CN110298509B - Large industrial industry electricity utilization optimization method combined with short-term load prediction - Google Patents
Large industrial industry electricity utilization optimization method combined with short-term load prediction Download PDFInfo
- Publication number
- CN110298509B CN110298509B CN201910575725.6A CN201910575725A CN110298509B CN 110298509 B CN110298509 B CN 110298509B CN 201910575725 A CN201910575725 A CN 201910575725A CN 110298509 B CN110298509 B CN 110298509B
- Authority
- CN
- China
- Prior art keywords
- load
- enterprise
- peak
- short
- index
- 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
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
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P80/00—Climate change mitigation technologies for sector-wide applications
- Y02P80/10—Efficient use of energy, e.g. using compressed air or pressurized fluid as energy carrier
Abstract
The invention provides a power consumption optimization method for large industrial industry in combination with short-term load prediction, which comprises the following steps: selecting a basic electric charge charging mode based on random forest short-term load prediction, and optimizing the basic electric charge; the load is adjusted based on peak valley period production of transverse and longitudinal comparison analysis; based on industry contrast analysis, the power factor is adjusted. The invention takes the electric charge of large industrial users as an access point, analyzes three factors of basic electric charge, electric power charge and power factor adjustment electric charge which influence the electric charge expenditure of the users, researches the reason of overhigh average unit price of a single customer from the whole industry perspective, and provides differentiated electricity utilization optimization suggestions by combining short-term load prediction results.
Description
Technical Field
The invention relates to an industrial electricity optimization method, in particular to an electricity optimization method for large industrial industry combined with short-term load prediction.
Background
With the deep electric change, the electric vending main body in each place continuously emerges, the market competition is more and more vigorous, the price difference is continuously reduced, and the electric vending company is urgently required to meet the electric consumption requirement of customers by expanding the value-added service of the customers, lock important electric power users and resist the competition of the electric vending company. From the aspect of saving electric charge, the construction of the electricity consumption optimization method guides users to reasonably use electricity, creates a good business environment for the users, and becomes an important cause. In the current power industry, power consumption optimization suggestions provided for customers are generally judged according to the power consumption and electricity charge of the customers, the historical power consumption characteristics of the customers are mainly considered, comparison analysis cannot be carried out from the general angle of the industry to which the customers belong, basic power consumption optimization cannot be effectively carried out by combining power consumption load prediction, and a comprehensive and scientific industry power consumption optimization method is provided.
Disclosure of Invention
In order to solve the technical problems in the prior art, the invention provides a power consumption optimization method for large industrial industry, which combines short-term load prediction, uses the power charge of large industrial users as an access point, analyzes three factors of basic power charge, electric power charge and power factor adjustment power charge which influence the power charge expenditure of the users, researches the reason that the average unit price of a single customer is too high from the whole industry perspective, and combines the short-term load prediction result to provide differentiated power consumption optimization suggestions.
The invention is realized by adopting the following technical scheme: a power consumption optimization method for large industrial industry combined with short-term load prediction comprises the following steps:
s1, selecting a basic electric charge charging mode based on random forest short-term load prediction, and optimizing the basic electric charge;
s2, peak valley period production based on transverse and longitudinal comparison analysis is carried out, and load is adjusted;
s3, adjusting the power factor based on industry comparison analysis.
In a preferred embodiment, step S1 comprises:
s11, acquiring load data of monthly frequency of enterprises in large industrial industry and related influence factor indexes, and obtaining a broad table of short-term load and factor indexes;
s12, searching data and screening key influence factor indexes;
s13, establishing a random forest load prediction model, and predicting the short-term load demand of an enterprise;
s14, selecting an enterprise basic electricity charge charging mode according to the prediction result of the enterprise short-term load demand.
Wherein, step S12 includes: index data consistency checking, anomaly identification and deletion filling, and screening out key influence factor indexes influencing load change;
the consistency check is as follows: consistency check is carried out on the occurrence time of the load data, the occurrence time is taken as a critical value, the recorded abnormal load data with the occurrence time exceeding the system time is identified, the abnormal load data is regarded as dirty data, the dirty data is directly removed and processed, and the normal data recorded in the occurrence time of the load is reserved;
the anomaly identification is as follows: identifying abnormal values of the historical load by adopting a box diagram of a space method, identifying the abnormal values of the load, and filling load data identified as the abnormal values with a historical synchronous average value;
the deletion filling is as follows: filling the missing values of the associated influence factor indexes; wherein, the missing value of the seasonal classified index is filled by the synchronous value in the latest period; seasonal numerical indicators, the missing values of which are filled with historical contemporaneous averages; non-seasonal classified indexes, wherein the missing values are filled by index value modes; a non-seasonal numerical indicator, replacing with an average load of the first 12 months of occurrence of the missing value;
the screening process of the key influence factor index comprises the following steps: the method comprises the steps of screening the data type indexes by calculating a correlation coefficient matrix among variables and carrying out the numerical type index screening of key influence factors; the method comprises the steps of discretizing a load index, calculating the correlation between the discretized load index and a classified index by adopting chi-square test, and screening the classified index of key influencing factors.
As can be seen from the above technical solutions, the core content of the present invention includes: designing a set of index system for influencing short-term load variation, and establishing an enterprise short-term load prediction model by adopting a random forest algorithm; based on the short-term load prediction result, designing basic electric charge optimization suggestions; comparing the whole industry level with the self electricity consumption condition of the enterprise, and designing an electricity degree and electricity fee optimization suggestion; and designing power factor adjustment electricity charge optimization suggestions according to the electricity consumption condition of the enterprise in the last several months.
Compared with the prior art, the invention has the following advantages and beneficial effects: the invention takes the electric charge of large industrial users as an access point, analyzes three factors influencing the basic electric charge, the electric power charge and the power factor adjustment electric charge of the electric charge expenditure of the users, researches the reason of overhigh average unit price of a single customer from the whole industry angle, and provides differentiated electricity consumption optimization suggestions by combining short-term load prediction results, thereby reducing the electricity consumption cost of industry enterprises, providing value-added service for the industry enterprises, and achieving the purpose of improving the viscosity of the users.
Drawings
FIG. 1 is a flow chart of the optimization method of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but embodiments of the present invention are not limited thereto.
Examples
Based on short-term load prediction, the invention designs power consumption optimization suggestions in industry and realizes power consumption optimization based on three factors of basic electric charge, electric power charge and power factor adjustment electric charge which influence the electric charge expenditure of a user; the method for optimizing the electricity consumption of the large industrial industry comprises three parts: the first part, based on the random forest short-term load prediction, selects a basic electric charge charging mode, and optimizes the basic electric charge; the second part is used for adjusting the production load based on the peak valley period of the transverse and longitudinal comparison analysis and optimizing the electricity charge; and the second part optimizes the power factor to adjust the electric charge based on the power factor adjustment of the industry standard comparison. The invention provides an electric charge optimization suggestion by using a system acquisition system (namely an electric charge acquisition system), a marketing application system, a production management system and the like to acquire data and adopting a random forest model and a comparison analysis method respectively, and comprises the following specific steps:
and the first part is used for selecting a basic electric charge charging mode based on random forest short-term load prediction and optimizing the basic electric charge.
1. Acquiring load data of monthly frequency of enterprises in large industrial industry and associated influence factor indexes, and acquiring a broad table of short-term load and factor indexes;
the method is implemented on a large data platform (ODPS), time and user number are used as key fields to carry out influence factor index splicing, the influence factor index change of each enterprise user is ensured to correspond to electricity load one by one, a data frame table containing month load index data and related factor index data of a large industrial enterprise is formed, and modeling index table data is pushed into a database (RDS).
2. Searching data and screening key influence factor indexes;
the step is realized on R analysis software (namely statistical analysis R software), and index data consistency check, anomaly identification and deletion filling are carried out, and key influence factor indexes influencing load change are screened out. Specifically, the R analysis software is connected with the database, reads modeling index table data, and performs the following data exploration and index screening:
consistency check: and carrying out consistency check on the load data occurrence time, wherein the load occurrence time is earlier than the current system time normally. And identifying the recorded abnormal load data with the occurrence time exceeding the system time by taking the system time as a critical value, regarding the abnormal load data as dirty data, directly removing and processing, and keeping the normal data recorded by the occurrence time of the load.
Abnormality identification: and identifying the abnormal value of the historical load by adopting a box diagram of a space method. And identifying abnormal values of the load by taking Q3+3/2 (Q3-Q1) and Q1-3/2 (Q3-Q1) as demarcation thresholds respectively. Defining outliers greater than Q3+3/2 (Q3-Q1) and less than Q1-3/2 (Q3-Q1), wherein Q1 is the smaller quartile and Q3 is the larger quartile. And load data identified as outliers is populated with historical contemporaneous averages.
Deletion filling: and filling the missing values of the associated influence factor indexes. The seasonal classified index (such as weather type in weather index) is filled with the missing value by adopting the synchronous value in the latest period, such as formula (1); seasonal numerical indicators (such as the highest month temperature in weather indicators), missing values are filled in by historical contemporaneous averages, as in formula (2); non-seasonal classified indexes, wherein the missing values are filled by index value modes, such as a formula (3); non-seasonal numerical indicators (e.g., business monthly output values) are replaced with an average load of the missing values occurring month-to-month 12-old, as in equation (4). The calculation formulas of the class 4 indexes of the related influence factors are respectively as follows:
x t ＝mean(x t-12 ，x t-24 ，x t-36 ，Λ，x t-12n ) (2)
x t ＝mode(x t-1 ，x t-2 ，x t-3 ，Λ，x t-n ) (3)
x t ＝mean(x t-1 ，x t-2 ，x t-3 ，Λ，x t-12 ) (4)
wherein x is t With x in the set which is not empty and (t-12 n) is the largest t-12n And (5) taking a value to replace.
Screening key influencing factor indexes: for the data type index, the key influence factor numerical type index screening is carried out by calculating a correlation coefficient matrix between variables, such as a formula (5); the classified indexes are subjected to discretization, and then chi-square inspection, such as a formula (6), is adopted to calculate the correlation between the discretized load indexes and the classified indexes, and key influence factor classified index screening is carried out.
In the formula 5, E XY, E X, E Y, var X, var Y represent the expected product of any two variables XY, the expected product of the variable X, the expected product of the variable Y, the variance of the variable X, the variance of the variable Y; in the formula (6), A is the actual observed value in the cross-linked list of the discretized load index and the classified index, and T is the theoretical value deduced by the hypothesis test in combination with the cross-linked list.
3. Establishing a random forest load prediction model to predict the short-term load demand of an enterprise;
the method is implemented on R analysis software, a random forest model for load prediction is established, and short-term load demands of enterprises are predicted, wherein the specific process is as follows:
randomly extracting 70% of samples as a test set, carrying out random forest model test, taking the remaining 30% of samples as a verification set, taking model prediction accuracy as an optimization target, and evaluating model performance. Firstly, putting the key influence factor indexes screened in the previous step into a random forest model for model test; then, model optimization is carried out in the modes of changing, eliminating and the like on key influence factor indexes continuously; and finally, reserving an index system with highest prediction accuracy of the random forest model, and predicting the short-term load demand of the enterprise.
4. Selecting a basic electric charge charging mode of an enterprise;
the method comprises the steps of realizing the steps on a big data platform, calculating the short-term (for example, the next month) load rate condition of an enterprise according to the short-term load demand prediction result of the enterprise and combining with the contract capacity information of the enterprise transformer, and recommending the enterprise to handle capacity reduction or suspension if the load rate is less than 40%; if the load rate > =75%, recommending the enterprise to adopt the capacity to charge the basic electricity fee; if the load rate > =100%, recommending the enterprise to handle capacity increasing; if the load rate is between 40% and 75%, and the load prediction accuracy of the enterprise in the last 3 months reaches 95% or more, recommending the enterprise to collect the basic electricity fee according to the maximum demand of the contract; if the load rate is between 40% and 75%, and the load prediction accuracy of the enterprise for 3 months is less than 95%, recommending the enterprise to collect the basic electricity fee according to the actual maximum demand. Wherein, the load factor index and the accuracy index are respectively represented by the following formulas (7) and (8):
wherein eta j For the load factor of enterprise j, W j C for predicting the month load of enterprise j j Contract capacity for enterprise j; MAPE (MAPE) j For the prediction accuracy of enterprise j, Y i Is the i-th actual load;
the i-th predicted load.
And the second part is produced based on peak valley time periods of transverse and longitudinal comparison analysis, and the load is adjusted.
1. The peak valley electricity consumption in the same industry is transversely compared and analyzed;
the step is realized on a big data platform, compared with the average electricity price of the same industry in the same period and the peak Gu Dianliang duty ratio, and according to the comparison between the average electricity price of the electricity of the enterprise for 3 months and the industry optimal result, a peak-to-valley ratio electric quantity optimization suggestion is given, and if the peak, peak and peak ratio of the electricity consumption of the near 3 periods are lower than the peak and peak ratio of the same industry in the local area, the suggestion is made to adjust the production period, the load of the valley section is improved, and the peak is staggered to fill the valley. Wherein, the average electricity price of electricity and peak Gu Dianliang ratio are respectively expressed as the following formulas (9) and (10):
R ji ＝p ji /TP j
(10)
wherein E is j Average electricity charge for enterprise j, e ji For the electricity charge of different time periods of enterprise j, i respectively represents peak, peak and valley periods, TP j The total electric quantity of the enterprise j; r is R ji For the electric quantity duty ratio of different time periods of enterprise j, p ji Power is supplied to different time periods of enterprise j.
2. Longitudinally comparing historical peak average loads of enterprises;
the step is formed on a big data platform, average production load conditions of peak time periods, peak time periods and valley time periods of enterprises in historical period 3 are longitudinally compared, peak-to-valley specific electric quantity optimization suggestions are given, and if the peak average load and the peak average load of the enterprises in the period 3 are both larger than the average load of the valley time periods, the production time periods are suggested to be adjusted, the load of the valley time periods is improved, and peak shifting and valley filling are carried out. Wherein, the average load formula of different time periods is as follows:
AL ji ＝p ji /t ji
(11)
wherein, AL ji For different time periods average load of enterprise j, p ji For the electric quantity of different time periods of enterprise j, i respectively represents peak, peak and valley, t ji For the duration of enterprise j in peak, valley periods.
And the third part is used for adjusting the power factor based on industry comparison analysis.
The step is formed on a big data platform, the enterprise is compared and analyzed for 3 months, the industry optimal power factor standard electric charge and the actual power factor electric charge are compared, a power factor adjustment suggestion is given, if the power factor of the user is lower than the optimal standard 0.95 for 3 months, the power factor is suggested to be improved, and the power factor electric charge is reduced.
Combining the three parts, namely the short-term electricity consumption optimization suggestion for an enterprise.
The above examples are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above examples, and any other changes, modifications, substitutions, combinations, and simplifications that do not depart from the spirit and principle of the present invention should be made in the equivalent manner, and the embodiments are included in the protection scope of the present invention.
Claims (3)
1. The electricity utilization optimization method for the industry in combination with short-term load prediction is characterized by comprising the following steps of:
s1, selecting a basic electric charge charging mode based on random forest short-term load prediction, and optimizing the basic electric charge;
s2, peak valley period production based on transverse and longitudinal comparison analysis is carried out, and load is adjusted;
s3, adjusting the power factor based on industry comparison analysis; the step is formed on a big data platform, and the optimal power factor standard electric charge and the actual power factor electric charge in the industry are compared and analyzed to give a power factor adjustment suggestion;
the step S1 comprises the following steps:
s11, acquiring load data of monthly frequency of enterprises in large industrial industry and related influence factor indexes, and obtaining a broad table of short-term load and factor indexes;
s12, searching data and screening key influence factor indexes;
s13, establishing a random forest load prediction model, and predicting the short-term load demand of an enterprise;
s14, selecting a basic electric charge charging mode of the enterprise according to a prediction result of the short-term load demand of the enterprise;
the step S13 includes: firstly, putting the key influence factor indexes screened in the previous step into a random forest model for model test; then, continuously transforming and eliminating key influence factor indexes to optimize the model; finally, reserving an index system with highest prediction accuracy of the random forest model, and predicting the short-term load demand of the enterprise;
step S14 includes: calculating the short-term load rate condition of an enterprise according to the short-term load demand prediction result of the enterprise and combining with the contract capacity information of the enterprise transformer, and recommending the enterprise to handle volume reduction or suspension if the load rate is less than 40%; if the load rate > =75%, recommending the enterprise to adopt the capacity to charge the basic electricity fee; if the load rate > =100%, recommending the enterprise to handle capacity increasing; if the load rate is between 40% and 75%, and the load prediction accuracy of the enterprise in the last 3 months reaches 95% or more, recommending the enterprise to collect the basic electricity fee according to the maximum demand of the contract; if the load rate is between 40% and 75%, and the load prediction accuracy of the enterprise in the last 3 months is less than 95%, recommending the enterprise to collect the basic electricity fee according to the actual maximum demand;
step S12 includes: index data consistency checking, anomaly identification and deletion filling, and screening out key influence factor indexes influencing load change;
the consistency check is as follows: consistency check is carried out on the occurrence time of the load data, the occurrence time is taken as a critical value, the recorded abnormal load data with the occurrence time exceeding the system time is identified, the abnormal load data is regarded as dirty data, the dirty data is directly removed and processed, and the normal data recorded in the occurrence time of the load is reserved;
the anomaly identification is as follows: identifying abnormal values of the historical load by adopting a box diagram of a space method, identifying the abnormal values of the load, and filling load data identified as the abnormal values with a historical synchronous average value;
the deletion filling is as follows: filling the missing values of the associated influence factor indexes; wherein, the missing value of the seasonal classified index is filled by the synchronous value in the latest period; seasonal numerical indicators, the missing values of which are filled with historical contemporaneous averages; non-seasonal classified indexes, wherein the missing values are filled by index value modes; a non-seasonal numerical indicator, replacing with an average load of the first 12 months of occurrence of the missing value;
the screening process of the key influence factor index comprises the following steps: the method comprises the steps of screening the data type indexes by calculating a correlation coefficient matrix among variables and carrying out the numerical type index screening of key influence factors; the method comprises the steps of discretizing a load index, calculating the correlation between the discretized load index and a classified index by adopting chi-square test, and screening the classified index of key influencing factors.
2. The electricity optimization method for the industry of large scale according to claim 1, wherein the formula of the inter-variable correlation coefficient matrix is:
the formula of chi-square test is:
wherein E < XY >, E < X >, E < Y >, var < X >, var < Y > represent the expected product of any two variables XY, the expected product of the variable X, the expected product of the variable Y, the variance of the variable X, and the variance of the variable Y, respectively; a is the actual observed value in the cross-linked list of the discretized load index and the classified index, and T is the theoretical value deduced by hypothesis testing in combination with the cross-linked list.
3. The electricity optimization method for large industrial industries according to claim 1, wherein step S2 comprises:
s21, transversely comparing and analyzing peak valley electricity consumption in the same industry; comparing average electricity price and peak Gu Dianliang duty ratio of the same industry in the same period, and according to comparison between average electricity price of the enterprise for a few months and industry optimal result, giving peak-to-valley ratio electric quantity optimization suggestion, if the peak, peak and proportion of electricity consumption in the near period are lower than the peak and peak proportion of the same industry in the local area, suggesting to adjust production period, improving load of valley section and filling peak and valley with staggering;
s22, longitudinally comparing historical peak-to-valley average loads of enterprises; longitudinally comparing the peak time, peak time and valley time average production load conditions of enterprises in historic periods, giving peak-to-valley specific electric quantity optimization suggestions, and if the peak average load and peak average load of the enterprises in recent periods are larger than the valley time average load, suggesting to adjust the production time, improving the valley time load and staggering peak and valley filling.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910575725.6A CN110298509B (en) | 2019-06-28 | 2019-06-28 | Large industrial industry electricity utilization optimization method combined with short-term load prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910575725.6A CN110298509B (en) | 2019-06-28 | 2019-06-28 | Large industrial industry electricity utilization optimization method combined with short-term load prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110298509A CN110298509A (en) | 2019-10-01 |
| CN110298509B true CN110298509B (en) | 2023-05-23 |
Family
ID=68029457
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910575725.6A Active CN110298509B (en) | 2019-06-28 | 2019-06-28 | Large industrial industry electricity utilization optimization method combined with short-term load prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110298509B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111008726B (en) * | 2019-10-28 | 2023-08-29 | 武汉理工大学 | Class picture conversion method in power load prediction |
| CN112734137B (en) * | 2021-01-27 | 2022-12-16 | 国网电力科学研究院武汉能效测评有限公司 | Short-term building power load prediction method and system based on intelligent electric meter data |
| CN112950408A (en) * | 2021-03-30 | 2021-06-11 | 为式(苏州)智能科技有限公司 | Energy management method, system and device with electric charge optimization function and storage medium |
| CN113095595A (en) * | 2021-05-06 | 2021-07-09 | 广东鹰视能效科技有限公司 | Energy efficiency optimization method and system based on power distribution operation and maintenance |
| CN113080986B (en) * | 2021-05-07 | 2022-11-15 | 中国科学院深圳先进技术研究院 | Method and system for detecting exercise fatigue based on wearable equipment |
| CN113610174A (en) * | 2021-08-13 | 2021-11-05 | 中南大学 | Power grid host load prediction method, equipment and medium based on Phik feature selection |
| CN115545569B (en) * | 2022-11-28 | 2023-08-01 | 国网江苏省电力有限公司营销服务中心 | Peak electricity price mechanism implementation effect evaluation method based on quantitative fee analysis |
| CN116562507A (en) * | 2023-07-05 | 2023-08-08 | 广东鹰视能效科技有限公司 | Electric bill analysis method and system |
| CN117495056A (en) * | 2023-12-28 | 2024-02-02 | 西安民为电力科技有限公司 | Power consumption data monitoring and optimizing method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105303262A (en) * | 2015-11-12 | 2016-02-03 | 河海大学 | Short period load prediction method based on kernel principle component analysis and random forest |
| CN108320053A (en) * | 2018-01-23 | 2018-07-24 | 国网冀北电力有限公司经济技术研究院 | A kind of region electricity demand forecasting method, apparatus and system |
| CN108520306A (en) * | 2018-03-23 | 2018-09-11 | 华翔翔能（湖南）能源科技有限公司 | Energy management method |
| JP2018191505A (en) * | 2017-04-28 | 2018-11-29 | ダイキン工業株式会社 | Power supply power factor control system, phase modifier and active filter device |
| CN109727066A (en) * | 2018-12-27 | 2019-05-07 | 浙江华云信息科技有限公司 | A kind of big industrial electricity consumers load forecasting method based on XGBoost algorithm |
-
2019
- 2019-06-28 CN CN201910575725.6A patent/CN110298509B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105303262A (en) * | 2015-11-12 | 2016-02-03 | 河海大学 | Short period load prediction method based on kernel principle component analysis and random forest |
| JP2018191505A (en) * | 2017-04-28 | 2018-11-29 | ダイキン工業株式会社 | Power supply power factor control system, phase modifier and active filter device |
| CN108320053A (en) * | 2018-01-23 | 2018-07-24 | 国网冀北电力有限公司经济技术研究院 | A kind of region electricity demand forecasting method, apparatus and system |
| CN108520306A (en) * | 2018-03-23 | 2018-09-11 | 华翔翔能（湖南）能源科技有限公司 | Energy management method |
| CN109727066A (en) * | 2018-12-27 | 2019-05-07 | 浙江华云信息科技有限公司 | A kind of big industrial electricity consumers load forecasting method based on XGBoost algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110298509A (en) | 2019-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110298509B (en) | Large industrial industry electricity utilization optimization method combined with short-term load prediction | |
| Hsu et al. | Forecasting integrated circuit output using multivariate grey model and grey relational analysis | |
| Joskow | The determination of the allowed rate of return in a formal regulatory hearing | |
| CN109165763A (en) | A kind of potential complained appraisal procedure and device of 95598 customer service work order | |
| CN111260198A (en) | Method and system for judging degree of rationality of line loss in transformer area synchronization and terminal equipment | |
| CN112308462A (en) | Power consumer classification method and device | |
| CN109272140A (en) | A kind of transformer equipment runtime forecasting of cost method based on big data analysis | |
| Guo et al. | Designing a sustainable-remanufacturing closed-loop supply chain under hybrid uncertainty: Cross-efficiency sorting multi-objective optimization | |
| CN105678414A (en) | Data processing method of predicting resource consumption | |
| CN111401804A (en) | Simulation-based engineering material supply chain network planning method and system | |
| CN111178957B (en) | Method for early warning sudden increase of electric quantity of electricity consumption customer | |
| CN112561364A (en) | Power grid operation and maintenance cost level rationality evaluation technology based on combination weight | |
| CN113469595A (en) | Intelligent supply chain system and server platform | |
| CN111105218A (en) | Power distribution network operation monitoring method based on holographic image technology | |
| CN114298538A (en) | Investment scheme evaluation method, system and storage medium for power grid retail project | |
| CN111967684B (en) | Metering asset active distribution method based on big data analysis | |
| CN112734274A (en) | Mining and comprehensive evaluation method for low-carbon power grid operation leading influence factors | |
| CN116912034A (en) | Power distribution network engineering cost data splitting method and auxiliary computing system | |
| CN109255470A (en) | A kind of non-item class forecasting of cost method of transformer equipment based on big data analysis | |
| CN115689386A (en) | Enterprise operation monitoring method based on knowledge graph and FP (Fabry-Perot) growth algorithm | |
| CN112001551B (en) | Ground and commercial power grid sales electricity quantity prediction method based on large-user electricity quantity information | |
| CN114139770A (en) | Metal industry economic estimation system and method based on Solo growth and stock recursion | |
| CN111797924B (en) | Three-dimensional garden portrait method and system based on clustering algorithm | |
| CN114254806A (en) | Power distribution network heavy overload early warning method and device, computer equipment and storage medium | |
| Li et al. | Validation techniques and performance metrics for loss given default models |
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 |