CN112865093B - Combined prediction method for short-time power load - Google Patents
Combined prediction method for short-time power load Download PDFInfo
- Publication number
- CN112865093B CN112865093B CN202110226784.XA CN202110226784A CN112865093B CN 112865093 B CN112865093 B CN 112865093B CN 202110226784 A CN202110226784 A CN 202110226784A CN 112865093 B CN112865093 B CN 112865093B
- Authority
- CN
- China
- Prior art keywords
- load
- season
- time
- day
- attribute
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000012706 support-vector machine Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 24
- 230000001932 seasonal effect Effects 0.000 claims description 22
- 230000007704 transition Effects 0.000 claims description 22
- 238000010438 heat treatment Methods 0.000 claims description 12
- 238000003066 decision tree Methods 0.000 claims description 10
- 238000001816 cooling Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 241001123248 Arma Species 0.000 claims description 5
- 238000002790 cross-validation Methods 0.000 claims description 3
- 241000207923 Lamiaceae Species 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000005057 refrigeration Methods 0.000 claims 1
- 238000012795 verification Methods 0.000 claims 1
- 238000012360 testing method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000010248 power generation Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 1
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 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
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A30/00—Adapting or protecting infrastructure or their operation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a short-time power load prediction method, which is characterized in that according to the correlation between meteorological factors and power loads, total power loads are split into basic loads and meteorological sensitive loads, the basic loads are subjected to load prediction by adopting an autoregressive time sequence method, the meteorological sensitive loads are subjected to load prediction by adopting a support vector machine algorithm, and the two predicted loads are overlapped to realize the prediction of the total power loads. Compared with the traditional short-time power load prediction method, the method has the advantages that the power load is split into the base load and the weather-sensitive load, the prediction model is respectively built aiming at different load types, the accuracy of the power load prediction model is improved, and the annual applicability of the prediction method is improved.
Description
Technical Field
The invention relates to the technical field of power load prediction, in particular to a short-time power load combined prediction method.
Background
The electric power is the basic energy for the national life, and the safe and reliable electric power system is an important guarantee for the national economic growth and the social development. Unlike the general industrial type industry, the electric power system cannot store electric energy in large scale, which also requires that the electric energy is produced and consumed in real time, i.e. the on-demand supply at the moment is maintained. The essence of load prediction is to predict the demands of users in the power system, and accurate load prediction is closely related to each layer of the power system, and plays a role in normal operation of the power system. The prediction of the power load on the demand side is a basis for power planning of various departments such as power generation enterprises, power transmission and distribution departments, dispatching centers and the like, making a power generation plan, arranging a maintenance plan of a generator set, planning regional power transactions and the like, and is a basis for ensuring the safety, reliability and economy of power supply.
The power load prediction is classified into the following four types according to the length of the future prediction time: 1. ultra-short-term load prediction, namely predicting load within 1 hour in the future, is mainly used for emergency dispatching and demand side response adjustment of the electric power market; 2. short-term load prediction, which aims at power load of one day to several days in the future, is mainly used for optimizing a power supply side power generation plan, efficiently arranges a generator set to operate, is convenient for a power transmission and distribution side to reasonably arrange a scheduling plan, and provides a basis for short-term power price fluctuation; 3. mid-term load prediction refers to predicting power load of one month to one year in the future, so that a power supply side is convenient to arrange a power generation plan with a longer period, and a unit maintenance plan is convenient to arrange; 4. the long-term load prediction refers to load prediction 3 to 5 years in advance, and is mainly used for long-term planning of a power system, and is convenient for planning projects such as power supply capacity expansion, power grid construction and the like.
The prior short-time power load prediction model mainly adopts a data-driven prediction model, and ignores physical meanings such as power load self-constitution, change characteristics and the like. The establishment of the data driving model depends on the quality of data samples, and in practical situations, the data sample size is often limited, which also causes the situation that the generalization of the existing model algorithm is low. Meanwhile, the existing short-time power load prediction model is incomplete in consideration of power load influence factors, and only the influence of air temperature on the power load is considered. Therefore, it is necessary to develop a short-time power load prediction model that comprehensively considers the influence of various weather factors on the power load from the viewpoint of the physical characteristics of the power load composition, so as to improve the generalization and prediction accuracy of the algorithm.
Disclosure of Invention
The invention aims to solve the technical problems of overcoming the defects existing in the prior art and providing a short-time power load combined prediction method.
In order to solve the technical problems, the invention adopts the following solutions:
a short-time power load combined prediction method specifically comprises the following steps as shown in figure 1:
step 1, judging the season attribute of the current day;
step 2, load prediction;
in step 2, the load prediction step specifically includes:
step 2.1, predicting a base load;
and 2.2, weather-sensitive load prediction.
In step 1, the current day season attribute determination is performed by using a current day season attribute determination model for determining the current day season of the future day, and the specific flow is as shown in fig. 2, and the season attribute S of the predicted day is determined by a mode weighted average of the time-by-time season attributes of the future day D ：
Wherein S is T Is a seasonal attribute at a certain moment; p is the season attribute weighting coefficient of different time periods; n is the number of season attribute modes at different times of the future day; i represents each time corresponding to the season attribute mode.
The seasonal attribute of the moment is calculated by a time-by-time seasonal attribute judging method, a decision tree model is adopted for judging the time-by-time seasonal attribute, and input parameters are a moving average effective temperature (mET, DEG C) of the first three hours, a moving average air enthalpy value (mEnth, kJ/kg) of the first three hours, a moving average moisture content (md, g/kg) of the first three hours and a moving average dew point temperature (mTS, DEG C) of the first three hours. Obtaining seasonal division S at a certain moment through judging rules T Seasonal divisions are defined as discrete attributes, corresponding physical meanings are shown in the following formula:
the season division is shown in fig. 3, the judging rule of the season division is that the root node takes the effective temperature as the root node attribute according to the information gain degree of each attribute, when the current time moving average effective temperature is less than 9.8 ℃, the data is led to the left branch, and the heat supply season and the transition season are divided downwards. According to the study on the thermal comfort of the human body, the human body can generate cold feeling when the effective temperature is less than 9 ℃. The second-level node examines the enthalpy value of the air, and when the enthalpy value of the left branch is smaller than 43kJ/kg, the left branch reaches the leaf node, so that the heating is required at the current moment and the heating season is judged. And on the contrary, the unique attribute category is obtained by the fact that the moisture content of the third-stage node is larger than or smaller than 10g/kg, so that the heating season and the transition season are divided. The effective temperature of the root node data is more than 9.8 ℃, and the data branches to the right side to divide the refrigerating season and the transition season finally. The maximum gain of the right-side secondary node information is also an enthalpy value, and when the enthalpy value is smaller than 20.8kJ/kg, the transition season can be judged. And otherwise, continuing to downwards reach the third-season node, and judging the cooling season when the effective temperature is higher than 22 ℃. The subset of data for the fourth layer node is divided into a transition season and a cooling season by whether the moisture content is greater than 15g/kg when the effective temperature is less than 22 ℃.
In the time-by-time seasonal attribute decision tree model, n-fold cross validation is adopted in model training in order to reduce the overfitting phenomenon of the decision tree. The training samples were randomly partitioned into n subsets, where n-1 subsets were used for training and another individual sample was used as data for the validation model, repeated n times across, and the n-time result optimization model was synthesized.
The season property weighting coefficients of different periods are determined according to the possibility that people are air-conditioned in different periods, and the daytime is divided into night low-valley periods (0:00-7:00), daytime peak periods (8:00-18:00) and evening peak periods (19:00-23:00) according to daily peak-valley changes. The season attribute weighting coefficients of different time periods are:
in step 2.1, the base load is a load which is greatly influenced by life work and rest and has small correlation with weather change, and is divided into a workday base load and a holiday base load according to the types of days. The daily load of the weighted average of the historical data of the corresponding daily type in the transition season is adopted as the basic load part of the power load. Adopting the seasonal attribute determination model, the daily base load Q F_D The calculation formula is as follows:
wherein, consider the day-season attribute S D When=0, the average daily load in the transition season is the sameDaily base load; when 0 < |S D When the level is less than 0.5, the time of day base load is the average value of the base load at each time of the day, and the base load Q at each time of the day is the incomplete transition season F_Ti The method comprises the following steps:
wherein T is i Indicating a certain moment; s is S Ti The season attribute value corresponding to a certain moment; n is T i A plurality of moments before the moment for calculating;is T i-m Base load at time.
In step 2.1, the base load is predicted by adopting a base load prediction model through a time sequence method, and input data is the base load of the same type of day adjacent to the predicted day calculated by the method. For stationary time series data, the main prediction models are MA (q), AR (p) and ARMA (p, q); for non-stationary time sequence data, an ARMA prediction model is built for the obtained new stationary time sequence sample by removing trend change or seasonal change of the data through difference, wherein the ARMA prediction model is mainly used as the prediction model. When determining the model order, estimating p and q by adopting a red pool information criterion (AIC, akaike information criterion), and obtaining p and q under the minimum AIC as model parameters.
The weather-sensitive load refers to that in the same daily type group, the current day season attribute is judged, if the current day load is a heating season or a cooling season, the current day load is used for subtracting the base load part, and the rest part is the weather-sensitive load. The meteorological sensitive load is predicted by adopting a support vector machine (SVR) model, the model parameter optimization adopts a grid search method, and the optimized parameters in the model comprise a punishment factor C and a kernel function parameter gamma, and the formula is as follows:
wherein σ is the kernel width.
In step 2.2, the input parameters of the weather-sensitive load prediction model are changed according to the type of the current day season of the prediction day, the type of the current day season is heat supply Ji Shi (sd=1), and the variables of the input model are shown in the following formula:
S D ＝1：
wherein T is i Is the predicted time; mET is the effective temperature of four points of moving average before the predicted time;is T i-j Weather-sensitive loads at time. The input space is six-dimensional.
When the type of the season of the day is cold supply season (S D = -1), the variables of the input model are as shown in the following formula:
wherein mT is the four-point moving average temperature before the predicted time;the four points before the predicted moment are the moving average relative humidity; mν is the four-point sliding average wind speed before the predicted time. The input space is eight dimensions.
The short-time power load combined prediction method can be used for predicting the power load 15 minutes a day in the future, and the predicted total load at each moment is as follows:
Q All_T ＝Q F_T +Q W_T
wherein Q is All_T A predicted value of the power load at a certain time of day in the future; q (Q) F_T A base load predicted value for the power load at a time of day in the future; q (Q) W_T And weather-sensitive load forecast values for the power load at a certain time of the future day.
Drawings
FIG. 1 is a flow chart of a short-term power load combined prediction method.
Fig. 2 is a flowchart of a method for determining the season of the day attribute.
Fig. 3 shows a season division determination rule at a certain time.
Fig. 4 model confusion matrix.
FIG. 5 is a timing diagram of predicted and target values of the data according to the embodiment.
FIG. 6 is a graph of the optimization results of the SVR model parameter penalty factor C and the kernel function distribution width gamma.
Figure 7 shows how the model fits and deviates on the training set.
Figure 8 shows how the model fits and deviates over the test set.
Detailed Description
FIG. 1 is a flow chart of a short-term power load combined prediction method of the invention.
The invention will be further described in detail (training and predicting power load every 15 minutes in 2014 in Shanghai, data derived from Shanghai economic and information Commission data platform) with reference to the accompanying drawings and specific examples, comprising the following steps:
step 1: and training a current day seasonal attribute judgment model. Based on background knowledge of building energy consumption, the heating season determined by Shanghai is one month, the cooling season determined by Shanghai is July, and transition seasons can occur in four and October. Therefore, the seasonal attribute of the data in July is assigned to be 1, the seasonal attribute of the data in July is assigned to be-1, and according to the comfort index of the human body, the human body feels more comfortable when the effective temperature ET falls between 15 and 23 ℃, so that the effective temperature time period of four and October within the range can be considered as the transition season. 4. The effective temperature of the month of October is not in the range, and the season attribute is considered to be indistinguishable. The period data may be classified by rule mining of training data by a subsequent decision tree.
The input training data are data of time by time of July and July, and data of effective temperatures of four and October falling at 15-23deg.C, and the season discrimination rules generated by decision tree are shown in figure 3. Through training, the classification accuracy of the generated decision tree is 92.5%. The model confusion matrix is shown in fig. 4, the error of the model on the whole classification of the transition season is larger than 10%, the classification error of the transition season and the heating season is larger than 7%, and the model confusion matrix is mainly characterized in that the transition season weather parameters and the heating season weather parameters in training data overlap more, and the number of branch stages on the left side of the decision tree is smaller. But in order to prevent data from being over fitted and for time-by-time seasonal decisions, the model accuracy has reached the requirements.
Step 2: and training a base load prediction model. Training modeling is carried out by taking continuous five working days from 3 months 17 days to 3 months 21 days as an example, 3 months 24 days load is predicted, and the model prediction performance is analyzed, wherein the interval date is holidays. After the data were subjected to stationarity judgment, model identification and scaling, the model was determined to be ARIMA (5,2,3), and the coefficients obtained by the least squares method are shown in table 1:
table 1 workday sample prediction model parameter determination
After calculating the specific form of the prediction model, 96 point data of 24 days of 3 months are predicted and compared with the actual daily base load so as to evaluate the performance of the prediction model. The model evaluation indexes of the test set are shown in table 2, and the prediction accuracy of the time series method on the basis load in the transition season is high, the relative error is less than 2%, and the fitting degree of the model is high. The method is shown to have better base load performance for predicting the stability of daytime variation.
Table 2 workday sample test set prediction model evaluation table
Fig. 5 shows a timing diagram of the predicted value and the target value of the 96-point data of 24 days of 3 months, and the predicted value and the target value shown in the lower diagram are highly fitted, so that the change characteristic of the base load is predicted well.
Step 3: and training a weather-sensitive load prediction model to predict the total power load. Taking a heating season prediction model as an example, in order to ensure sufficient training sample numbers, 960 items of data from 1 month, 7 days to 1 month, 16 days and ten days are selected as training sets, 17 days and 20 days of data from 1 month are selected as test sets, and the interval date is the holiday of the rejection. Fig. 6 shows an optimization result diagram of the SVR model parameter penalty factor C and the kernel function distribution width γ, where the model minimum normalized Mean Square Error (MSE) is targeted by a grid search method, and the obtained C and γ are the optimal parameters under the input sample. For the training sample optimal equation parameters c=2.828, γ=5.67.
Table 3 shows a summary of the performance parameters of the weather-sensitive load prediction model based on the training set and the prediction set. The training set adopts a 5-fold cross validation method when performing model fitting, and the trained model has better performance, namely R is firstly 2 Higher, have higher fitting degree. And secondly, the RMSE of the predicted value and the target value is smaller, which indicates that the residual distribution is more concentrated, the MAE is 170.8, and the average relative error is 12%. In contrast, the model has reduced prediction performance for the test set, but the comparison of average relative errors, combined with the fact that the input parameter dimension is small and the training sample is large enough, shows that the model has better generalization and excludes the situation of overfitting. The test set RMSE and MAE are larger, mainly because the fluctuation of weather sensitive load is more random and the periodic variability is not strong. However, the data of the addition of the predicted weather-sensitive load and the base load, compared with the total power load sequence, shows that MAPE is significantly reduced, and the relative error on the whole prediction set is greatly optimized.
Table 3 Meteorological sensitive load prediction model Performance summary
Figures 7 and 8 show how the model fits and deviates on the training set and test set, respectively, and overall the model can better predict the changing trend of the power load.
Claims (4)
1. A short-time power load combined prediction method is characterized in that the flow comprises the following steps:
step 1, judging the season attribute of the current day;
in step 1, the current day season attribute determination is performed by using a current day season attribute determination model for determining the current day season of the future day, and determining the season attribute S of the predicted day by a mode weighted average of the time-by-time season attributes of the future day D ：
Wherein S is T Is a seasonal attribute at a certain moment; p is the season attribute weighting coefficient of different time periods; n is the number of season attribute modes at different times of the future day; i represents each moment corresponding to the season attribute mode;
step 2, load prediction; in step 2, the load prediction step specifically includes:
step 2.1, predicting a base load;
in the step 2.1, the base load is a load which is greatly influenced by life work and rest and has smaller correlation with weather change, and is divided into a workday base load and a holiday base load according to different day types; adopting daily load of weighted average of each moment of historical data of corresponding daily type in transition season as a basic load part of the power load; adopting the seasonal attribute determination model, the daily base load Q F_D The calculation formula is as follows:
wherein, consider the day-season attribute S D When=0, the average daily load in the transition season is the base daily load; when 0 < |S D When the level is less than 0.5, the time of day base load is the average value of the base load at each time of the day, and the base load Q at each time of the day is the incomplete transition season F_Ti The method comprises the following steps:
wherein T is i Indicating a certain moment; s is S Ti The season attribute value corresponding to a certain moment; n is T i A plurality of moments before the moment for calculating;is T i-m A moment base load;
and 2.2, weather-sensitive load prediction.
2. The short-term power load combined prediction method according to claim 1, wherein in step 1,
the seasonal attribute at the moment is calculated by a time-by-time seasonal attribute judging method, a decision tree model is adopted for judging the time-by-time seasonal attribute, and input parameters are a first three-hour moving average effective temperature mET (DEG C), a first three-hour moving average air enthalpy value mEnth (kJ/kg), a first three-hour moving average moisture content md (g/kg) and a first three-hour moving average dew point temperature mTs (DEG C); obtaining seasonal division S at a certain moment through judging rules T Seasonal divisions are defined as discrete attributes, corresponding physical meanings are shown in the following formula:
according to the season division, the root node takes the effective temperature as the root node attribute according to the information gain degree of each attribute, when the current time sliding average effective temperature is less than 9.8 ℃, the data is led to the left branch, and the heat supply season and the transition season are divided downwards; according to the related study of the thermal comfort of the human body, when the effective temperature is less than 9 ℃, the human body can generate cold feeling; the second-level node examines the enthalpy value of the air, and when the enthalpy value is less than 43kJ/kg, the left branch reaches the leaf node, so that the heating is required at the current moment and the heating season is judged; on the contrary, the unique attribute category is obtained through whether the moisture content of the third-stage node is more than or less than 10g/kg, so that a heating season and a transition season are divided; the effective temperature of the root node data is more than 9.8 ℃, and the data branches to the right side to flow, so that a refrigerating season and a transition season are finally divided; the maximum gain of the right-side secondary node information is also an enthalpy value, and when the enthalpy value is smaller than 20.8kJ/kg, the transition season can be judged; otherwise, continuing to downwards reach the third-stage node, and judging that the cooling season is performed when the effective temperature is higher than 22 ℃; dividing the subset of data of the fourth level node into a transition season and a refrigeration season by whether the moisture content is greater than 15g/kg when the effective temperature is less than 22 ℃;
in the time-by-time seasonal attribute decision tree model, n-fold cross validation is adopted in model training in order to reduce the overfitting phenomenon of the decision tree; randomly dividing the training samples into n subsets, wherein n-1 subsets are used for training, another single sample is used as data of a verification model, the training samples are alternately repeated for n times, and the result optimization model is synthesized for n times;
the season attribute weighting coefficients of different time periods are determined according to the possibility that people turn on air conditioner in different time periods, and the time periods are divided into night valley time periods 0:00-7:00, daytime peak time periods 8:00-18:00 and evening peak time periods 19:00-23:00 according to daily peak-valley changes; the season attribute weighting coefficients of different time periods are:
3. the short-term power load combined prediction method according to claim 2, characterized in that,
in step 2.1, the base load is predicted by adopting a base load prediction model through a time sequence method, and input data is the base load of the same type of day adjacent to the predicted day calculated by the method; for stationary time series data, the prediction model is MA (q), AR (p) and ARMA (p, q); for non-stable time sequence data, establishing an ARMA prediction model for the obtained new stable time sequence sample through differential removal of trend change or seasonal change of the data; when determining the model order, estimating p and q by adopting a red pool information criterion AIC (Akaike information criterion) to obtain p and q under the minimum AIC as model parameters;
the weather-sensitive load refers to the weather-sensitive load, the weather-sensitive load is divided into groups of the same day type, the current day season attribute is judged, if the weather-sensitive load is a heating season or a cooling season, the current day load is used for subtracting the base load part, and the excess part is the weather-sensitive load; the meteorological sensitive load is predicted by adopting a support vector machine (SVR) model, the model parameter optimization adopts a grid search method, and the optimized parameters in the model comprise a punishment factor C and a kernel function parameter gamma, and the formula is as follows:
wherein σ is the kernel width.
4. The short-term power load combined prediction method according to claim 3, characterized in that,
in step 2.2, the input parameters of the weather-sensitive load prediction model are changed along with the different types of the seasons of the day of the prediction, and when the types of the seasons of the day are heat supply seasons, the variables of the input model are shown in the following formula:
wherein T is i Is the predicted time; mET is the effective temperature of four points of moving average before the predicted time;is T i-j Weather-sensitive load at time;
when the type of the day season is the cold season, the variables of the input model are as follows:
wherein mT is the four-point moving average temperature before the predicted time;the four points before the predicted moment are the moving average relative humidity; mν is the four-point sliding average wind speed before the predicted time;
the short-time power load combined prediction method can be used for predicting the power load 15 minutes a day in the future, and the predicted total load at each moment is as follows:
Q All_T ＝Q F_T +Q W_T
wherein Q is All_T A predicted value of the power load at a certain time of day in the future; q (Q) F_T A base load predicted value for the power load at a time of day in the future; q (Q) W_T And weather-sensitive load forecast values for the power load at a certain time of the future day.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110226784.XA CN112865093B (en) | 2021-03-01 | 2021-03-01 | Combined prediction method for short-time power load |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110226784.XA CN112865093B (en) | 2021-03-01 | 2021-03-01 | Combined prediction method for short-time power load |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112865093A CN112865093A (en) | 2021-05-28 |
| CN112865093B true CN112865093B (en) | 2024-02-23 |
Family
ID=75990672
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110226784.XA Active CN112865093B (en) | 2021-03-01 | 2021-03-01 | Combined prediction method for short-time power load |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112865093B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113850432A (en) * | 2021-09-24 | 2021-12-28 | 广东电网有限责任公司 | Prediction method, system, equipment and medium of power load data |
| CN115860797B (en) * | 2022-12-08 | 2023-07-18 | 国网江苏省电力有限公司南通供电分公司 | Electric quantity demand prediction method suitable for new electricity price reform situation |
| CN116817415B (en) * | 2023-08-28 | 2024-01-12 | 国网浙江省电力有限公司宁波供电公司 | Air conditioner load management and adjustment method, computing equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2016073155A (en) * | 2014-10-01 | 2016-05-09 | 中国電力株式会社 | Demand prediction device, demand prediction method, and program |
| CN109242189A (en) * | 2018-09-12 | 2019-01-18 | 国网安徽省电力有限公司合肥供电公司 | A kind of Short-Term Load Forecasting of Electric Power System based on meteorologic factor |
| CN109670670A (en) * | 2018-11-13 | 2019-04-23 | 上海电器科学研究所（集团）有限公司 | A kind of electric load method for splitting dividing season attribute |
| CN110135617A (en) * | 2019-03-31 | 2019-08-16 | 中南林业科技大学 | A kind of prediction technique of improvement short term under meteorological factor influence based on part throttle characteristics |
-
2021
- 2021-03-01 CN CN202110226784.XA patent/CN112865093B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2016073155A (en) * | 2014-10-01 | 2016-05-09 | 中国電力株式会社 | Demand prediction device, demand prediction method, and program |
| CN109242189A (en) * | 2018-09-12 | 2019-01-18 | 国网安徽省电力有限公司合肥供电公司 | A kind of Short-Term Load Forecasting of Electric Power System based on meteorologic factor |
| CN109670670A (en) * | 2018-11-13 | 2019-04-23 | 上海电器科学研究所（集团）有限公司 | A kind of electric load method for splitting dividing season attribute |
| CN110135617A (en) * | 2019-03-31 | 2019-08-16 | 中南林业科技大学 | A kind of prediction technique of improvement short term under meteorological factor influence based on part throttle characteristics |
Non-Patent Citations (1)
| Title |
|---|
| 考虑多因素气象的电网短期负荷预测建模研究;覃芳璐;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》(第01期);第C042-787页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112865093A (en) | 2021-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112865093B (en) | Combined prediction method for short-time power load | |
| Luo et al. | Feature extraction and genetic algorithm enhanced adaptive deep neural network for energy consumption prediction in buildings | |
| CN111932402B (en) | Short-term power load bidirectional combination prediction method based on similar day and LSTM | |
| Wu et al. | A novel hybrid model for short‐term forecasting in PV power generation | |
| Aras et al. | Forecasting residential natural gas demand | |
| CN109670670B (en) | Power load splitting method for dividing seasonal attributes | |
| CN105117810A (en) | Residential electricity consumption mid-term load prediction method under multistep electricity price mechanism | |
| CN112149890A (en) | Comprehensive energy load prediction method and system based on user energy label | |
| Kim et al. | Time-series clustering and forecasting household electricity demand using smart meter data | |
| CN111723085B (en) | Method and system for cleaning electric energy consumption data of public building air conditioner | |
| Kwon et al. | Weekly peak load forecasting for 104 weeks using deep learning algorithm | |
| Zhang et al. | Similarity-based grouping method for evaluation and optimization of dataset structure in machine-learning based short-term building cooling load prediction without measurable occupancy information | |
| CN115860797A (en) | Electric quantity demand prediction method suitable for new electricity price reform situation | |
| Yakut et al. | Modeling of energy consumption forecast with economic indicators using particle swarm optimization and genetic algorithm: an application in Turkey between 1979 and 2050 | |
| CN113313312A (en) | Spring festival load rolling prediction method based on short, medium and long time scales | |
| Kareem et al. | Removing seasonal effect on city based daily electricity load forecasting with linear regression | |
| CN108154259A (en) | Load forecasting method and device, storage medium, the processor of heat pump | |
| CN117200352A (en) | Photovoltaic power generation regulation and control method and system based on cloud edge fusion | |
| Ma et al. | Residential high-power load prediction based on optimized LSTM network | |
| Nguyen et al. | A machine learning-based approach for the prediction of electricity consumption | |
| CN115829418A (en) | Power consumer load characteristic portrait construction method and system suitable for load management | |
| Ramnath et al. | A statistical and predictive modeling study to analyze impact of seasons and COVID-19 factors on household electricity consumption | |
| CN115271168A (en) | Method and device for predicting response potential of electrical load and storage medium | |
| Zhou et al. | LSTM Based Quantile Regression Method for Holiday Load Forecasting | |
| Fan et al. | Short-term multi-region load forecasting based on weather and load diversity analysis |
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 |