CN113159394A - Intelligent power distribution network partition cold and hot load prediction method and device - Google Patents
Intelligent power distribution network partition cold and hot load prediction method and device Download PDFInfo
- Publication number
- CN113159394A CN113159394A CN202110344348.2A CN202110344348A CN113159394A CN 113159394 A CN113159394 A CN 113159394A CN 202110344348 A CN202110344348 A CN 202110344348A CN 113159394 A CN113159394 A CN 113159394A
- Authority
- CN
- China
- Prior art keywords
- cold
- load
- distribution network
- characteristic
- heat
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- 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
Abstract
The invention belongs to the technical field of cold and hot load prediction, and particularly relates to a method and a device for predicting a partitioned cold and hot load of an intelligent power distribution network. And acquiring characteristic parameter data corresponding to the optimal characteristic set of the intelligent distribution network partition, and inputting the characteristic parameter data into the trained heat/cold load prediction model to obtain a heat/cold load result of the intelligent distribution network partition. The method comprises the steps of firstly processing characteristic parameters by using a characteristic engineering method, removing the characteristic parameters with low heat load/cold load correlation, avoiding the reduction of model prediction precision caused by redundant information caused by multiple collinearity, further carrying out full arrangement on the selected characteristic parameters with high correlation to obtain a plurality of characteristic sets, determining the precision of the prediction model when the characteristic sets are respectively used as the prediction model, finding out the optimal characteristic set and the corresponding prediction model, and effectively improving the accuracy of the prediction model for the heat/cold load of the intelligent power distribution network.
Description
Technical Field
The invention belongs to the technical field of cold and hot load prediction, and particularly relates to a method and a device for predicting a partitioned cold and hot load of an intelligent power distribution network.
Background
The cold and heat load prediction is a main basis of economic dispatching, is the basis of power production planning, is necessary data for smooth development of a power market, is one of main factors of power system safety analysis, is an important means for realizing the target energy-saving control of the transformer, is an important aspect for realizing scientific management and dispatching of a power grid, and plays an important role in power system planning and power grid operation.
With the gradual establishment of the power market and the development of the power market and the intelligent power distribution network, the cold and heat load prediction is more and more emphasized, and the requirement on the load prediction level is increased day by day. The improvement of the load prediction technical level is beneficial to planning power utilization management, reducing energy consumption and generating cost, reasonably arranging the operation mode of a power grid and establishing a unit maintenance plan, and improving the economic benefit and social benefit of a power system. Therefore, the level of cold and heat load prediction has become one of the significant indicators for measuring whether the management of a power enterprise moves toward modernization.
The traditional cold and heat load index calculation method is characterized in that the cold and heat load indexes of typical buildings of different types are directly multiplied by the areas of the corresponding buildings, so that the cold and heat loads of the buildings of various types are obtained, and then the cold and heat loads of the buildings of various types are directly added to obtain the cold and heat loads of a certain partition. The cold and heat load index used in the method is generally determined through experience, and the experience and the actual situation generally have a difference, so that the cold and heat load prediction is inaccurate.
Disclosure of Invention
The invention provides a method and a device for forecasting a partitioned cold and hot load of an intelligent power distribution network, which are used for solving the problem that a method in the prior art is inaccurate in forecasting the cold and hot load.
In order to solve the technical problems, the technical scheme and the corresponding beneficial effects of the technical scheme are as follows:
the invention provides a method for forecasting the partition cold and hot load of an intelligent power distribution network, which comprises the following steps:
acquiring characteristic parameter data corresponding to the optimal characteristic set of the heat/cold load prediction model of the intelligent power distribution network partition, and inputting the characteristic parameter data into the trained heat/cold load prediction model to obtain a heat/cold load result of the intelligent power distribution network partition; the heat/cold load prediction model is constructed and trained by adopting the following method:
selecting characteristic parameters with high correlation with the heat load from the selected characteristic parameters by using a characteristic engineering method;
the selected characteristic parameters with high correlation are arranged completely to obtain a plurality of characteristic sets;
and respectively taking each feature set as input and taking the heat/cold load as output to construct a plurality of prediction models, training the constructed prediction models by utilizing the acquired feature parameter data and the heat/cold load data, comparing the prediction precision of each prediction model after training, taking the feature set with the highest prediction precision as an optimal feature set, and taking the prediction model corresponding to the optimal feature set as the heat/cold load prediction model.
The beneficial effects of the above technical scheme are: the method comprises the steps of firstly processing characteristic parameters by using a characteristic engineering method, removing the characteristic parameters with low heat load/cold load correlation, effectively eliminating potential multiple collinearity among parameter variables, avoiding reduction of model prediction precision caused by redundant information caused by the multiple collinearity, further carrying out full arrangement on the selected characteristic parameters with high correlation to obtain a plurality of characteristic sets, determining the precision of the prediction model when the characteristic sets are respectively used as the prediction models to be input, finding out the optimal characteristic set and the corresponding prediction model, and effectively improving the accuracy of the prediction model for the cold and heat loads of the intelligent power distribution network.
Further, the selected characteristic parameters include: meteorological parameter variables, power consumption, distribution network load, plot area, and planned energy usage time.
Further, after the heat/cold load prediction model is obtained, the method further comprises the step of utilizing the training set to train the heat/cold load prediction model again.
Further, the prediction model is a neural network model.
Furthermore, a characteristic engineering method corresponding to the thermal load prediction model is a correlation analysis method.
Furthermore, the characteristic engineering method corresponding to the cold load prediction model is a wiener filtering method.
Further, the method also comprises the step of carrying out data cleaning and/or normalization processing on the characteristic parameter data.
The invention also provides a device for forecasting the partition cold and hot load of the intelligent power distribution network, which comprises a memory and a processor, wherein the processor is used for executing instructions stored in the memory to realize the method for forecasting the partition cold and hot load of the intelligent power distribution network introduced above and achieve the same beneficial effects as the method.
Drawings
FIG. 1 is a flow chart of a method for forecasting the partitioned cold and hot loads of the intelligent power distribution network;
fig. 2 is a block diagram of the device for predicting the cooling and heating loads of the zones of the intelligent distribution network according to the present invention.
Detailed Description
The method comprises the following steps:
the flow of the embodiment of the method for predicting the cold and heat loads of the partition of the intelligent power distribution network is shown in fig. 1, and the cold load and the heat load are calculated separately by adopting different prediction models.
The heat load calculation will be described in detail below.
Step one, taking the optimal feature set as input, taking the heat load as output, and constructing a heat load prediction model based on a neural network model. The specific process comprises the following steps:
1. acquiring heat load data of a plurality of groups of distribution network partitions and characteristic parameter data influencing the heat load of the distribution network, wherein the five characteristic parameter data are respectively meteorological parameter variables (such as outdoor temperature, humidity, wind power, air pressure, solar radiation intensity and the like), power consumption, distribution network load, land area and planned energy utilization time.
2. And preprocessing the acquired parameter data, wherein the preprocessing comprises data cleaning and normalization processing.
Data cleaning: and eliminating wrong values and filling up missing data through data cleaning.
Normalization treatment: listing the maximum value Max and the minimum value Min of the obtained characteristic parameter data, enabling the normalized minimum characteristic parameter data to be 0, the normalized maximum characteristic parameter data to be 1, compressing all the characteristic parameter data in the range of [0,1], and enabling the normalized characteristic parameter data to be: and R is (i-Min)/(Max-Min), and i is characteristic parameter data before normalization processing. This process eliminates the dimension and makes the mathematical units of the data consistent.
3. When testing load, under the influence of various accidental factors in the test instrument record, manual operation or production environment, the superposition of noise is brought to monitored time sequence data, so that load signal identification is difficult. Therefore, the cold and hot load data is composed of two parts, one part is data determined by the fluctuation rule, and the other part is data of nonlinear fluctuation generated by noise interference. Therefore, the correlation between the five characteristic parameters and the heat load is evaluated by adopting a correlation analysis method, and a correlation coefficient used for representing the high and low correlation between each characteristic parameter and the heat load is obtained, wherein the correlation coefficient is greater than 0 to represent positive correlation, the correlation coefficient is less than 0 to represent negative correlation, and the correlation coefficient is equal to 0 to represent no correlation. And selecting the characteristic parameter with high heat load correlation from the five characteristic parameters according to the correlation coefficient of each characteristic parameter. The threshold value set in this embodiment is 0.3, and finally, it is found that the correlation coefficients of these five parameters are all greater than 0.3, so that the characteristic parameters which are finally screened out and have high correlation with the heat load are meteorological parameter variables, power consumption, distribution network load, land area, and planned energy consumption time.
4. And (3) fully arranging the selected characteristic parameters with higher correlation, and finally constructing a basic characteristic set comprising five characteristic parameters with higher correlation and 30 characteristic subsets, wherein the characteristic subsets comprise 1, 2, 3 or 4 characteristic parameters with higher correlation. The base feature set and the 30 feature subsets are collectively referred to as a feature set, and the number of feature sets is 31.
5. And respectively taking each feature set as input and taking the heat load as output to construct 31 neural network models, training the constructed neural network models by using the acquired feature parameter data and the heat load data, comparing the prediction precision of each neural network model after training, taking the feature set with the highest prediction precision as an optimal feature set, and taking the neural network model corresponding to the optimal feature set as the heat load prediction model.
And step two, taking the characteristic parameter data corresponding to the obtained optimal feature set and the corresponding heat load data as a training set, training a heat load prediction model, inputting the obtained predicted value into a loss function calculation loss index, comparing the difference between the predicted value and an actual measurement value in the training set, inputting the calculation result of the loss function into an optimizer to update a weight error, reducing the adoption rate of error data by the learning capacity of a neural network algorithm, repeating iteration until a specified iteration step number is reached or the loss function value is lower than a threshold value, and obtaining the optimal parameter configuration under the specific hyper-parameter configuration. And testing the loss functions under different super-parameter configurations through the verification set, and selecting the corresponding parameters and super-parameter configurations when the loss functions are minimum to obtain the trained heat load prediction model.
And step three, after the trained thermal load prediction model is obtained, acquiring the characteristic parameter data of the intelligent distribution network partition corresponding to the real-time optimal characteristic set, and inputting the characteristic parameter data into the trained thermal load prediction model to obtain the thermal load result of the intelligent distribution network partition.
The cooling load calculation will be described in detail below.
Step one, the selected optimal feature set is used as input, the cold load is used as output, and a cold load prediction model is constructed based on the neural network model. The specific process comprises the following steps:
1. acquiring cold load data of a plurality of groups of distribution network partitions and characteristic parameter data influencing the cold load of the distribution network, wherein the five characteristic parameter data are respectively meteorological parameter variables (such as outdoor temperature, humidity, wind power, air pressure, solar radiation intensity and the like), power consumption, distribution network load, land area and planned energy utilization time.
2. And preprocessing the acquired parameter data, wherein the preprocessing comprises data cleaning and normalization processing.
Data cleaning: and eliminating wrong values and filling up missing data through data cleaning.
Normalization treatment: listing the maximum value Max and the minimum value Min of the obtained characteristic parameter data, enabling the normalized minimum characteristic parameter data to be 0, the normalized maximum characteristic parameter data to be 1, compressing all the characteristic parameter data in the range of [0,1], and enabling the normalized characteristic parameter data to be: and R is (i-Min)/(Max-Min), and i is characteristic parameter data before normalization processing. This process eliminates the dimension and makes the mathematical units of the data consistent.
3. When testing load, under the influence of various accidental factors in the test instrument record, manual operation or production environment, the superposition of noise is brought to monitored time sequence data, so that load signal identification is difficult. Therefore, the cold and hot load data is composed of two parts, one part is data determined by the fluctuation rule, and the other part is data of nonlinear fluctuation generated by noise interference. Therefore, a filtering algorithm (for example, a wiener filtering algorithm) is adopted to perform noise reduction processing on the five characteristic parameter data, suppress and prevent interference, remove useless influence parameters, evaluate the correlation between the five characteristic parameters and the cold load by using a filtering method, and select the characteristic parameters with high correlation with the cold load from the five characteristic parameters. The characteristic parameters which are finally screened out and have high correlation with the cold load in the embodiment are meteorological parameter variables, power consumption, distribution network load, land area and planned energy consumption time.
4. And (3) fully arranging the selected characteristic parameters with higher correlation, and finally constructing a basic characteristic set comprising five characteristic parameters with higher correlation and 30 characteristic subsets, wherein the characteristic subsets comprise 1, 2, 3 or 4 characteristic parameters with higher correlation. The base feature set and the 30 feature subsets are collectively referred to as a feature set, and the number of feature sets is 31.
5. And taking each feature set as input and cold load as output to construct 31 neural network models, training the constructed neural network models by using the acquired feature parameter data and cold load data, comparing the prediction precision of each neural network model after training, taking the feature set with the highest prediction precision as an optimal feature set, and taking the neural network model corresponding to the optimal feature set as the cold load prediction model.
And step two, taking the characteristic parameter data corresponding to the obtained optimal feature set and the corresponding cold load data as a training set, training a cold load prediction model, inputting the obtained predicted value into a loss function to calculate a loss index, comparing the difference between the predicted value and an actual measurement value in the training set, inputting a loss function calculation result into an optimizer to update a weight error, reducing the adoption rate of error data by the learning capacity of a neural network algorithm, and repeating iteration until a specified iteration step number is reached or the loss function value is lower than a threshold value, so as to obtain the optimal parameter configuration under the specific hyper-parameter configuration. And testing the loss functions under different super-parameter configurations through the verification set, and selecting the corresponding parameter and super-parameter configuration when the loss function is the minimum value to obtain the trained cold load prediction model.
And step three, after the trained cold load prediction model is obtained, acquiring the characteristic parameter data of the intelligent power distribution network partition corresponding to the real-time optimal characteristic set, and inputting the characteristic parameter data into the trained cold load prediction model to obtain the cold load result of the intelligent power distribution network partition.
It should be noted that, in the present embodiment, both step 5 and step two of the thermal load prediction step have model training processes. And step 5 in the first step is only rough training to find the feature set with the highest prediction precision, and after the feature set with the highest prediction precision is found, the selected heat load prediction model is trained again by using the training sets with more data sets to optimize parameters in the heat load prediction model so as to obtain a better heat load prediction model. Namely, the process of the step two is a process for making the heat load prediction model more optimal. In addition, the same applies to the process principle of cold load prediction.
Moreover, the invention takes longer in the early prediction model to find the highest-precision heat/cold load prediction model. The final prediction models may be different for different regions, for example, the optimal feature set of the final thermal load prediction model for a certain region is < distribution network load, area of a land, power consumption >, and the optimal feature set of the final thermal load prediction model for another region is < gas phase variable parameter, power consumption >, etc., but whichever region, whether thermal load or cold load, is selected from five feature parameters of meteorological parameter variables, power consumption, distribution network load, area of a land, and planned energy utilization time.
In conclusion, no matter the heat load prediction or the cold load prediction is carried out, the characteristic engineering method (the heat load prediction is a correlation analysis method, and the cold load prediction is a wiener filtering method) is firstly utilized to process the characteristic parameters, so that the potential multiple collinearity among parameter variables is effectively eliminated, the model prediction precision reduction caused by redundant information brought by the multiple collinearity is avoided, the acquired parameter data of the intelligent distribution network can be converted into the characteristic set matched with the prediction model, and the accuracy of the prediction model on the cold and heat loads of the intelligent distribution network is effectively improved. In addition, according to the influence degree of each characteristic parameter on the prediction precision of the prediction model, the parameter variable with small influence degree is deleted, so that the optimal characteristic set can be selected according to the type and precision requirements of the prediction model, the calculation speed can be effectively increased, the prediction instantaneity is ensured, and the model prediction precision can be improved.
The embodiment of the device is as follows:
an embodiment of the device for predicting the partitioned cold and hot loads of the intelligent distribution network, disclosed by the invention, is shown in fig. 2 and comprises a memory, a processor and an internal bus, wherein the processor and the memory are communicated and interacted with each other through the internal bus. The memory comprises at least one software functional module stored in the memory, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, so as to realize the method for predicting the cold and hot loads of the partitions of the intelligent power distribution network, which is introduced in the method embodiment of the invention.
The processor can be a microprocessor MCU, a programmable logic device FPGA and other processing devices.
The memory can be various memories for storing information by using an electric energy mode, such as RAM, ROM and the like; various memories for storing information by magnetic energy, such as hard disk, floppy disk, magnetic tape, magnetic core memory, bubble memory, U disk, etc.; various memories for storing information optically, such as CDs, DVDs, etc.; of course, other forms of memory are possible, such as quantum memory, graphene memory, and the like.
Claims (8)
1. A method for forecasting the partition cold and hot load of an intelligent power distribution network is characterized by comprising the following steps:
acquiring characteristic parameter data corresponding to the optimal characteristic set of the heat/cold load prediction model of the intelligent power distribution network partition, and inputting the characteristic parameter data into the trained heat/cold load prediction model to obtain a heat/cold load result of the intelligent power distribution network partition; the heat/cold load prediction model is constructed and trained by adopting the following method:
selecting characteristic parameters with high correlation with the heat load from the selected characteristic parameters by using a characteristic engineering method;
the selected characteristic parameters with high correlation are arranged completely to obtain a plurality of characteristic sets;
and respectively taking each feature set as input and taking the heat/cold load as output to construct a plurality of prediction models, training the constructed prediction models by utilizing the acquired feature parameter data and the heat/cold load data, comparing the prediction precision of each prediction model after training, taking the feature set with the highest prediction precision as an optimal feature set, and taking the prediction model corresponding to the optimal feature set as the heat/cold load prediction model.
2. The method for forecasting the partitioned cold and hot loads of the intelligent power distribution network according to claim 1, wherein the selected characteristic parameters comprise: meteorological parameter variables, power consumption, distribution network load, plot area, and planned energy usage time.
3. The method for forecasting the partitioned cold and hot loads of the intelligent power distribution network according to claim 1, further comprising the step of training the hot/cold load forecasting model again by using a training set after the hot/cold load forecasting model is obtained.
4. The method for forecasting the partitioned cold and hot load of the intelligent power distribution network according to claim 1, wherein the forecasting model is a neural network model.
5. The method for forecasting the partitioned cold and hot loads of the intelligent power distribution network according to claim 1, wherein a characteristic engineering method corresponding to the heat load forecasting model is a correlation analysis method.
6. The method for forecasting the partitioned cold and hot loads of the intelligent power distribution network according to claim 1, wherein a characteristic engineering method corresponding to the cold load forecasting model is a wiener filtering method.
7. The intelligent power distribution network partition cold and heat load prediction method according to any one of claims 1 to 6, further comprising the step of performing data cleaning and/or normalization processing on the characteristic parameter data.
8. The device for forecasting the cold and hot load of the partition of the intelligent power distribution network is characterized by comprising a memory and a processor, wherein the processor is used for executing instructions stored in the memory to realize the method for forecasting the cold and hot load of the partition of the intelligent power distribution network according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110344348.2A CN113159394A (en) | 2021-03-30 | 2021-03-30 | Intelligent power distribution network partition cold and hot load prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110344348.2A CN113159394A (en) | 2021-03-30 | 2021-03-30 | Intelligent power distribution network partition cold and hot load prediction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113159394A true CN113159394A (en) | 2021-07-23 |
Family
ID=76885982
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110344348.2A Pending CN113159394A (en) | 2021-03-30 | 2021-03-30 | Intelligent power distribution network partition cold and hot load prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113159394A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704875A (en) * | 2017-09-30 | 2018-02-16 | 山东建筑大学 | Based on the building load Forecasting Methodology and device for improving IHCMAC neutral nets |
| CN110619389A (en) * | 2019-09-23 | 2019-12-27 | 山东大学 | Load prediction method and system of combined cooling heating and power system based on LSTM-RNN |
| CN111639823A (en) * | 2020-06-10 | 2020-09-08 | 天津大学 | Building cold and heat load prediction method constructed based on feature set |
| CN112001439A (en) * | 2020-08-19 | 2020-11-27 | 西安建筑科技大学 | GBDT-based shopping mall building air conditioner cold load prediction method, storage medium and equipment |
-
2021
- 2021-03-30 CN CN202110344348.2A patent/CN113159394A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704875A (en) * | 2017-09-30 | 2018-02-16 | 山东建筑大学 | Based on the building load Forecasting Methodology and device for improving IHCMAC neutral nets |
| CN110619389A (en) * | 2019-09-23 | 2019-12-27 | 山东大学 | Load prediction method and system of combined cooling heating and power system based on LSTM-RNN |
| CN111639823A (en) * | 2020-06-10 | 2020-09-08 | 天津大学 | Building cold and heat load prediction method constructed based on feature set |
| CN112001439A (en) * | 2020-08-19 | 2020-11-27 | 西安建筑科技大学 | GBDT-based shopping mall building air conditioner cold load prediction method, storage medium and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109659933B (en) | Electric energy quality prediction method for power distribution network with distributed power supply based on deep learning model | |
| CN111695731B (en) | Load prediction method, system and equipment based on multi-source data and hybrid neural network | |
| CN109376960A (en) | Load Forecasting based on LSTM neural network | |
| CN107274067B (en) | Distribution transformer overload risk assessment method | |
| CN102063550A (en) | Intelligent design system of cold extrusion piece with machine intelligence involved design decision | |
| CN105138717A (en) | Transformer state evaluation method by optimizing neural network with dynamic mutation particle swarm | |
| CN113656938A (en) | Monitoring method, control device and storage medium for distribution network transformer | |
| CN108121215B (en) | Process control loops method of evaluating performance and device based on full loop reconstruct emulation | |
| CN104037943A (en) | Method and system for monitoring voltage and capable of improving power grid voltage quality | |
| CN111680841B (en) | Short-term load prediction method, system and terminal equipment based on principal component analysis | |
| CN113159341A (en) | Power distribution network aid decision-making method and system integrating deep reinforcement learning and expert experience | |
| CN115099594A (en) | Building energy consumption management method and system based on Internet of things | |
| CN111445065A (en) | Energy consumption optimization method and system for refrigeration group control of data center | |
| Xu et al. | Interval prediction method for wind power based on VMD-ELM/ARIMA-ADKDE | |
| CN116826745B (en) | Layered and partitioned short-term load prediction method and system in power system background | |
| CN117634678A (en) | Low-carbon park carbon emission prediction method based on actual operation scene | |
| CN112363012A (en) | Power grid fault early warning device and method | |
| CN113159394A (en) | Intelligent power distribution network partition cold and hot load prediction method and device | |
| CN116706992A (en) | Self-adaptive power prediction method, device and equipment for distributed photovoltaic cluster | |
| CN115864644A (en) | Relay protection device state evaluation method, system, equipment and medium | |
| CN111756031A (en) | Power grid operation trend estimation method and system | |
| CN114840591A (en) | Method and device for determining sectional switch power data | |
| CN114066068A (en) | Short-term power load prediction method, device, equipment and storage medium | |
| CN114091325A (en) | Power distribution station load prediction method and system based on multi-model layered learning | |
| CN106777494B (en) | Method for calculating sensitivity of reliability influence factors of power system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210723 |
|
| RJ01 | Rejection of invention patent application after publication |