CN110490385A - The unified prediction of electric load and thermic load in a kind of integrated energy system - Google Patents
The unified prediction of electric load and thermic load in a kind of integrated energy system Download PDFInfo
- Publication number
- CN110490385A CN110490385A CN201910778722.2A CN201910778722A CN110490385A CN 110490385 A CN110490385 A CN 110490385A CN 201910778722 A CN201910778722 A CN 201910778722A CN 110490385 A CN110490385 A CN 110490385A
- Authority
- CN
- China
- Prior art keywords
- data
- load
- network model
- neural network
- training
- 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
- 238000012549 training Methods 0.000 claims abstract description 27
- 238000003062 neural network model Methods 0.000 claims abstract description 19
- 230000005855 radiation Effects 0.000 claims abstract description 16
- 238000012360 testing method Methods 0.000 claims abstract description 12
- 238000013135 deep learning Methods 0.000 claims abstract description 6
- 238000010606 normalization Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 28
- 230000006870 function Effects 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 230000004913 activation Effects 0.000 claims description 6
- 230000001186 cumulative effect Effects 0.000 claims description 6
- 210000002569 neuron Anatomy 0.000 claims description 6
- 230000009467 reduction Effects 0.000 claims description 4
- 238000010438 heat treatment Methods 0.000 claims description 3
- 238000012847 principal component analysis method Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 3
- 238000011425 standardization method Methods 0.000 claims description 3
- 238000001816 cooling Methods 0.000 claims 1
- 238000000513 principal component analysis Methods 0.000 abstract description 4
- 238000004393 prognosis Methods 0.000 abstract 1
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 241001123248 Arma Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005485 electric heating Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000013277 forecasting method Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Strategic Management (AREA)
- Evolutionary Computation (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Entrepreneurship & Innovation (AREA)
Abstract
The embodiment of the invention discloses the unified predictions of electric load and thermic load in a kind of integrated energy system, first collect integrated energy system electric load, thermic load historical data；And corresponding solar radiation, wind-force, temperature, humidity, barometric history data；It tests and normalizes to historical data, obtain sample data；The important meteorologic factor for influencing electric load and thermic load is chosen using Principal Component Analysis；Sample data after normalization is divided into training data and test data, training sample and training network model are inputted into LSTM deep neural network model；LSTM deep neural network model is trained in TensorFlow deep learning frame, saves neural network model weight；The LSTM deep neural network model saved is loaded, prognosis modelling, the electric load power and thermic load power predicted are carried out to test data；The prediction result of electric load and thermic load is evaluated using mean absolute error percentage and the average value of root-mean-square error.
Description
Technical Field
The embodiment of the invention relates to the technical field of comprehensive energy systems, in particular to a combined prediction method for electric load and thermal load in a comprehensive energy system.
Background
Under the current large background of energy conservation and emission reduction and rapid growth of renewable energy sources, the construction of a clean, low-carbon and environment-friendly comprehensive energy system is improved, and the system is an important energy utilization mode in the energy transformation process. The comprehensive energy system is an energy comprehensive network with a plurality of energy sources in interactive coupling, and is an important form for the development of the energy field at present.
Therefore, accurate energy demand prediction will become an important part in the economic dispatching and optimized operation of the integrated energy system. The technical level at present can accurately predict the power load, but the prediction of the heat load is rough, so that the prediction requirements of the whole electricity and heat utilization of the comprehensive energy system are difficult to meet.
On one hand, the electric load and the heat load in the comprehensive energy system often have a certain coupling relationship and can be mutually converted and mutually influenced, for example, household heating can adopt a heat supply form and can also be converted into an electric load through an electric heating form; on the other hand, according to the prediction results of the electric load and the thermal load, the consumption of the renewable energy power generation and the adjustment of the operation mode of the cogeneration unit can be matched, and the peak regulation capacity and the operation economy of the comprehensive energy system are improved.
Currently, common electric load prediction methods include an ARMA time series prediction method, a gray prediction method, a BP neural network method and the like, and the prediction methods for the heat load are few. The heating system is a complex system with time lag, nonlinearity and uncertainty, and has high coupling with the electrical load in the comprehensive energy system, so that the accurate prediction is difficult. Meanwhile, the electrical load and the thermal load are sensitive to weather conditions such as solar radiation, wind power, temperature and the like, and have strong relevance, and the electrical load and the thermal load in the comprehensive energy system are difficult to accurately predict by using the traditional method.
The time series method is only suitable for short-term prediction and requires uniform changes of electrical load and thermal load, and cannot timely react to sudden and abnormal conditions, such as suddenly changing weather and the like; although the gray prediction method has strong adaptability, the error of the medium-short term heat load prediction is large, and the method is suitable for the long-term heat load prediction; the single BP neural network is easy to fall into the defect of local minimum, and the prediction effect precision is not high.
Disclosure of Invention
Therefore, the embodiment of the invention provides a combined prediction method of an electric load and a heat load in an integrated energy system, so as to solve the problems in the prior art.
In order to achieve the above object, an embodiment of the present invention provides the following:
in a first aspect of the embodiments of the present invention, there is provided a method for joint prediction of electrical load and thermal load in an integrated energy system, including the steps of:
step 100, collecting historical data of electric load and heat load of the comprehensive energy system and corresponding historical data of solar radiation, wind power, temperature, humidity and air pressure;
step 200, checking and normalizing the collected historical data to obtain sample data;
300, selecting important meteorological factors influencing the electric load and the heat load by adopting a principal component analysis method;
step 400, dividing the normalized sample data into training data and test data, inputting the training sample into the LSTM deep neural network model and training the network model;
500, training the LSTM deep neural network model in a TensorFlow deep learning framework, and when the error of the loss function of the LSTM deep neural network model is smaller than a set value epsilon2Or when the iteration times reach the training maximum times, stopping training and storing the weight of the neural network model.
And step 600, loading the stored LSTM deep neural network model, and performing prediction simulation on the test data to obtain predicted electric load power and thermal load power.
And 700, evaluating the prediction results of the electrical load and the thermal load by adopting the average absolute error percentage and the average value of the root mean square error.
Preferably, the historical data of the electrical load and the thermal load collected in the step 100 is data with a time resolution of 15 minutes in at least one year or more in the integrated energy system.
Preferably, the step 400 further includes: inputting the historical data of the electrical load and the thermal load of the integrated energy system and the historical data of the corresponding solar radiation, wind power, temperature, humidity and air pressure in the step 100 as sample data, wherein the sample data comprises the electrical load data PX (t), the thermal load data HX (t), and the meteorological data X (t) (x ═ x (t)), (x:)1(t),x2(t),...,xn(T)) (T ═ 1, 2, 3, …, T, n ═ 5), T denotes the width of the data point, x1(t) to x5(t) in turn represents solar radiation, wind, temperature, humidity and barometric pressure at time t.
Preferably, the step 200 specifically includes: removing the numerical points of which the electrical load and the thermal load data are negative numbers and are continuously 0, removing more than 8 continuous unchanged data points, and simultaneously removing the data points of which the meteorological data correspond to the loss of the electrical load and the thermal load; adopting a minimum and maximum value standardization method to carry out normalization processing on electric load, heat load, solar radiation, wind power, temperature, humidity and air pressure data:
wherein,xiis the actual value of the data, ximinIs the minimum value of data, ximaxIs the maximum value of the data, x* iIs a normalized standard value.
Preferably, the step 300 specifically includes:
reducing the dimension of n-dimensional data of an input sample of acquired sample data, selecting m principal components (m is less than n) of the n-dimensional data, wherein data information contained in each principal component is mainly reflected on covariance which is shown as a formula (2); judging the value of m by using the accumulated covariance contribution rate, as shown in the formula (3) and the formula (4); first m eigenvalues μ1,μ2…μmCorresponding feature vector Z1,Z2,...,ZmThe vector is used as a principal component vector after dimensionality reduction and is used as the input of m meteorological principal components of the next LSTM neural network;
cov(Xi，PX)＝μiZi (2)
in the formula, muiFor meteorological data sample XiCharacteristic value of covariance with electric load data sample PX, mu1≥μ2≥...≥μn，λiIs the variance contribution ratio, λ∑(m) is the cumulative variance contribution, ε, of the first m principal components1And (5) 90%, namely selecting the m value with the cumulative variance contribution rate exceeding 90% as the selected main component value.
Preferably, the step 400 specifically includes:
the designed LSTM deep neural network comprises an input layer, a hidden layer and an output layer, wherein a sigmoid activation function is adopted as the activation function, as shown in a formula (5), and a loss function is adopted as a mean square error, as shown in a formula (6);
in the formula: y iskIs the actual value, y ', of the kth data in a batch of data batch'kFor the predicted value, l is the number of data in one batch.
Preferably, the hidden layer has a two-layer hidden layer structure, the input vector includes m meteorological principal components, and the electrical load and the thermal load at t-2, t-1 and t time, that is, the number of nodes of the input layer is m +3+3, the number of time steps is 16, the first hidden layer includes 32 neurons, the second hidden layer includes 64 neurons, and the number of nodes of the output layer is 2, that is, the output of the network is the predicted electrical load and the thermal load at t +1 time.
Preferably, in training the network model, the parameters are set as follows:
the maximum number of training times is set to 10000 times, the learning rate is 0.02, the batch size is 60, and dropout is set to 0.5;
the predicted time period is 24 hours into the future and the time resolution is 15 min.
Preferably, the predicted electrical load power is obtained according to said step 600And thermal load powerStep 700 is performed as shown in formula (7) and formula (8):
wherein, PEL(t)，PHL(t)，(T ═ 1, 2, 3, …, T) represents the actual electrical load and the actual thermal load at time T, respectively, the predicted electrical load and the predicted thermal load, and N represents the length of the test data.
The embodiment of the invention has the following advantages:
the invention considers the influence of meteorological factors such as solar radiation, wind power, temperature and the like on the electrical load and the thermal load in the prediction model, adopts the LSTM deep learning theory, establishes the dynamic association relation of the meteorological data, the electrical load and the thermal load, and can accurately predict the electrical load and the thermal load in the comprehensive energy system.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It should be apparent that the drawings in the following description are merely exemplary, and that other embodiments can be derived from the drawings provided by those of ordinary skill in the art without inventive effort.
FIG. 1 is a flow chart of a method according to an embodiment of the present invention.
Detailed Description
The present invention is described in terms of particular embodiments, other advantages and features of the invention will become apparent to those skilled in the art from the following disclosure, and it is to be understood that the described embodiments are merely exemplary of the invention and that it is not intended to limit the invention to the particular embodiments disclosed. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
A combined prediction method for electric load and heat load in an integrated energy system comprises the following steps:
step 100, collecting historical data of electric load and heat load of the comprehensive energy system and corresponding historical data of solar radiation, wind power, temperature, humidity and air pressure;
the collected historical data of the electrical load and the thermal load is data with time resolution of 15 minutes in at least one year in the comprehensive energy system.
All historical data are input as samples, including electrical load data px (t), thermal load data hx (t), and meteorological data x (t) (x)1(t),x2(t),...,xn(T)) (T ═ 1, 2, 3, …, T, n ═ 5), T denotes the width of the data point, x1(t) to x5(t) in turn represents solar radiation, wind, temperature, humidity and barometric pressure at time t.
Step 200, checking and normalizing the collected historical data to obtain sample data;
removing the numerical points of which the electrical load and the thermal load data are negative numbers and are continuously 0, removing more than 8 continuous unchanged data points, and simultaneously removing the data points of which the meteorological data correspond to the loss of the electrical load and the thermal load; adopting a minimum and maximum value standardization method to carry out normalization processing on electric load, heat load, solar radiation, wind power, temperature, humidity and air pressure data:
wherein x isiIs the actual value of the data, ximin is the minimum value of the data, ximaxIs the maximum value of the data, x* iIs a normalized standard value.
300, selecting important meteorological factors influencing the electric load and the heat load by adopting a principal component analysis method;
reducing the dimension of n-dimensional data of an input sample of acquired sample data, selecting m principal components (m is less than n) of the n-dimensional data, wherein data information contained in each principal component is mainly reflected on covariance which is shown as a formula (2); judging the value of m by using the accumulated covariance contribution rate, as shown in the formula (3) and the formula (4); first m eigenvalues μ1,μ2...μmCorresponding feature vector Z1,Z2,...,ZmThe vector is used as a principal component vector after dimensionality reduction and is used as the input of m meteorological principal components of the next LSTM neural network;
cov(Xi，PX)＝μiZi (2)
in the formula, muiFor meteorological data sample XiCharacteristic value of covariance with electric load data sample PX, mu1≥μ2≥...≥μn，λiIs the variance contribution ratio, λ∑(m) is the cumulative variance contribution, ε, of the first m principal components1And (5) 90%, namely selecting the m value with the cumulative variance contribution rate exceeding 90% as the selected main component value.
Step 400, dividing the normalized sample data into training data and test data, inputting the training sample into the LSTM deep neural network model and training the network model;
in the step, historical data of the electrical load and the thermal load of the comprehensive energy system in the step 100 and corresponding historical data of solar radiation, wind power, temperature, humidity and air pressure are used as sample data to be input; the designed LSTM deep neural network comprises an input layer, a hidden layer and an output layer, wherein a sigmoid activation function is adopted as the activation function, as shown in a formula (5), and a loss function is adopted as a mean square error, as shown in a formula (6);
in the formula: y iskIs the actual value, y ', of the kth data in a batch of data batch'kFor the predicted value, l is the number of data in one batch.
The hidden layer has a two-layer hidden layer structure, an input vector comprises m meteorological principal components, and electric loads and heat loads at t-2, t-1 and t moments, namely the number of nodes of the input layer is m +3+3, the number of time steps is 16, the first hidden layer comprises 32 neurons, the second hidden layer comprises 64 neurons, the number of nodes of the output layer is 2, namely the output of the network is the predicted electric load and heat load at the t +1 moment
500, training the LSTM deep neural network model in a TensorFlow deep learning framework, and when the error of the loss function of the LSTM deep neural network model is smaller than a set value epsilon2Or when the iteration times reach the training maximum times, stopping training and storing the weight of the neural network model;
in training the network model, the parameters are set as follows:
the maximum number of training times is set to 10000 times, the learning rate is 0.02, the batch size is 60, and dropout is set to 0.5;
the predicted time period is 24 hours into the future and the time resolution is 15 min.
And step 600, loading the stored LSTM deep neural network model, and performing prediction simulation on the test data to obtain predicted electric load power and thermal load power.
And 700, evaluating the prediction results of the electrical load and the thermal load by adopting the average absolute error percentage and the average value of the root mean square error.
Step 600 obtains predicted electrical load powerAnd thermal load powerStep 700 is performed as shown in formula (7) and formula (8):
wherein, PEL(t)，PHL(t)，(T ═ 1, 2, 3, …, T) represents the actual electrical load and the actual thermal load at time T, respectively, the predicted electrical load and the predicted thermal load, and N represents the length of the test data.
The invention is essentially a combined forecasting method of the electrical load and the thermal load of a comprehensive energy system based on the combination of Principal Component Analysis (PCA) and long-short term memory cycle neural network (LSTM), the influence of meteorological factors such as solar radiation, wind power, temperature and the like on the electrical load and the thermal load is considered in a forecasting model, the PCA is adopted in the model to reduce the dimension of characteristics and data, the LSTM deep learning theory is adopted, the dynamic association relation of the meteorological data, the electrical load and the thermal load is established, and the electrical load and the thermal load in the comprehensive energy system can be accurately forecasted.
Although the invention has been described in detail above with reference to a general description and specific examples, it will be apparent to one skilled in the art that modifications or improvements may be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.
Claims (9)
1. A combined prediction method for electric load and heat load in an integrated energy system is characterized by comprising the following steps:
step 100, collecting historical data of electric load and heat load of the comprehensive energy system and corresponding historical data of solar radiation, wind power, temperature, humidity and air pressure;
step 200, checking and normalizing the collected historical data to obtain sample data;
300, selecting important meteorological factors influencing the electric load and the heat load by adopting a principal component analysis method;
step 400, dividing the normalized sample data into training data and test data, inputting the training sample into the LSTM deep neural network model and training the network model;
500, training the LSTM deep neural network model in a TensorFlow deep learning framework, and when the error of the loss function of the LSTM deep neural network model is smaller than a set value epsilon2Or when the iteration times reach the training maximum times, stopping training and storing the weight of the neural network model;
step 600, loading a stored LSTM deep neural network model, and performing prediction simulation on test data to obtain predicted electric load power and thermal load power;
and 700, evaluating the prediction results of the electrical load and the thermal load by adopting the average absolute error percentage and the average value of the root mean square error.
2. The method according to claim 1, wherein the historical data of electrical load and thermal load collected in step 100 is data of 15 minutes time resolution for at least one year in the integrated energy system.
3. The method of claim 2, wherein the step 400 further comprises: inputting the historical data of the electrical load and the thermal load of the integrated energy system and the historical data of the corresponding solar radiation, wind power, temperature, humidity and air pressure in the step 100 as sample data, wherein the sample data comprises the electrical load data PX (t), the thermal load data HX (t), and the meteorological data X (t) (x ═ x (t)), (x:)1(t),x2(t),...,xn(T)) (T ═ 1, 2, 3, …, T, n ═ 5), T denotes the width of the data point, x1(t) to x5(t) in turn represents solar radiation, wind, temperature, humidity and barometric pressure at time t.
4. The method according to claim 3, wherein the step 200 is specifically as follows: removing the numerical points of which the electrical load and the thermal load data are negative numbers and are continuously 0, removing more than 8 continuous unchanged data points, and simultaneously removing the data points of which the meteorological data correspond to the loss of the electrical load and the thermal load; adopting a minimum and maximum value standardization method to carry out normalization processing on electric load, heat load, solar radiation, wind power, temperature, humidity and air pressure data:
wherein x isiIs the actual value of the data, ximinIs the minimum value of data, ximaxIs the maximum value of the data, x* iIs a normalized standard value.
5. The method for calculating the power flow of the grid-connected operation cooling, heating and power comprehensive energy system according to claim 4, wherein the step 300 specifically comprises:
reducing the dimension of n-dimensional data of an input sample of acquired sample data, selecting m principal components (m is less than n) of the n-dimensional data, wherein data information contained in each principal component is mainly reflected on covariance which is shown as a formula (2); judging the value of m by using the accumulated covariance contribution rate, as shown in the formula (3) and the formula (4); first m eigenvalues μ1,μ2…μmCorresponding feature vector Z1,Z2,...,ZmThe vector is used as a principal component vector after dimensionality reduction and is used as the input of m meteorological principal components of the next LSTM neural network;
cov(Xi，PX)＝μiZi (2)
in the formula, muiFor meteorological data sample XiCharacteristic value of covariance with electric load data sample PX, mu1≥μ2≥...≥μn，λiIs the variance contribution ratio, λ∑(m) is the cumulative variance contribution, ε, of the first m principal components1And (5) 90%, namely selecting the m value with the cumulative variance contribution rate exceeding 90% as the selected main component value.
6. The method according to claim 5, wherein the step 400 is specifically:
the designed LSTM deep neural network comprises an input layer, a hidden layer and an output layer, wherein a sigmoid activation function is adopted as the activation function, as shown in a formula (5), and a loss function is adopted as a mean square error, as shown in a formula (6);
in the formula: y iskIs the actual value, y ', of the kth data in a batch of data batch'kFor the predicted value, l is the number of data in one batch.
7. The method of claim 6, wherein the method comprises the steps of:
the hidden layer has a two-layer hidden layer structure, an input vector comprises m meteorological principal components, and electric loads and heat loads at t-2, t-1 and t moments, namely the number of nodes of the input layer is m +3+3, the number of time steps is 16, the first hidden layer comprises 32 neurons, the second hidden layer comprises 64 neurons, and the number of nodes of the output layer is 2, namely the output of the network is the predicted electric load and heat load at the t +1 moment.
8. The method of claim 7, wherein the method comprises the steps of:
in training the network model, the parameters are set as follows:
the maximum number of training times is set to 10000 times, the learning rate is 0.02, the batch size is 60, and dropout is set to 0.5;
the predicted time period is 24 hours into the future and the time resolution is 15 min.
9. The method of claim 7, wherein the method comprises the steps of: obtaining a predicted electrical load power from said step 600And thermal load powerStep 700 is performed as shown in formula (7) and formula (8):
wherein, PEL(t)，PHL(t)，The actual electrical load and the actual thermal load at the t-th time point, the predicted electrical load and the predicted thermal load, respectively, and N represents the length of the test data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910778722.2A CN110490385A (en) | 2019-08-22 | 2019-08-22 | The unified prediction of electric load and thermic load in a kind of integrated energy system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910778722.2A CN110490385A (en) | 2019-08-22 | 2019-08-22 | The unified prediction of electric load and thermic load in a kind of integrated energy system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110490385A true CN110490385A (en) | 2019-11-22 |
Family
ID=68551707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910778722.2A Pending CN110490385A (en) | 2019-08-22 | 2019-08-22 | The unified prediction of electric load and thermic load in a kind of integrated energy system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110490385A (en) |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111191856A (en) * | 2020-02-17 | 2020-05-22 | 天津大学 | Regional comprehensive energy system multi-energy load prediction method considering time sequence dynamic characteristics and coupling characteristics |
| CN111561734A (en) * | 2020-05-18 | 2020-08-21 | 瑞纳智能设备股份有限公司 | Heating household valve adjusting method, system and equipment based on artificial intelligence |
| CN111932015A (en) * | 2020-08-12 | 2020-11-13 | 武汉中电节能有限公司 | Method and system for predicting cold and heat load of regional cooling and heating |
| CN112116153A (en) * | 2020-09-18 | 2020-12-22 | 上海电力大学 | Park multivariate load joint prediction method for coupling Copula and stacked LSTM network |
| CN112365056A (en) * | 2020-11-12 | 2021-02-12 | 云南电网有限责任公司 | Electrical load joint prediction method and device, terminal and storage medium |
| CN112747477A (en) * | 2021-01-25 | 2021-05-04 | 华南理工大学 | Intelligent control system and method for safe and energy-saving gas-fired hot water boiler |
| CN113065680A (en) * | 2020-01-02 | 2021-07-02 | 中国电力科学研究院有限公司 | Energy demand prediction method and system for energy Internet |
| CN113112094A (en) * | 2021-05-10 | 2021-07-13 | 瑞纳智能设备股份有限公司 | Thermal load prediction method, system, device and storage medium for quantifying wind condition influence |
| CN113139605A (en) * | 2021-04-27 | 2021-07-20 | 武汉理工大学 | Power load prediction method based on principal component analysis and LSTM neural network |
| CN113222231A (en) * | 2021-04-30 | 2021-08-06 | 洛阳双瑞精铸钛业有限公司 | Wisdom heating system based on internet of things |
| CN113283774A (en) * | 2021-06-07 | 2021-08-20 | 润电能源科学技术有限公司 | Deep peak regulation method and device for heating unit, electronic equipment and storage medium |
| CN113515898A (en) * | 2021-08-09 | 2021-10-19 | 国网湖北省电力有限公司电力科学研究院 | Refined monitoring and modeling method for comprehensive energy utilization network of' electricity, cold and hot gas |
| CN113657660A (en) * | 2021-08-12 | 2021-11-16 | 杭州英集动力科技有限公司 | Heat source load prediction method based on substation load and heat supply network hysteresis model |
| CN113988393A (en) * | 2021-10-21 | 2022-01-28 | 青岛联众芯云科技有限公司 | Energy internet load prediction method |
| CN114021848A (en) * | 2021-11-24 | 2022-02-08 | 西安热工研究院有限公司 | Generating capacity demand prediction method based on LSTM deep learning |
| CN114118571A (en) * | 2021-11-24 | 2022-03-01 | 西安西热电站信息技术有限公司 | Heating heat load prediction method based on LSTM deep learning |
| CN114676941A (en) * | 2022-05-30 | 2022-06-28 | 国网天津市电力公司经济技术研究院 | Electric-thermal load combined self-adaptive prediction method and device for park comprehensive energy system |
| CN114676979A (en) * | 2022-03-04 | 2022-06-28 | 南方科技大学 | Energy scheduling method and device, computer equipment and storage medium |
| CN115034424A (en) * | 2021-03-05 | 2022-09-09 | 气体产品与化学公司 | Method and apparatus for managing predicted power resources of an industrial gas plant complex |
| CN115310888A (en) * | 2022-10-13 | 2022-11-08 | 国网天津市电力公司城东供电分公司 | Comprehensive energy user energy consumption behavior correlation analysis method based on multi-element data processing |
| CN116187601A (en) * | 2023-05-04 | 2023-05-30 | 华北理工大学 | Comprehensive energy system operation optimization method based on load prediction |
| CN117078047A (en) * | 2023-10-16 | 2023-11-17 | 华能济南黄台发电有限公司 | LSTM-based heat load prediction and distribution optimization method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104505824A (en) * | 2014-12-12 | 2015-04-08 | 国家电网公司 | Method and device for making electricity-generating load plan of thermal power plant under running mode of ordering power by heat |
| CN109740826A (en) * | 2019-01-30 | 2019-05-10 | 广东工业大学 | A kind of cooling heating and power generation system load forecasting method based on Dynamic Data Mining |
-
2019
- 2019-08-22 CN CN201910778722.2A patent/CN110490385A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104505824A (en) * | 2014-12-12 | 2015-04-08 | 国家电网公司 | Method and device for making electricity-generating load plan of thermal power plant under running mode of ordering power by heat |
| CN109740826A (en) * | 2019-01-30 | 2019-05-10 | 广东工业大学 | A kind of cooling heating and power generation system load forecasting method based on Dynamic Data Mining |
Non-Patent Citations (1)
| Title |
|---|
| 谢明磊: ""基于LSTM网络的住宅负荷短期预测"", 《广东电力》 * |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065680A (en) * | 2020-01-02 | 2021-07-02 | 中国电力科学研究院有限公司 | Energy demand prediction method and system for energy Internet |
| CN113065680B (en) * | 2020-01-02 | 2024-09-24 | 中国电力科学研究院有限公司 | Energy demand prediction method and system for energy Internet |
| CN111191856A (en) * | 2020-02-17 | 2020-05-22 | 天津大学 | Regional comprehensive energy system multi-energy load prediction method considering time sequence dynamic characteristics and coupling characteristics |
| CN111191856B (en) * | 2020-02-17 | 2023-12-08 | 天津大学 | Regional comprehensive energy system multi-energy load prediction method considering time sequence dynamic characteristics and coupling characteristics |
| CN111561734A (en) * | 2020-05-18 | 2020-08-21 | 瑞纳智能设备股份有限公司 | Heating household valve adjusting method, system and equipment based on artificial intelligence |
| CN111932015A (en) * | 2020-08-12 | 2020-11-13 | 武汉中电节能有限公司 | Method and system for predicting cold and heat load of regional cooling and heating |
| CN112116153A (en) * | 2020-09-18 | 2020-12-22 | 上海电力大学 | Park multivariate load joint prediction method for coupling Copula and stacked LSTM network |
| CN112116153B (en) * | 2020-09-18 | 2022-10-04 | 上海电力大学 | Park multivariate load joint prediction method coupling Copula and stacked LSTM network |
| CN112365056A (en) * | 2020-11-12 | 2021-02-12 | 云南电网有限责任公司 | Electrical load joint prediction method and device, terminal and storage medium |
| CN112747477A (en) * | 2021-01-25 | 2021-05-04 | 华南理工大学 | Intelligent control system and method for safe and energy-saving gas-fired hot water boiler |
| CN115034424A (en) * | 2021-03-05 | 2022-09-09 | 气体产品与化学公司 | Method and apparatus for managing predicted power resources of an industrial gas plant complex |
| CN113139605A (en) * | 2021-04-27 | 2021-07-20 | 武汉理工大学 | Power load prediction method based on principal component analysis and LSTM neural network |
| CN113222231A (en) * | 2021-04-30 | 2021-08-06 | 洛阳双瑞精铸钛业有限公司 | Wisdom heating system based on internet of things |
| CN113112094B (en) * | 2021-05-10 | 2024-02-13 | 瑞纳智能设备股份有限公司 | Method, system, device and storage medium for predicting thermal load by quantifying influence of wind condition |
| CN113112094A (en) * | 2021-05-10 | 2021-07-13 | 瑞纳智能设备股份有限公司 | Thermal load prediction method, system, device and storage medium for quantifying wind condition influence |
| CN113283774A (en) * | 2021-06-07 | 2021-08-20 | 润电能源科学技术有限公司 | Deep peak regulation method and device for heating unit, electronic equipment and storage medium |
| CN113515898A (en) * | 2021-08-09 | 2021-10-19 | 国网湖北省电力有限公司电力科学研究院 | Refined monitoring and modeling method for comprehensive energy utilization network of' electricity, cold and hot gas |
| CN113657660A (en) * | 2021-08-12 | 2021-11-16 | 杭州英集动力科技有限公司 | Heat source load prediction method based on substation load and heat supply network hysteresis model |
| CN113657660B (en) * | 2021-08-12 | 2024-07-19 | 浙江英集动力科技有限公司 | Heat source load prediction method based on substation load and heat supply network hysteresis model |
| CN113988393A (en) * | 2021-10-21 | 2022-01-28 | 青岛联众芯云科技有限公司 | Energy internet load prediction method |
| CN114021848A (en) * | 2021-11-24 | 2022-02-08 | 西安热工研究院有限公司 | Generating capacity demand prediction method based on LSTM deep learning |
| CN114118571A (en) * | 2021-11-24 | 2022-03-01 | 西安西热电站信息技术有限公司 | Heating heat load prediction method based on LSTM deep learning |
| CN114676979A (en) * | 2022-03-04 | 2022-06-28 | 南方科技大学 | Energy scheduling method and device, computer equipment and storage medium |
| CN114676941B (en) * | 2022-05-30 | 2022-09-30 | 国网天津市电力公司经济技术研究院 | Electric-thermal load combined self-adaptive prediction method and device for park comprehensive energy system |
| CN114676941A (en) * | 2022-05-30 | 2022-06-28 | 国网天津市电力公司经济技术研究院 | Electric-thermal load combined self-adaptive prediction method and device for park comprehensive energy system |
| CN115310888A (en) * | 2022-10-13 | 2022-11-08 | 国网天津市电力公司城东供电分公司 | Comprehensive energy user energy consumption behavior correlation analysis method based on multi-element data processing |
| CN116187601A (en) * | 2023-05-04 | 2023-05-30 | 华北理工大学 | Comprehensive energy system operation optimization method based on load prediction |
| CN116187601B (en) * | 2023-05-04 | 2023-06-23 | 华北理工大学 | Comprehensive energy system operation optimization method based on load prediction |
| CN117078047A (en) * | 2023-10-16 | 2023-11-17 | 华能济南黄台发电有限公司 | LSTM-based heat load prediction and distribution optimization method and system |
| CN117078047B (en) * | 2023-10-16 | 2024-02-23 | 华能济南黄台发电有限公司 | LSTM-based heat load prediction and distribution optimization method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110490385A (en) | The unified prediction of electric load and thermic load in a kind of integrated energy system | |
| CN110909912B (en) | Park electric power system net load combination prediction method based on self-adaptive error feedback | |
| Wu et al. | A novel hybrid model for short‐term forecasting in PV power generation | |
| CN106779223B (en) | Photovoltaic system power generation real-time prediction method and device | |
| CN111008728A (en) | Method for predicting short-term output of distributed photovoltaic power generation system | |
| CN107766990A (en) | A kind of Forecasting Methodology of photovoltaic power station power generation power | |
| CN114792156B (en) | Photovoltaic output power prediction method and system based on curve characteristic index clustering | |
| CN111079989B (en) | DWT-PCA-LSTM-based water supply amount prediction device for water supply company | |
| CN111695736B (en) | Photovoltaic power generation short-term power prediction method based on multi-model fusion | |
| CN102102626A (en) | Method for forecasting short-term power in wind power station | |
| CN114970952B (en) | Photovoltaic output short-term prediction method and system considering environmental factors | |
| CN110380444B (en) | Capacity planning method for distributed wind power orderly access to power grid under multiple scenes based on variable structure Copula | |
| CN110766200A (en) | Method for predicting generating power of wind turbine generator based on K-means mean clustering | |
| CN110929953A (en) | Photovoltaic power station ultra-short term output prediction method based on cluster analysis | |
| CN110796307A (en) | Distributed load prediction method and system for comprehensive energy system | |
| Ku et al. | Building electric energy prediction modeling for BEMS using easily obtainable weather factors with Kriging model and data mining | |
| CN115545333A (en) | Method for predicting load curve of multi-load daily-type power distribution network | |
| CN110991747A (en) | Short-term load prediction method considering wind power plant power | |
| CN113887833A (en) | Distributed energy user side time-by-time load prediction method and system | |
| Chen et al. | Short interval solar power prediction for energy harvesting with low computation cost on edge computation network | |
| CN113112085A (en) | New energy station power generation load prediction method based on BP neural network | |
| CN110309978B (en) | Neural network photovoltaic power prediction model and method based on secondary dynamic adjustment | |
| CN110648011B (en) | Feeder line short-term load prediction method considering photovoltaic users | |
| CN116167508B (en) | Short-term photovoltaic output rapid prediction method and system based on meteorological factor decomposition | |
| CN116432830A (en) | Photovoltaic power prediction method, related computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191122 |
|
| RJ01 | Rejection of invention patent application after publication |