate the heat supply system, but cannot maximally utilize the clean energy and reduce the energy waste.
Disclosure of Invention
The invention provides an energy management system based on big data, which solves the technical problems that the management strategy of multi-type multi-energy complementation is complex in the related technology, and the traditional interference supply management strategy cannot maximally utilize clean energy and reduce energy waste.
The invention provides an energy management system based on big data, comprising: a policy pre-generation module for establishing an energy management policy matrix,and (3) representing elements of an ith row and a tth column of the energy management strategy matrix, wherein i is less than or equal to n, n is the number of heat consumption units, t is less than or equal to m, and m is the number of moments in one management period.
Wherein->The ith solar heat source unit at time tHeat supply demand, ->Representing the heat supply demand of the ith heat consumption unit at time t, which is assigned to the b-th waste heat source device,/>And (3) representing the heat supply demand of the ith heat consumption unit at the moment t, wherein the values of a, b and c are the total number of the solar heat source device, the waste heat source device and the natural gas heat source device respectively.
And the policy constraint generation module is used for generating constraint conditions of the energy management policy.
A policy generator for generating a plurality of initial energy management policy matrices conforming to constraints of an energy management policy.
A policy set generation module that generates a first policy set based on the initial energy management policy matrix, the first policy set including N first energy management policy matrices that conform to constraints of the energy management policies.
And the first processing module is used for calculating the first strategy set to generate a second strategy set.
The second processing module is used for inputting the second energy management strategy matrixes in the second strategy set into the energy optimization evaluation model, outputting evaluation scores of each second energy management strategy matrix, and selecting the first K second energy management strategy matrixes as final energy management strategy matrixes according to the descending order of the evaluation scores.
Further, the method further comprises the following steps: and the information extraction module is used for extracting the information of the solar heat source device, the waste heat source device and the natural gas heat source device, the information of the heat consumption unit and the information of the heat supply system.
Further, constraints of the energy management strategy include energy balance constraints and power constraints.
Energy balance constraint:，/>the heat supplied at the time t for all solar heat sources; />The heat supplied by all waste heat source devices at the time t; />Heat supplied at time t for all natural gas heat sources; />The heat dissipated to the outside at the time t by the heat supply system; />Is the total heating demand at time t.
Power constraint:；/>；/>。
for maximum heating power of solar heat source at one moment, < >>The heat supply power at the time t of the solar heat source device is provided; />For maximum heating power of the waste heat source at one moment,/->The heat supply power at the time t of the waste heat source device is used for supplying heat; />For maximum heating power of the natural gas heat source at one moment, < >>The heating power at the t moment of the natural gas heat source device.
Further, the generation policy matrix is generated by a historical energy management policy.
Further, the method of generating the first energy management policy based on the initial energy management policy matrix includes: step 101, two initial energy management strategy matrixes are randomly selected from the initial energy management strategy matrixes, namely a first initial energy management strategy matrix and a second initial energy management strategy matrix.
Step 102, randomly selecting an element of column A from the first initial energy management strategy matrix to replace the element of the corresponding column of the second initial energy management strategy matrix to obtain a third initial energy management strategy matrix.
Step 103, randomly selecting the element of the B row from the third initial energy management strategy matrix to replace the element of the corresponding row of the second initial energy management strategy matrix to obtain a fourth initial energy management strategy matrix.
Step 104, iteratively executing the steps 101-103 until M fourth initial energy management strategy matrixes are obtained; and then selecting N fourth initial energy management strategy matrixes meeting the constraint conditions of the energy management strategies as the first energy management strategy matrixes.
Further, the energy optimization evaluation model is a neural network.
Further, m columns of LSTM cells, each column including n LSTM cells, an i-th row and a t-th column of LSTM cells input elements of the i-th row and the t-th column of the second energy management policy matrix.
LSTM cell output of mth columnInputting a classifier, wherein the classification space of the classifier is +.>The value range of the evaluation score +.>The mean value is discretized, and one classification of the classification space Q corresponds to one score after the mean value is discretized.
Further, the operation procedure of the LSTM cells of the ith row and the nth column is as follows: forgetting doorThe calculation formula of (2) is as follows:wherein->Indicating the preposition status,/->；/>Representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; />Representation->Transfer to->Corresponding weight matrix, < >>Representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing the bias term.
Input doorThe calculation formula of (2) is as follows: />Wherein->The state of the front-end is indicated,；/>representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; />Representation input +.>Transfer to->Corresponding weight matrix, < >>Representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing the bias term.
Intermediate stateCan be expressed as follows: />Wherein->Indicating the preposition status,/->；/>Representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; />Representation input +.>Transfer to->Corresponding weight matrix, < >>Representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing the bias term.
Output stateThe calculation formula of (2) is as follows: />Wherein->Is the output state transferred by LSTM of row i, column t-1, +.>、/>、/>Is the calculation result of forget gate, input gate and intermediate state.
Indicating forgetfulness door->And the output state of LSTM of row i, column t-1 +.>The multiplication is performed point by point,；/>representing I/O gate->And intermediate state->Point-wise multiplication is performed, ">。
Output doorThe calculation formula of (2) is as follows: />Wherein->Indicating the preposition status,/->；/>Representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; wherein->Representation input +.>Transfer to->Corresponding weight matrix, < >>Representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing the bias term.
Output ofThe calculation formula of (2) is as follows: />Will output door->And->Multiplying point by point to obtain the output of LSTM unit of ith row and t column>。
Definition of the definition、/>、/>All assigned a value of 0.
Further, the method further comprises the following steps: and a policy generation module that generates management policy information based on the final energy management policy matrix.
The management policy information includes: m times, each time corresponds to a total power meter and an output power meter, and each unit cell of the total power meter comprises an ID of a heat source device and the total heat supply power of the heat source device.
The unit cell of the first row and the j-th column of the output power meter comprises the heat supply power provided by the heat source device with the ID of j to the heat consumption unit with the ID of l.
Further, management information of the heating system is obtained directly based on the information of the final energy management policy matrix.
The invention has the beneficial effects that: the invention can comprehensively consider the supply loss, the heat supply requirement of the management period and the change condition of the heat supply power to generate a refined management strategy, and the energy distribution of the multi-energy supply side and the multi-energy requirement side can maximally utilize clean energy and reduce energy waste.
Drawings
Fig. 1 is a schematic diagram of a big data based energy management system according to the present invention.
Fig. 2 is a flow chart of a method of generating a first energy management policy based on an initial energy management policy matrix of the present invention.
Fig. 3 is a schematic diagram of a second module of the big data based energy management system of the present invention.
In the figure: the system comprises an information extraction module 101, a strategy pre-generation module 102, a strategy constraint generation module 103, a strategy generator 104, a strategy set generation module 105, a first processing module 106, a second processing module 107 and a strategy generation module 108.
Detailed Description
The subject matter described herein will now be discussed with reference to example embodiments. It is to be understood that these embodiments are merely discussed so that those skilled in the art may better understand and implement the subject matter described herein and that changes may be made in the function and arrangement of the elements discussed without departing from the scope of the disclosure herein. Various examples may omit, replace, or add various procedures or components as desired. In addition, features described with respect to some examples may be combined in other examples as well.
As shown in fig. 1, an energy management system based on big data includes: and the information extraction module 101 is used for extracting information of the solar heat source device, the waste heat source device and the natural gas heat source device, information of a heat consumption unit and information of a heat supply system.
A policy pre-generation module 102 for establishing an energy management policy matrix,representing energy source pipeThe ith row and the tth column of the management strategy matrix are elements, wherein i is less than or equal to n, n is the number of heat consumption units, t is less than or equal to m, and m is the number of moments in one management period.
For example, one management period is 24 hours, and the time is divided by hours, and then 24 times are included, and one hour executes one time energy management strategy.
Wherein->Indicating the heat supply demand of the ith heat consumption unit at time t, which is assigned to the a-th solar heat source device,/->Representing the heat supply demand of the ith heat consumption unit at time t, which is assigned to the b-th waste heat source device,/>And (3) representing the heat supply demand of the ith heat consumption unit at the moment t, wherein the values of a, b and c are the total number of the solar heat source device, the waste heat source device and the natural gas heat source device respectively.
Policy constraint generation module 103 for generating constraints of the energy management policy, including energy balance constraints and power constraints.
Energy balance constraint:，/>the heat supplied at the time t for all solar heat sources; />The heat supplied by all waste heat source devices at the time t; />For all natural gas heat sourcesHeat supplied at time t; />The heat dissipated to the outside at the time t by the heat supply system; />Is the total heating demand at time t.
Power constraint:；/>；/>。
for maximum heating power of solar heat source at one moment, < >>The heat supply power at the time t of the solar heat source device is provided; />For maximum heating power of the waste heat source at one moment,/->The heat supply power at the time t of the waste heat source device is used for supplying heat; />For maximum heating power of the natural gas heat source at one moment, < >>The heating power at the t moment of the natural gas heat source device.
A policy generator 104 for generating a plurality of initial energy management policy matrices that conform to constraints of the energy management policies.
The generation policy matrix may be generated by a historical energy management policy.
Generally, the reference available contemporaneous historical energy management policies are limited, and an initial energy management policy matrix that meets the constraints of the energy management policies may be generated by way of random generation.
A policy set generation module 105 that generates a first policy set based on the initial energy management policy matrix, the first policy set including N first energy management policy matrices that conform to constraints of the energy management policies.
As shown in fig. 2, the method of generating the first energy management policy based on the initial energy management policy matrix includes: step 101, two initial energy management strategy matrixes are randomly selected from the initial energy management strategy matrixes, namely a first initial energy management strategy matrix and a second initial energy management strategy matrix.
Step 102, randomly selecting an element of column A from the first initial energy management strategy matrix to replace the element of the corresponding column of the second initial energy management strategy matrix to obtain a third initial energy management strategy matrix.
Step 103, randomly selecting the element of the B row from the third initial energy management strategy matrix to replace the element of the corresponding row of the second initial energy management strategy matrix to obtain a fourth initial energy management strategy matrix.
Step 104, iteratively executing the steps 101-103 until M fourth initial energy management strategy matrixes are obtained; and then selecting N fourth initial energy management strategy matrixes meeting the constraint conditions of the energy management strategies as the first energy management strategy matrixes.
The value of M defaults to twice of N, and the constraint condition is difficult to meet under the condition that a heating system is complex, so that the value of M can be increased.
A first processing module 106 for computing the first set of policies to generate a second set of policies.
The second processing module 107 is configured to input the second energy management policy matrices in the second policy set into the energy optimization evaluation model, output an evaluation score of each second energy management policy matrix, and select the first K second energy management policy matrices as final energy management policy matrices according to the order of the evaluation scores from big to small.
In one embodiment of the invention, the energy optimization evaluation model is a neural network.
The energy optimization evaluation model comprises: m (number of times within one management period) columns of LSTM cells, each column including n (number of heat consumption units) LSTM cells, the LSTM cells of the ith row and the tth column inputting elements of the ith row and the tth column of the second energy management policy matrix.
The operation procedure of the LSTM unit of the ith row and the tth column is as follows: forgetting doorThe calculation formula of (2) is as follows:wherein->Indicating the preposition status,/->；/>Representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; />Representation->Transfer to->Corresponding weight matrix, < >>Representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing the bias term.
Input doorThe calculation formula of (2) is as follows: />Wherein->The state of the front-end is indicated,；/>representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; />Representation input +.>Transfer to->Corresponding weightThe matrix is formed by a matrix of,representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing the bias term.
Intermediate stateCan be expressed as follows: />Wherein->Indicating the preposition status,/->；/>Representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; />Representation input +.>Transfer to->Corresponding weight matrix, < >>Representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing the bias term.
Output stateThe calculation formula of (2) is as follows: />Wherein->Is the output state transferred by LSTM of row i, column t-1, +.>、/>、/>Is the calculation result of forget gate, input gate and intermediate state.
Indicating forgetfulness door->And the output state of LSTM of row i, column t-1 +.>The multiplication is performed point by point,；/>representing I/O gate->And intermediate state->Point-wise multiplication is performed, ">。
Output doorThe calculation formula of (2) is as follows: />Wherein->Indicating the preposition status,/->；/>Representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; wherein->Representation input +.>Transfer to->Corresponding weight momentArray (S)>Representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing the bias term.
Output ofThe calculation formula of (2) is as follows: />Will output door->And->Multiplying point by point to obtain the output of LSTM unit of ith row and t column>。
LSTM cell output of mth columnInputting a classifier, wherein the classification space of the classifier is +.>The value range of the evaluation score +.>The mean value is discretized, and one classification of the classification space Q corresponds to one score after the mean value is discretized.
In one embodiment of the present invention, all LSTM cells of the mth column are outputAfter vectorization, a classifier is input.
In one embodiment of the present invention, all LSTM cells of the mth column are outputAnd respectively inputting the vectorized evaluation scores into a classifier, and taking the average value of the output evaluation scores as a final evaluation score.
In the above-mentioned formula(s),and->Representing an activation function, in one embodiment of the invention,/->For sigmoid function, +.>As a hyperbolic tangent function.
Definition of the definition、/>、/>All assigned a value of 0.
In one embodiment of the invention, the element rows of the second energy management policy matrix are vectorized and then input as feature vectors into the energy optimization assessment model.
The training of the neural network model is a conventional technical means, and will not be described herein.
In one embodiment of the invention, the management information of the heating system is obtained directly based on the information of the final energy management policy matrix.
In one embodiment of the present invention, as shown in fig. 3, an energy management system based on big data further includes: a policy generation module 108 that generates management policy information based on the final energy management policy matrix.
The management policy information includes: m times, each time corresponds to a total power meter and an output power meter, and each unit cell of the total power meter comprises an ID of a heat source device and the total heat supply power of the heat source device.
The unit cell of the first row and the j-th column of the output power meter comprises the heat supply power provided by the heat source device with the ID of j to the heat consumption unit with the ID of l.
The embodiment has been described above with reference to the embodiment, but the embodiment is not limited to the above-described specific implementation, which is only illustrative and not restrictive, and many forms can be made by those of ordinary skill in the art, given the benefit of this disclosure, are within the scope of this embodiment.
Claims (10)
1. An energy management system based on big data, comprising: a policy pre-generation module for establishing an energy management policy matrix,elements representing the ith row and the nth column of the energy management strategy matrix, wherein i is less than or equal to n, n is the number of heat consumption units, t is less than or equal to m, and m is the number of moments in one management period; />Wherein->Indicating the heat supply demand of the ith heat consumption unit at time t, which is assigned to the a-th solar heat source device,/->Representing the heat supply demand of the ith heat consumption unit at time t, which is assigned to the b-th waste heat source device,/>The heat supply demand of the ith heat consumption unit at the moment t is distributed to the c natural gas heat source device, wherein the values of a, b and c are the total number of the solar heat source device, the waste heat source device and the natural gas heat source device respectively;
a policy constraint generation module for generating constraint conditions of an energy management policy;
a policy generator for generating a plurality of initial energy management policy matrices conforming to constraints of an energy management policy;
a policy set generation module that generates a first policy set based on the initial energy management policy matrix, the first policy set including N first energy management policy matrices that conform to constraints of the energy management policies;
the first processing module is used for calculating the first strategy set to generate a second strategy set;
the second processing module is used for inputting the second energy management strategy matrixes in the second strategy set into the energy optimization evaluation model, outputting evaluation scores of each second energy management strategy matrix, and selecting the first K second energy management strategy matrixes as final energy management strategy matrixes according to the descending order of the evaluation scores.
2. The big data based energy management system of claim 1, further comprising: and the information extraction module is used for extracting the information of the solar heat source device, the waste heat source device and the natural gas heat source device, the information of the heat consumption unit and the information of the heat supply system.
3. The big data based energy management system of claim 1, wherein the constraints of the energy management strategy include energy balance constraints and power constraints;
energy balance constraint:，/>the heat supplied at the time t for all solar heat sources; />The heat supplied by all waste heat source devices at the time t; />Heat supplied at time t for all natural gas heat sources; />The heat dissipated to the outside at the time t by the heat supply system; />Is the total heating demand at time t;
power constraint:；/>；/>；/>for maximum heating power of solar heat source at one moment, < >>The heat supply power at the time t of the solar heat source device is provided; />For maximum heating power of the waste heat source at one moment,/->The heat supply power at the time t of the waste heat source device is used for supplying heat; />For maximum heating power of the natural gas heat source at one moment, < >>The heating power at the t moment of the natural gas heat source device.
4. The big data based energy management system of claim 1, wherein the generation policy matrix is generated by a historical energy management policy.
5. The big data based energy management system of claim 1, wherein the method of generating the first energy management policy based on the initial energy management policy matrix comprises:
step 101, randomly selecting two initial energy management strategy matrixes, namely a first initial energy management strategy matrix and a second initial energy management strategy matrix;
102, randomly selecting an element of column A from the first initial energy management strategy matrix to replace the element of a corresponding column of the second initial energy management strategy matrix to obtain a third initial energy management strategy matrix;
step 103, randomly selecting B rows of elements from the third initial energy management strategy matrix to replace the elements of the corresponding rows of the second initial energy management strategy matrix to obtain a fourth initial energy management strategy matrix;
step 104, iteratively executing the steps 101-103 until M fourth initial energy management strategy matrixes are obtained; and then selecting N fourth initial energy management strategy matrixes meeting the constraint conditions of the energy management strategies as the first energy management strategy matrixes.
6. The big data based energy management system of claim 1, wherein the energy optimization evaluation model is a neural network.
7. The big data based energy management system of claim 6, wherein m columns of LSTM cells, each column comprising n LSTM cells, the LSTM cells of the ith row and the nth column inputting elements of the ith row and the nth column of the second energy management policy matrix;
LSTM cell output of mth columnInputting a classifier, wherein the classification space of the classifier is +.>The value range of the evaluation score +.>The mean value is discretized, and one classification of the classification space Q corresponds to one score after the mean value is discretized.
8. The big data based energy management system of claim 1, wherein the LSTM unit of the ith row and the nth column operates as follows:
forgetting doorThe calculation formula of (2) is as follows: />Wherein->The state of the front-end is indicated,；/>representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; />Representation->Transfer to->The corresponding weight matrix is used to determine the weight matrix,representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing a bias term;
input doorThe calculation formula of (2) is as follows: />Wherein->The state of the front-end is indicated,；/>representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; />Representation input +.>Transfer to->Corresponding weight matrix, < >>Representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing a bias term;
intermediate stateCan be expressed as follows: />Wherein->Indicating the preposition status,/->；/>Representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; />Representation input +.>Transfer to->Corresponding weight matrix, < >>Representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing a bias term;
output stateThe calculation formula of (2) is as follows: />Wherein->Is the output state transferred by LSTM of row i, column t-1, +.>、/>、/>Is the calculation result of the forget gate, the input gate and the intermediate state; />Door for indicating forgetfulnessAnd the output state of LSTM of row i, column t-1 +.>Point-by-point multiplication is performed to make->；/>Representing I/O gate->And intermediate state->Point-wise multiplication is performed, ">；
Output doorThe calculation formula of (2) is as follows: />Wherein->The state of the front-end is indicated,；/>representing the output of the i-1 th row and t th column LSTM cell, < >>Representing the output of the ith row, t-1 th column LSTM cell; />An input representing an LSTM cell of an ith row and a nth column; wherein->Representation input +.>Transfer to->Corresponding weight matrix, < >>Representing the preposition status +.>Transfer to->Corresponding weight matrix, < >>Representing a bias term;
output ofThe calculation formula of (2) is as follows: />Will output door->And->Multiplying point by point to obtain the output of LSTM unit of ith row and t column>；
Definition of the definition、/>、/>All assigned a value of 0.
9. The big data based energy management system of claim 1, further comprising: a policy generation module that generates management policy information based on the final energy management policy matrix;
the management policy information includes: m times, each time corresponds to a total power meter and an output power meter, and each unit cell of the total power meter comprises an ID of a heat source device and the total heat supply power of the heat source device;
the unit cell of the first row and the j-th column of the output power meter comprises the heat supply power provided by the heat source device with the ID of j to the heat consumption unit with the ID of l.
10. The big data based energy management system of claim 1, wherein the management information of the heating system is obtained directly based on the information of the final energy management policy matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310868139.7A CN116595896B (en) | 2023-07-17 | 2023-07-17 | Big data-based energy management system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310868139.7A CN116595896B (en) | 2023-07-17 | 2023-07-17 | Big data-based energy management system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116595896A true CN116595896A (en) | 2023-08-15 |
| CN116595896B CN116595896B (en) | 2023-09-22 |
Family
ID=87601241
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310868139.7A Active CN116595896B (en) | 2023-07-17 | 2023-07-17 | Big data-based energy management system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116595896B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120071102A1 (en) * | 2010-09-16 | 2012-03-22 | The Hong Kong University Of Science And Technology | Multiple-input, multiple-output cognitive radio |
| US20150248118A1 (en) * | 2014-02-26 | 2015-09-03 | Board Of Trustees Of The University Of Alabama | Systems and methods for modeling energy consumption and creating demand response strategies using learning-based approaches |
| CN110225075A (en) * | 2019-03-25 | 2019-09-10 | 北京快电科技有限公司 | A kind of building energy internet wisdom operation cloud operating system |
| CN111245025A (en) * | 2020-02-04 | 2020-06-05 | 国网河北省电力有限公司经济技术研究院 | Optimization method, terminal device and storage medium for comprehensive energy system operation strategy |
| US20200242400A1 (en) * | 2019-01-25 | 2020-07-30 | Oath Inc. | Systems and methods for hyper parameter optimization for improved machine learning ensembles |
| CN114676920A (en) * | 2022-03-30 | 2022-06-28 | 天津津电供电设计所有限公司 | Electric heating comprehensive energy system optimized operation method considering external support capacity |
| US20230177235A1 (en) * | 2021-12-02 | 2023-06-08 | Peking University | Energy flow optimization in multi-energy system based on spatiotemporal network flows |
| CN116313127A (en) * | 2023-03-23 | 2023-06-23 | 珠海市安克电子技术有限公司 | Decision support system based on pre-hospital first-aid big data |
-
2023
- 2023-07-17 CN CN202310868139.7A patent/CN116595896B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120071102A1 (en) * | 2010-09-16 | 2012-03-22 | The Hong Kong University Of Science And Technology | Multiple-input, multiple-output cognitive radio |
| US20150248118A1 (en) * | 2014-02-26 | 2015-09-03 | Board Of Trustees Of The University Of Alabama | Systems and methods for modeling energy consumption and creating demand response strategies using learning-based approaches |
| US20200242400A1 (en) * | 2019-01-25 | 2020-07-30 | Oath Inc. | Systems and methods for hyper parameter optimization for improved machine learning ensembles |
| CN110225075A (en) * | 2019-03-25 | 2019-09-10 | 北京快电科技有限公司 | A kind of building energy internet wisdom operation cloud operating system |
| CN111245025A (en) * | 2020-02-04 | 2020-06-05 | 国网河北省电力有限公司经济技术研究院 | Optimization method, terminal device and storage medium for comprehensive energy system operation strategy |
| US20230177235A1 (en) * | 2021-12-02 | 2023-06-08 | Peking University | Energy flow optimization in multi-energy system based on spatiotemporal network flows |
| CN114676920A (en) * | 2022-03-30 | 2022-06-28 | 天津津电供电设计所有限公司 | Electric heating comprehensive energy system optimized operation method considering external support capacity |
| CN116313127A (en) * | 2023-03-23 | 2023-06-23 | 珠海市安克电子技术有限公司 | Decision support system based on pre-hospital first-aid big data |
Non-Patent Citations (2)
| Title |
|---|
| 宋宏升;: "大型办公区分布式多能耦合能源站运行策略优化", 分布式能源, no. 01, pages 54 - 61 * |
| 赵青;: "基于遗传算法的智能电网需求侧管理", 通信电源技术, no. 02, pages 205 - 207205 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116595896B (en) | 2023-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Qin et al. | Macroscopic–microscopic attention in LSTM networks based on fusion features for gear remaining life prediction | |
| Assaf et al. | Explainable deep neural networks for multivariate time series predictions. | |
| Xiong et al. | Short-term wind power forecasting based on attention mechanism and deep learning | |
| Adeli et al. | A probabilistic neural network for earthquake magnitude prediction | |
| CN110738344B (en) | Distributed reactive power optimization method and device for power system load forecasting | |
| Idrissi et al. | Genetic algorithm for neural network architecture optimization | |
| CN110889603A (en) | Power system economic dispatching method considering wind power correlation based on PCA-Copula theory | |
| Tang et al. | Wind power forecasting: A hybrid forecasting model and multi-task learning-based framework | |
| Bhaumik et al. | Hidden Markov models for wind farm power output | |
| CN113313306A (en) | Elastic neural network load prediction method based on improved wolf optimization algorithm | |
| CN114091615A (en) | A method and system for electric energy metering data completion based on generative adversarial network | |
| Boybat et al. | Improved deep neural network hardware-accelerators based on non-volatile-memory: The local gains technique | |
| Sivanand et al. | A comparative study of different deep learning models for mid-term solar power prediction | |
| CN116595896B (en) | Big data-based energy management system | |
| CN116523001A (en) | Method, device and computer equipment for constructing weak line identification model of power grid | |
| CN116090635A (en) | A Weather-Driven New Energy Power Generation Power Forecasting Method | |
| Ansari et al. | Sequential combination of statistics, econometrics and Adaptive Neural-Fuzzy Interface for stock market prediction | |
| Prema et al. | LSTM based Deep Learning model for accurate wind speed prediction | |
| CN113705086A (en) | Ultra-short-term wind power prediction method based on Elman error correction | |
| CN116706907B (en) | Photovoltaic power generation prediction method based on fuzzy reasoning and related equipment | |
| Rafik et al. | Learning and Predictive Energy Consumption Model based on LSTM recursive neural networks | |
| Chung et al. | An intelligent control strategy for energy storage systems in solar power generation based on long-short-term power prediction | |
| CN116402134A (en) | Knowledge tracking method and system based on behavior perception | |
| CN116205148A (en) | Data center cooling load prediction method based on Wide & Deep model | |
| Ahmed et al. | Echo state network optimization using hybrid-structure based gravitational search algorithm with square quadratic programming for time series prediction. |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |