68 results about "Electricity demand forecasting" patented technology

A method and system for defining and optimizing demand response events is disclosed where at least one initial input parameter is received for selection of a demand response event, a power demand forecast is retrieved, a demand response event is automatically calculated based on the at least one initial input parameter and the power demand forecast, and an interactive user interface, the interactive user interface is generated and includes the power demand forecast, an expected power capacity forecast, and the demand response event.

The present invention relates to a kind of LSSVM annual power consumption prediction method based on antlion optimization, the annual power consumption prediction method comprises the following steps: determine the input variable of least square support vector machine (Least Square Support Vector Machines, LSSVM) prediction model; Initialize the antlion optimization algorithm; calculate the fitness value of the initial antlion to obtain the initial elite antlion; update the position of the ant, calculate the fitness value of the current antlion, and compare it with the fitness value of the corresponding antlion to determine whether to update the antlion Lion position; compare the fitness value of the antlion after the updated position with the fitness value of the previous elite antlion one by one, keep the antlion corresponding to the smaller fitness value, and obtain the elite antlion of this iteration; judge whether The maximum number of iterations is reached, if yes, then output the location of the elite antlion and the corresponding predicted value of annual electricity consumption, if no, continue to iterate. Compared with the prior art, the present invention has the advantages of higher prediction accuracy and higher prediction efficiency.

A demand-supply planning device includes: a power demand predicting unit that predicts power demand; an expected rebate value predicting unit that calculates an expected profit value per unit amount of reduced power based on an occurrence probability of a demand response and a rebate value to be acquired through the demand response; and an optimal operation plan creating unit that sets an addition result of a first cost for power purchase, a second cost for power generation in power generators and storage batteries, and a multiplication result of the expected profit value and reserve power, which is an amount of power capable of being further generated by the power generators and the storage batteries after an amount of power supplied satisfying the power demand is generated, as an evaluation function and determines an operation plan of the reserve power and the power generators and the storage batteries.

A weather prediction unit calculates a local numerical weather prediction model from GPV data and weather observation data and predicts a regional meteorological phenomenon in a fractionated region. A power demand prediction unit predicts a power demand in each region on the basis of the regional prediction result of the meteorological phenomenon. A generated heat value prediction unit predicts a generated heat value in each region on the basis of the regional prediction result of the power demand. The regional prediction result of the generated heat value is reflected on the numerical weather prediction model. Meteorological phenomena are predicted by adding the thermal energy as sensible heat from the earth's surface and taking artificial exhaust heat by electrical energy into consideration. Accordingly, weather prediction and power demand prediction are organically combined so that accurate weather prediction and power demand prediction are implemented while considering artificial heat exhaust.

The invention discloses a power distribution network plan method based on marketing and distribution information integration. The method comprises the steps of 1), integrating power distribution network plan basic data; 2), mapping power distribution network device and marketing customer information; 3), establishing a plan uniform device model; 4), analyzing current situations of a power network; 6), making a plan scheme; and 7) evaluating the plan scheme. The method has the beneficial effects that through marketing and distribution information integration, data required by a power network plan is formed, information is automatically collected through data interfaces, data sources are unified, and the data quality is improved; the plan uniform device model is established, automatic calculation analysis is realized, a traditional mode of collecting the information manually and making the plan scheme artificially is replaced, the artificial information collection cost is reduced, the physical power of planners can be liberated, the brain power of the planners is released, the problem that the plan is difficult to make is solved, and the quality of the plan scheme is improved; and the simple and convenient information collection process facilitates the routinized plan making work, and the plan rolling editing and revision timeliness is improved.

A plurality of forecast weather groups in a period comprising a plurality of days including a forecast target day for forecasting the electric power demand, and a plurality of actual weather groups in a period in a plurality of days in the past are set as the target period, and the similarity between the forecast weather group and the plurality of actual weather groups is calculated, a trend of a subsequent electric power demand is predicted based on the comparison of the plurality of calculated similarities whereby the electric power demand of a forecast target day is known.

A power control method for a power supplying unit for supplying power from a commercial power supply and a battery to a load. The method includes (a) calculating a deviation amount between an actual power demand and a predicted power demand in a predetermined unit period on the basis of power demand transition data and power demand prediction data, (b) correcting, by a computer, a leveling target value on the basis of the calculated deviation amount, and (c) controlling the power supplying unit so that the power supplying unit supplies power corresponding to the corrected leveling target value from the commercial power supply. The power demand transition data is stored in a power database, and the power demand prediction data is stored in a prediction data storing unit. The leveling target value is a target value of power to be supplied from the commercial power supply.

The invention relates to a leading industry based electricity demand prediction and early warning method, and belongs to the field of electricity system electricity demand analysis. The method includes performing time difference correlation analysis on industrial electricity consumption and a total electricity consumption sequence; predicting electricity consumption growth rates of each early-warning leading industry in following months virtually according to historical electricity consumption; calculating early warning success rate of each early-warning leading industry in history virtually; calculating early warning accuracy and coverage of each early-warning leading industry; calculating a mutual information value between each early-warning leading industry and the total electricity consumption sequence, and determining an electricity-demand-oriented early-warning leading industry set; calculating early warning threshold values of electricity demand growth rates of the early-warning leading industries; when the growth rates exceed the early warning numerical value range, sending out early warning signals, and performing rolling correction on follow-up electricity demand prediction. By the leading industry based electricity demand prediction and early warning method, electricity demand prediction accuracy can be improved, and supply and operation risks of a power grid can be reduced.

The invention discloses a monthly power consumption prediction method that comprehensively considers various economic factors, adopts the X-12-ARIMA model to decompose the monthly power and various economic factors into seasons; uses stepwise regression analysis to study the economic quantities and power consumption The correlation degree and regression model of electricity consumption were used to obtain preliminary prediction results; the annual power consumption forecast was carried out by using polynomial fitting, and the existing monthly power consumption forecast results were adjusted; the autoregressive integral sliding average Seasonal forecast corrections are carried out in each month to obtain a monthly electricity forecast model with good accuracy. Using the above technical solution, after seasonal decomposition of monthly electricity and economic volume, not only can use periodicity for forecasting, but also can effectively reduce the impact of volatility on regression analysis fitting accuracy and forecasting accuracy, and obtain good forecasting results.

Provided is a power supply grid load predicting method based on a geographic information system. A grid planning concept is introduced based on a power geographic information system. By combining the municipal planning and construction scheme and the distribution network planning technical guideline and according to the characteristics, a power supply area is divided in a scientific and reasonable manner based on the idea of large-to-small and divide-and-rule to form a plurality of relatively independent power supply grids with clear grid structure and self-healing potential. A strong support is provided for power demand prediction, distribution network status analysis, planning project scheme drafting, planning effect evaluation and investment decision making.

The invention provides a power demand change early warning method. The power demand change early warning method comprises the steps that stationarity test is carried out on preset index data in an alternative index bank and preset benchmark index data; index data are removed according to a stationarity test result if the index data and the benchmark index data are not integrated of the same order; if the index data and the benchmark index data are integrated of the same order, whether the index data and the benchmark index data are in co-integration or not is judged, if not, the index data are removed, and if yes, the fluctuation relationship between the index data and the benchmark index data is judged; the index data, the benchmark index data and the fluctuation relationship are input into a neural network prediction model, and a power demand prediction value is obtained; a historical power demand actual value within the preset period is obtained, a prosperity index is calculated according to the power demand prediction value, and an early warning message is generated. The invention further provides a power demand change early warning system. The prosperity index reflecting the power demand changes can be obtained rapidly, and the early warning message is given out for the power demand changes.

A method for predicting consumer power demand uses power consumption data measured over a long term and a power usage pattern immediately before a target time and for controlling ESS charge/discharge of an ESS based on the predicted power demand. A power demand prediction apparatus using the method includes respective components for collecting weather data and data on power used by the consumer; selecting data points according to a preset condition from among data collected by the data collector based on a specific time span; generating long term prediction data for the power demand in the specific time span based on the selected data points; analyzing a power usage pattern immediately before the specific time span and comparing the power usage pattern with the long-term prediction data, to determine whether prediction data correction is required; and correcting the prediction data based on the power usage pattern when correction is required.

A reserve power-supplying capacity transaction system of an embodiment of the present invention comprises a reserve capacity transaction unit, a determining unit, and a transaction-compensation distribution unit. The reserve capacity transaction unit obtains the reducible amount of power by a consumer, conducts a transaction thereof as reserve demand-adjustment capacity, and obtains compensation therefor. The determining unit determines the ranking of each of the consumers, such that consumers with little difference between a power-demand prediction amount thereof and the amount of power consumption thereof will have higher rankings. The transaction-compensation distribution unit distributes the compensation such that, when a request to reduce power consumption has not been sent out, consumers having higher rankings will have higher ratios, and when a request to reduce power consumption has been sent out, the consumers having higher rankings and whose estimated reduction amount will be great after the requested amount of power to be reduced is notified to the consumers in response to the request to reduce power consumption will have higher ratios.

The invention relates to a power demand forecast method based on a vector autoregression model. Step one, the economic data and the power demand data of a region (country, province or city) to be forecasted are collected; step two, the data are compiled and cleaned and stored in a database; step three, the time period required to be forecasted and the minimum time interval of the data are definedand logarithmic processing is performed on the corresponding data; step four, stationarity testing is performed on the four columns of processed data (fixed asset investment, total retail sales of consumer goods, total net export trade and the power demand); step five, an equation based on the vector autoregression model is constructed; step six, the lag order of the model is determined by using the AIC Akaike information criterion and the SIC Schwarz criterion; and step seven, the vector autoregression model is corrected and the power demand is forecasted by using the vector autoregression model. The method has the characteristics of high generalization capability and high forecast accuracy.

The invention discloses a method for building evaluation models of power distribution networks after rolling planning. The method includes establishing six primary indexes of power source planning, power demand prediction, power grid planning execution conditions, investment execution conditions, power grid development level and equipment utilization rates and 26 corresponding secondary indexes; determining weights of the primary indexes and the secondary indexes by the aid of hierarchical analysis processes; establishing equipment utilization rate influence factors; computing primary index quantization values and total quantization values by the aid of the weights of the primary indexes and the secondary indexes and the equipment utilization rate influence factors. The method has the advantages that the power distribution networks can be evaluated by the aid of evaluation values and total evaluation values of the various indexes after rolling planning, accordingly, scientific quantitative analysis bases can be provided for power distribution network planning and investment decision making in late periods, the problems of deficiency of evaluation models of existing power distribution networks after rolling planning and inaccurate and unscientific evaluation can be effectively solved, planning can be adjusted scientifically, and the method is favorable for quantitative horizontal comparison on power distribution network planning work in various regions.

The invention provides an LEAP-model-based method for predicating medium-and-long-term electricity demand distribution of a power grid. The method comprises the following steps that an LEAP model of a medium-and-long-term terminal energy demand and energy processing and converting link of a region is established; based on the LEAP model, a medium-and-long-term electricity demand predicted value of the power grid of the region is obtained; the distribution of medium-and-long-term region total electricity consumption in all administrative areas is predicated according to electricity utilization departments; a predicated result of the distribution of the medium-and-long-term region total electricity consumption in all the administrative areas is obtained; the distribution of the medium-and-long-term region total maximum load in all the administrative areas is predicated. The method offers a reliable basis to regional power grid planners for mastering of the medium-and-long-term region electricity demand distribution, thereby serving power grid planning better.

It is an object of the present invention to provide a load leveling system of a power system capable of attaining load leveling in a distribution substation unit and a method thereof. Disclosed is a load leveling system of a power system with a controller having a major customer and lower transforming apparatus, and with a customer connected to the lower transforming apparatus via a low-voltage line. The system also has a communication unit between the customer and its own system, and control use time zones of devices within the customer; an electric demand prediction unit, and a first load leveling processing unit. By employing such an arranged system, it becomes possible to provide a load leveling system of a power system capable of realizing load leveling of both a distribution substation unit and a pole transformer unit.

Provided are a method and apparatus for forecasting power demand. The method of forecasting power demand includes forming weighted power demand data by assigning different weights to power demand data according to the frequency of the power demand data, and forming a power demand forecasting model by recurrent neural network (RNN)-based deep learning using the weighted power demand data. From the power demand forecasting model, a power demand forecasting value is extracted using a forecast target label or index information.

The invention discloses a method for predicting electricity consumption based on business expansion data and econometric model. Through analysis on regional business expansion data, by adopting econometric model, the relations among business expansion conditions,operation capacity, load utilization ratio and electricity demand is studied within industry classification to build electricity demand prediction model. First,cleaning business expansion data and eliminating information of non-production electricity demand application; then, predicting electricity consumed not in business expansion, electricity consumed in business expansion of all industries and residential electricity consumption, according to which, predicting regional electricity consumption. The invention can better grasp the trend of regional electricity demand and can serve to social development of regional economy as well as safe and stable operation of electricity network.

The invention provides an energy supply and demand analysis and prediction platform based on big data application. The platform comprises a basic database management module for storing a business database, an economic database, a meteorological database and an industrial database; a power demand prediction module used for counting the change characteristics of the electric quantity and electric load indexes and reasonably predicting and analyzing the influence on monthly electric quantity indexes; and a regional energy economy analysis module used for data variable setting, data modification,data storage, data operation model and visual display. According to the invention, complex prediction work is simplified into simple system operation of a single person from a large amount of originalmanual collection, arrangement, calculation and analysis processes. According to the invention, scene construction is carried out based on a full-service data center, in order to adapt to a new situation faced by energy supply and demand analysis and prediction work, the influence of factors such as economy and society on energy demands is fully considered, multi-dimensional mass data is comprehensively utilized, and an energy supply and demand analysis and prediction platform application scene is developed.

The invention discloses a grid division method of power distribution network gridding planning. The method comprises the following steps of status assessments of gridding power distribution networks;demand forecasting of gridding electric power; planning of technical principles and schemes; planning of sites and channels; investment estimation and planning assessments. The method accurately locates problems existing in the power distribution networks to each grid, each grid can be examined under a magnifying glass to ensure that the existing problems can be solved through investments, and accurate delivery is achieved.