CN112508246A - Photovoltaic power generation power prediction method based on similar days - Google Patents
Photovoltaic power generation power prediction method based on similar days Download PDFInfo
- Publication number
- CN112508246A CN112508246A CN202011344728.8A CN202011344728A CN112508246A CN 112508246 A CN112508246 A CN 112508246A CN 202011344728 A CN202011344728 A CN 202011344728A CN 112508246 A CN112508246 A CN 112508246A
- Authority
- CN
- China
- Prior art keywords
- day
- similar
- days
- prediction
- irradiance
- 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
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000010248 power generation Methods 0.000 title claims abstract description 30
- 238000013528 artificial neural network Methods 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 10
- 230000001174 ascending effect Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- 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
-
- 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
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Marketing (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Water Supply & Treatment (AREA)
- Evolutionary Biology (AREA)
- Primary Health Care (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Photovoltaic Devices (AREA)
Abstract
The invention discloses a photovoltaic power generation power prediction method based on similar days, which depends on historical power generation data and historical meteorological data of a photovoltaic power station and adopts a combined prediction method to predict the photovoltaic power generation power. The method comprises the following specific steps: firstly, matching the predicted daily irradiance curve with the historical daily irradiance curve of the previous n days, and respectively calculating the dissimilarity of the predicted daily irradiance curve and the historical daily irradiance curve of the previous n days to obtain an m-day-history similar day and a historical most similar day (m < n). On one hand, training data of the m calendar history similar days by using a neural network algorithm to obtain a fitting prediction model, and substituting irradiance, temperature, humidity and the like of the prediction day into the fitting prediction model to obtain a fitting prediction value; on the other hand, the actual generated power of the most similar day of the history is used as the predicted value of the similar day. And finally, weighting and adding the fitting predicted value and the similar day predicted value to obtain a combined predicted value.
Description
Technical Field
The invention belongs to the technical field of photovoltaic power generation, and particularly relates to a photovoltaic power generation power prediction method based on similar days.
Background
The new energy represented by photovoltaic is connected into the power grid, so that the generated energy of the traditional thermal power can be reduced, and the environment-friendly effect is remarkable. However, because photovoltaic power generation has the characteristics of volatility and intermittence, a series of problems such as power grid lines, section tidal currents, node voltage out-of-limit and fluctuation are brought by photovoltaic grid connection, a great challenge is formed on the traditional regulation and control mode of a power grid, and great influence is also caused on the safety and stability of the operation of a power system. The photovoltaic power generation prediction refers to the prediction of the photovoltaic power generation power in a future period of time through machine learning and artificial intelligence methods based on meteorological information, historical operation data and the like, and accurate photovoltaic power generation power prediction can effectively help a power grid dispatching department to adjust a dispatching plan in time and reasonably arrange the operation mode of a power grid, so that the safe and stable operation of the power grid is guaranteed.
Compared with abroad, China starts late in the field of photovoltaic prediction and is far behind developed countries. However, in recent years, there has been a certain achievement in power prediction technology. In the prior art, historical data are generally divided into multiple weather types (sunny days, cloudy days, rainy days and the like), then, the weather data, near-day data and similar day data are used as input characteristics, model training is performed by means of an intelligent algorithm, photovoltaic prediction models under different weather types are respectively established, and finally, the weather types of the prediction days are substituted into corresponding models to obtain prediction results.
In the prior art, historical data need to be classified according to weather conditions, a large amount of complete historical data are objectively needed, and the method is not suitable for power prediction of a photovoltaic power station which is operated soon; in addition, the multi-category prediction model established according to the weather category classification is difficult to cover extreme weather conditions, and the prediction accuracy of the weather conditions with less occurrence times needs to be improved.
Therefore, it is desired to solve the above technical problems.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to solve the technical defects that the existing photovoltaic power generation prediction technology is not suitable for a new production running unit and extreme weather conditions, and provides a photovoltaic power generation power prediction method based on similar days.
The technical scheme is as follows: in order to achieve the purpose, the invention discloses a photovoltaic power generation power prediction method based on similar days, which comprises the following steps:
(1) acquiring historical operating data comprising photovoltaic power generation power, irradiance, temperature and humidity n days before the prediction day; n is an integer greater than 0;
(2) calculating the dissimilarity degree of the predicted daily irradiance curve and the previous n-day history daily irradiance curve to obtain a dissimilarity degree sequence { X1,X2,…XnAnd (5) calculating the dissimilarity degree by the following method:
wherein: t is an integer greater than 0 and less than n; xtThe difference between the previous tth day and the prediction day is indicated, and the smaller the difference is, the higher the similarity is; y isiIrradiance sequences that are predicted days;
irradiance sequence for the previous t-th day; i is 1-96 time periods;
(3) the phase difference sequence { X }1,X2,…XnGet one-dimensional ordered sample { x ] by ascending order arrangement1,x2,…xnObtaining m days with similar calendar history by using an ordered sample clustering method; m is an integer greater than 0 and less than n;
(4) training data of the m days with similar history by using a neural network to obtain a fitting prediction model, wherein irradiance, temperature and humidity are input, photovoltaic power generation power is output, and substituting the irradiance, the temperature and the humidity in a prediction period into the fitting prediction model to obtain a fitting prediction value;
(5) taking the historical day with the minimum dissimilarity degree in the step (2) as a historical most similar day, and taking the actual generated power of the historical most similar day as a predicted value of the similar day;
(6) and weighting and adding the fitting predicted value and the similar day predicted value to obtain a combined predicted value.
In the step (2), the difference degree between the predicted day and the historical day is calculated according to the irradiance curve.
Preferably, the step (3) is performed on one-dimensional ordered samples { x }1,x2,…xnThe specific steps of obtaining m calendar history similar days by using an ordered clustering algorithm comprise the following steps:
(3.1) calculating the sum of squared deviations matrix Dn×n，
The equation for the sum of squared deviations D (i, j) is:
wherein
Is the mean vector of this class, Dn×nThe other elements are set to be 0;
(3.2) calculating a classification loss function L:
where b (n, k) denotes a method of dividing n ordered samples into k classes, 1 ═ j1<j2<…jk<n＝jk+1-1 is a classification point;
(3.3) constructing a minimum classification loss matrix Cn×nAnd a classification label matrix Jn×nThe specific method comprises the following steps:
to pair
Is provided with
J(l,k)＝jlk
In the formula, l is more than or equal to 3 and less than or equal to n, and k is more than or equal to j and less than or equal to n; p (l, k) represents the best fraction for classifying l samples into k classes; j is a function oflkRepresents the starting sample number of the kth class in p (l, k); matrix Cn×nAnd matrix Jn×nThe other elements are all set as 0;
(3.4) determining each discrete interval by letting k be 3, and obtaining m days with similar calendar history:
the 3 rd interval G3＝{j3,j3+1,…,n},j3＝J(n,3)，
The 2 nd interval G2＝{j2,j2+1,…,j3-1},j2＝J(j3-1,2)，
The 1 st interval G1＝{1,2,…,j2-1}，G1The historical days contained in the interval are m calendar history similar days.
And (4) inputting the fitted model in the step (4) into irradiance, temperature and humidity.
Further, the similar day prediction value in the step (5) adopts the actual generated power of the historical most similar day.
Finally, the step of calculating the combination prediction value in the step (6) is as follows:
(6.1) calculating a weighting coefficient:
wherein: λ is a weighting systemCounting;
predicting daily maximum irradiance;
the maximum irradiance for the most similar day of history.
(6.2) calculating a combination predicted value:
P＝λP1+(1-λ)P2
wherein: p is a combined predicted value; p1Is a fitting predicted value; p2The predicted value is similar day.
Advantageous effects
Compared with the prior art, the invention has the following remarkable advantages:
(1) the photovoltaic power generation power prediction method provided by the invention does not need a large amount of historical operating data, and is suitable for a new unit which is just put into production;
(2) the method considers the actual power generation power of the most similar historical day, can obviously improve the photovoltaic power generation prediction precision under the extreme weather condition, and has important significance for safe and stable operation of the power grid.
Drawings
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a graph of dissimilarity between historical days according to an embodiment of the present invention;
FIG. 3 is a comparison curve of the fit predicted value and the actual value in the embodiment of the present invention;
FIG. 4 is a comparison curve of the predicted value and the actual value of the similar day in the embodiment of the present invention;
FIG. 5 is a comparison graph of the combined predicted value and the actual value in the embodiment of the present invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings.
Taking a 30MW photovoltaic power station as an example, the power station is put into operation formally in 5 months and 1 day in 2020, and only 31 days of historical data exist by zero point of 6 months and 1 day in 2020. Due to the fact that training data are too few, prediction accuracy of a traditional single intelligent algorithm is difficult to guarantee. After the prediction method is adopted, the prediction accuracy is obviously improved, taking the photovoltaic power generation power prediction of 6 months and 1 day of 2020 as an example, the detailed flow is shown in fig. 1:
1) acquiring data such as photovoltaic power generation power, irradiance, temperature, humidity and the like 31 days before 1/6/2020, wherein the access interval is 15 minutes;
2) calculating the dissimilarity degree of the 96-point irradiance curve of 6-month-1-day in 2020 and the 96-point irradiance curve of the history day of the first 31 days;
3) arranging the dissimilarity curves in ascending order to obtain { x1,x2,…x31Obtaining 21 calendar history similar days and history most similar days by using an ordered sample clustering method;
4) training a neural network prediction model by using data of 21-day-history similar days, wherein irradiance, temperature and humidity are input, generated power is output, and substituting the irradiance, the temperature and the humidity in a prediction time period into a fitting model to obtain a fitting prediction value P1；
5) Taking the actual generating power of the most similar historical day as a predicted value P of the similar day2；
6) Calculating to obtain a weighting coefficient of lambda, and fitting a predicted value P1And similar daily prediction value P2And weighting and adding to obtain a combined predicted value P.
Examples of the design
Actual data of main parameters such as active power, irradiance, temperature and humidity are obtained from a historical database of a photovoltaic power station, the sampling time is from No. 5/month 1 in 2020 to No. 5/month 31 in 2020, the interval is 15min, the installed capacity of the unit is 30MW, and the maximum irradiance in 5 months is 1331.33W/m2。
Firstly, respectively calculating the difference between the irradiance curve of 96 points in 6 month and 1 day of 2020 and the irradiance curve of 31 days in 5 months and 31 days of 2020, and obtaining { x by arranging in ascending order1,x2,…x31}. And classifying the historical days by using an ordered clustering algorithm, wherein dissimilarity curves of the historical days are shown in fig. 2, and 21 historical similar days are obtained, wherein the historical most similar day is 5-month-30-day 2020.
On one hand, the actual generated power of the 21 days is used as output, the irradiance, the temperature and the humidity are used as input, model training is carried out by utilizing a neural network, and a fitting prediction model is established. And substituting the predicted irradiance, the predicted temperature and the predicted humidity of the day 1 and 6 months in 2020 into the fitting prediction model to obtain a fitting prediction value of 96 points in the predicted day. Fig. 3 shows a comparison of the fitted predicted values and the actual values, and the root mean square prediction error of this method is 0.0747.
On the other hand, as can be seen from fig. 2, the minimum dissimilarity is 1172, which corresponds to the day of 2020, 5, 30. The irradiance curve of the day has the minimum difference with the irradiance curve of the predicted day, and the similarity is the highest, so the actual generated power of 5, 30 and 30 days in 2020 is taken as the predicted value of the similar day, fig. 4 shows the comparison curve of the predicted value and the actual value of the similar day, and the root mean square error of the method is 0.0600.
And finally, weighting and adding the fitting predicted value and the similar day predicted value according to the weight to obtain a combined predicted value. Wherein the maximum irradiance of 6, 1 and 1 days in 2020 is 1051.32W/m2The maximum irradiance of 5 months and 30 days is 1096.67W/m2The weighting factor λ is calculated as 0.2034. Fig. 5 shows a comparison of the combined predicted value and the actual value, resulting in a root mean square error 0.0516, the combined prediction method having a higher prediction accuracy than the single prediction method.
In conclusion, the photovoltaic power generation method provided by the invention adds the weight of the most similar day predicted value on the basis of the traditional fitting prediction model, and the accuracy of the prediction result is obviously improved.
Claims (6)
1. A photovoltaic power generation power prediction method based on similar days is characterized by comprising the following steps:
(1) acquiring historical operating data comprising photovoltaic power generation power, irradiance, temperature and humidity n days before the prediction day; n is an integer greater than 0;
(2) calculating the dissimilarity degree of the predicted daily irradiance curve and the previous n-day history daily irradiance curve to obtain a dissimilarity degree sequence { X1，X2，…XnIn which the degree of dissimilarity is calculatedThe method comprises the following steps:
wherein: t is an integer greater than 0 and less than n; xtThe difference between the previous tth day and the prediction day is indicated, and the smaller the difference is, the higher the similarity is; y isiIrradiance sequences that are predicted days;
irradiance sequence for the previous t-th day; i is 1-96 time periods;
(3) the phase difference sequence { X }1，X2，…XnGet one-dimensional ordered sample { x ] by ascending order arrangement1，x2，…xnObtaining m days with similar calendar history by using an ordered sample clustering method; m is an integer greater than 0 and less than n;
(4) training data of the m days with similar history by using a neural network to obtain a fitting prediction model, wherein irradiance, temperature and humidity are input, photovoltaic power generation power is output, and substituting the irradiance, the temperature and the humidity in a prediction period into the fitting prediction model to obtain a fitting prediction value;
(5) taking the historical day with the minimum dissimilarity degree in the step (2) as a historical most similar day, and taking the actual generated power of the historical most similar day as a predicted value of the similar day;
(6) and weighting and adding the fitting predicted value and the similar day predicted value to obtain a combined predicted value.
2. The similar-day-based photovoltaic power generation power prediction method according to claim 1, characterized in that: in the step (2), the difference degree between the predicted day and the historical day is calculated according to the irradiance curve.
3. The similar-day-based photovoltaic power generation power prediction method according to claim 1, characterized in that: for one-dimensional ordered samples { x in step (3)1，x2，…xnThe specific steps of obtaining m calendar history similar days by using an ordered clustering algorithm comprise the following steps:
(3.1) calculating the sum of squared deviations matrix Dn×n，
The equation for the sum of squared deviations D (i, j) is:
wherein
Is the mean vector of this class, Dn×nThe other elements are set to be 0;
(3.2) calculating a classification loss function L:
where b (n, k) denotes a method of dividing n ordered samples into k classes, 1 ═ j1＜j2＜…jk＜n＝jk+1-1 is a classification point;
(3.3) constructing a minimum classification loss matrix Cn×nAnd a classification label matrix Jn×nThe specific method comprises the following steps:
to pair
Is provided with
J(l，k)＝jlk
In the formula, l is more than or equal to 3 and less than or equal to n, and k is more than or equal to j and less than or equal to n; p (l, k) represents the best fraction for classifying l samples into k classes; j is a function oflkRepresents the starting sample number of the kth class in p (l, k); matrix Cn×nAnd matrix Jn×nThe other elements are all set as 0;
(3.4) determining each discrete interval by letting k be 3, and obtaining m days with similar calendar history:
the 3 rd interval G3＝{j3，j3+1，…，n}，j3＝J(n，3)，
The 2 nd interval G2＝{j2，j2+1，…，j3-1}，j2＝J(j3-1，2)，
The 1 st interval G1＝{1，2，…，j2-1}，G1The historical days contained in the interval are m calendar history similar days.
4. The similar-day-based photovoltaic power generation power prediction method according to claim 1, characterized in that: and (4) inputting the fitted model in the step (4) into irradiance, temperature and humidity.
5. The similar-day-based photovoltaic power generation power prediction method according to claim 1, characterized in that: and (5) adopting actual generated power of the history most similar day by the similar day prediction value in the step (5).
6. The similar-day-based photovoltaic power generation power prediction method according to claim 1, characterized in that: the step of calculating the combined prediction value in the step (6) is as follows:
(6.1) calculating a weighting coefficient:
wherein: λ is a weighting coefficient;
predicting daily maximum irradiance;
maximum irradiance for the most similar day of history;
(6.2) calculating a combination predicted value:
P＝λP1+(1-λ)P2
wherein: p is a combined predicted value; p1Is a fitting predicted value; p2The predicted value is similar day.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011344728.8A CN112508246B (en) | 2020-11-26 | 2020-11-26 | Photovoltaic power generation power prediction method based on similar days |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011344728.8A CN112508246B (en) | 2020-11-26 | 2020-11-26 | Photovoltaic power generation power prediction method based on similar days |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112508246A true CN112508246A (en) | 2021-03-16 |
| CN112508246B CN112508246B (en) | 2022-09-16 |
Family
ID=74966721
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011344728.8A Active CN112508246B (en) | 2020-11-26 | 2020-11-26 | Photovoltaic power generation power prediction method based on similar days |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112508246B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113689057A (en) * | 2021-10-25 | 2021-11-23 | 广东电网有限责任公司佛山供电局 | Photovoltaic power generation power prediction method and device |
| CN115529002A (en) * | 2022-11-28 | 2022-12-27 | 天津海融科技有限公司 | Photovoltaic power generation power prediction method under various weather conditions |
| CN117060407A (en) * | 2023-10-12 | 2023-11-14 | 国网湖北省电力有限公司经济技术研究院 | Wind power cluster power prediction method and system based on similar day division |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109002915A (en) * | 2018-07-13 | 2018-12-14 | 福州大学 | Photovoltaic plant short term power prediction technique based on Kmeans-GRA-Elman model |
| CN109389305A (en) * | 2018-09-30 | 2019-02-26 | 南京地铁集团有限公司 | Method for judging passenger traffic flow state in urban rail transit section |
-
2020
- 2020-11-26 CN CN202011344728.8A patent/CN112508246B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109002915A (en) * | 2018-07-13 | 2018-12-14 | 福州大学 | Photovoltaic plant short term power prediction technique based on Kmeans-GRA-Elman model |
| CN109389305A (en) * | 2018-09-30 | 2019-02-26 | 南京地铁集团有限公司 | Method for judging passenger traffic flow state in urban rail transit section |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113689057A (en) * | 2021-10-25 | 2021-11-23 | 广东电网有限责任公司佛山供电局 | Photovoltaic power generation power prediction method and device |
| CN115529002A (en) * | 2022-11-28 | 2022-12-27 | 天津海融科技有限公司 | Photovoltaic power generation power prediction method under various weather conditions |
| CN117060407A (en) * | 2023-10-12 | 2023-11-14 | 国网湖北省电力有限公司经济技术研究院 | Wind power cluster power prediction method and system based on similar day division |
| CN117060407B (en) * | 2023-10-12 | 2023-12-22 | 国网湖北省电力有限公司经济技术研究院 | Wind power cluster power prediction method and system based on similar day division |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112508246B (en) | 2022-09-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112508246B (en) | Photovoltaic power generation power prediction method based on similar days | |
| CN110474339B (en) | Power grid reactive power control method based on deep power generation load prediction | |
| Hu et al. | A hybrid model based on CNN and Bi-LSTM for urban water demand prediction | |
| CN111008728B (en) | Prediction method for short-term output of distributed photovoltaic power generation system | |
| CN103390202B (en) | A kind of photovoltaic plant output power predicting method chosen based on set of metadata of similar data | |
| CN109165774A (en) | A kind of short-term photovoltaic power prediction technique | |
| CN110705789A (en) | Photovoltaic power station short-term power prediction method | |
| CN114792156B (en) | Photovoltaic output power prediction method and system based on curve characteristic index clustering | |
| CN111092451B (en) | Photovoltaic consumption early warning method based on power distribution network grid | |
| CN110866633B (en) | Micro-grid ultra-short-term load prediction method based on SVR support vector regression | |
| CN116629416A (en) | Photovoltaic power station power prediction method and device | |
| CN114648176A (en) | Wind-solar power consumption optimization method based on data driving | |
| CN110909310A (en) | Photovoltaic short-term power generation capacity prediction method and system based on model parameter optimization | |
| Dou et al. | Elman neural network based short-term photovoltaic power forecasting using association rules and kernel principal component analysis | |
| CN115688993A (en) | Short-term power load prediction method suitable for power distribution station area | |
| CN112669168A (en) | Short-term wind power prediction method | |
| CN114021783A (en) | Two-stage monthly unit combination and maintenance plan optimization method considering social carbon-like emission factors and short-term benefits | |
| CN108694475B (en) | Short-time-scale photovoltaic cell power generation capacity prediction method based on hybrid model | |
| CN110852492A (en) | Photovoltaic power ultra-short-term prediction method for finding similarity based on Mahalanobis distance | |
| CN113033136B (en) | Simplified photovoltaic cell physical parameter extraction optimization method and system | |
| CN117993611A (en) | Flexible heat source new energy consumption capability assessment method based on scene time sequence | |
| CN112288130B (en) | New energy consumption calculation method based on two-stage multi-objective optimization | |
| CN111242371B (en) | Photovoltaic power generation short-term prediction correction method based on non-iterative multi-model | |
| CN112836876A (en) | Power distribution network line load prediction method based on deep learning | |
| Xiaoyun et al. | Short-Term wind power prediction of regions based on deep long short-term memory network and gaussian process regression |
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 |