CN115392546A - Method and device for predicting cooling electric quantity and managing demand side based on meteorological factors - Google Patents
Method and device for predicting cooling electric quantity and managing demand side based on meteorological factors Download PDFInfo
- Publication number
- CN115392546A CN115392546A CN202210924740.9A CN202210924740A CN115392546A CN 115392546 A CN115392546 A CN 115392546A CN 202210924740 A CN202210924740 A CN 202210924740A CN 115392546 A CN115392546 A CN 115392546A
- Authority
- CN
- China
- Prior art keywords
- electric quantity
- day
- cooling
- load
- cooling electric
- 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
- 238000001816 cooling Methods 0.000 title claims abstract description 148
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000013528 artificial neural network Methods 0.000 claims abstract description 17
- 230000005611 electricity Effects 0.000 claims description 43
- 238000007726 management method Methods 0.000 claims description 40
- 239000003102 growth factor Substances 0.000 claims description 13
- 239000011159 matrix material Substances 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 6
- 230000002045 lasting effect Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 5
- 230000008878 coupling Effects 0.000 abstract description 4
- 238000010168 coupling process Methods 0.000 abstract description 4
- 238000005859 coupling reaction Methods 0.000 abstract description 4
- 230000002265 prevention Effects 0.000 abstract description 3
- 230000002354 daily effect Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000014759 maintenance of location Effects 0.000 description 2
- 229920000642 polymer Polymers 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 238000004378 air conditioning Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000000547 structure data Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- General Health & Medical Sciences (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Tourism & Hospitality (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Quality & Reliability (AREA)
- Educational Administration (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Operations Research (AREA)
- Artificial Intelligence (AREA)
- Power Engineering (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a device for predicting cooling electric quantity and managing a demand side based on meteorological factors, wherein a first preset load reference day set and a second preset load reference day set are obtained through historical data based on different preset conditions, a reference load curve is obtained by clustering the first preset load reference day set, day-by-day cooling electric quantity is obtained according to the second preset load reference day set and the reference load curve, historical cooling electric quantity data are accurately measured and calculated by a reference load method, a neural network is trained through the historical cooling electric quantity data and the meteorological factors, and a coupling relation between meteorological factors and cooling electric quantity is established, so that accurate cooling electric quantity data can be obtained when a subsequent day to be predicted is predicted, meteorological influencing factors are integrated in the cooling electric quantity prediction process, technical personnel can reasonably make a demand side management plan based on prediction results, the operation management of an electric power system is optimized, and the operation efficiency of the electric power system and the load loss risk prevention are improved.
Description
Technical Field
The invention relates to the technical field of electric quantity prediction, in particular to a cooling electric quantity prediction and demand side management method and device based on meteorological factors.
Background
With the development of industrial structures and the improvement of economic levels, the power consumption in different fields such as industrial power, commercial power, household power and the like is increasing. Therefore, management of power systems is important in economic development. In the management of power systems, it is most important to predict short-term loads, including, for example, the amount of cooling power. The cooling load refers to a load which is generated by cooling equipment such as an air conditioner and an ice chest and causes the load of a power grid to rapidly increase in a short period. With the increasing proportion of the cooling electric quantity in the power consumption of the whole society, the prediction of the cooling electric quantity has important significance for accurately predicting the monthly electric quantity in summer so as to reasonably arrange a monthly power generation plan. However, currently, prediction modes are usually based on single data, and high-precision prediction cannot be performed on multi-dimensional data in power grid data.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method and the device for predicting the cooling electric quantity and managing the demand side based on the meteorological factors are provided, the accuracy of predicting the cooling electric quantity is improved, a power generation plan is scientifically and reasonably formulated, and the operation efficiency of a power system is improved.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
a cooling electric quantity prediction and demand side management method based on meteorological factors is characterized by comprising the following steps:
acquiring historical load data to obtain a first preset load reference day set and a second preset load reference day set;
clustering the first preset load reference day set to obtain a reference load curve;
obtaining the day-by-day cooling electric quantity according to the second preset load reference day set and the reference load curve;
training a neural network by using a training set formed by taking meteorological factors corresponding to the second preset load reference day set as input and taking the day-by-day cooling electric quantity as output to obtain a cooling electric quantity measuring and calculating model;
and inputting the meteorological factors corresponding to the day set to be predicted into the cooling electric quantity measuring and calculating model to obtain the predicted cooling electric quantity.
In order to solve the technical problem, the invention adopts another technical scheme as follows:
the cooling electric quantity prediction and demand side management device based on the meteorological factor comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the computer program to realize the steps of the cooling electric quantity prediction and demand side management method based on the meteorological factor.
The invention has the beneficial effects that: the method comprises the steps of obtaining a first preset load reference day set and a second preset load reference day set through historical data based on different preset conditions, clustering the first preset load reference day set to obtain a reference load curve, obtaining day-by-day cooling electric quantity according to the second preset load reference day set and the reference load curve, accurately measuring and calculating historical cooling electric quantity data through a reference load method, training a neural network through the historical cooling electric quantity data in combination with meteorological factors, and establishing a coupling relation between meteorological factors and cooling electric quantity, so that accurate cooling electric quantity data can be obtained when prediction is carried out on a day to be predicted subsequently, influence factors of the meteorological factors are integrated in the cooling electric quantity prediction process, technicians can reasonably make a demand side management plan based on prediction results, operation management of an electric power system is optimized, and the improvement of the operation efficiency of the electric power system and the prevention of load loss risks are facilitated.
Drawings
FIG. 1 is a flowchart illustrating steps of a method for forecasting cooling power and managing demand side based on meteorological factors according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a cooling power prediction and demand side management device based on meteorological factors according to an embodiment of the present invention;
FIG. 3 is a schematic block flow diagram of a cooling power prediction and demand side management apparatus based on meteorological factors according to an embodiment of the present invention;
FIG. 4 is a schematic illustration of a reference load curve and a baseline load curve in an embodiment of the present invention;
fig. 5 is a schematic diagram of a predicted value and a measured value of the cooling power everyday according to the embodiment of the present invention;
FIG. 6 is a schematic diagram of an exemplary cooling load curve and demand side management period in accordance with an embodiment of the present invention;
FIG. 7 is a schematic diagram of the stepped electricity prices of an embodiment of the present invention;
fig. 8 is a schematic diagram of the effect of demand side management according to an embodiment of the present invention.
Detailed Description
In order to explain the technical contents, the objects and the effects of the present invention in detail, the following description is made with reference to the accompanying drawings in combination with the embodiments.
Referring to fig. 1, a method for predicting cooling power and managing demand side based on meteorological factors includes the steps:
acquiring historical load data to obtain a first preset load reference day set and a second preset load reference day set;
clustering the first preset load reference day set to obtain a reference load curve;
obtaining the day-by-day cooling electric quantity according to the second preset load reference day set and the reference load curve;
training a neural network by using a training set formed by taking meteorological factors corresponding to the second preset load reference day set as input and taking the day-by-day cooling electric quantity as output to obtain a cooling electric quantity measuring and calculating model;
and inputting the meteorological factors corresponding to the day set to be predicted into the cooling electric quantity measuring and calculating model to obtain the predicted cooling electric quantity.
As can be seen from the above description, the beneficial effects of the present invention are: the method comprises the steps of obtaining a first preset load reference day set and a second preset load reference day set through historical data based on different preset conditions, clustering the first preset load reference day set to obtain a reference load curve, obtaining day-by-day cooling electric quantity according to the second preset load reference day set and the reference load curve, accurately measuring and calculating historical cooling electric quantity data through a reference load method, training a neural network through accurate historical cooling electric quantity data and meteorological factors, and establishing a coupling relation between meteorological factors and the cooling electric quantity, so that accurate cooling electric quantity data can be obtained when a subsequent day to be predicted is predicted, influence factors of the meteorological factors are blended in the cooling electric quantity prediction process, technical personnel can reasonably make a demand side management plan based on prediction results, operation management of an electric power system is optimized, and improvement of operation efficiency of the electric power system and prevention of load loss risks are facilitated.
Further, the clustering the first preset load reference day set to obtain a reference load curve includes:
using membership matrix W = (W) ik ) 24×n L representing that the integral point load of the first preset load reference day belongs to each cluster center ref ＝[l 1 ,l 2 ,...，l 24 ]Degree;
optimizing clustering by minimizing an objective function J to obtain the reference load curve L ref (t)；
Wherein n is the number of samples to be clustered, w ik The membership degree of the kth sample belonging to the ith cluster; d is a radical of ik The distance from the kth sample to the ith class center.
According to the description, the load of the integral point of the reference day is clustered into the load curve points by adopting a fuzzy C-means clustering method, and the membership degree of each sample point to all class centers is obtained by optimizing the clustering through a minimized objective function J, so that the class of the sample points is determined to achieve the purpose of automatically classifying the sample data, the reference load curve can be accurately generated, and the accuracy of calculating the cooling electric quantity is improved.
Further, the obtaining of the day-to-day cooling electric quantity according to the second preset load reference day set and the reference load curve includes:
obtaining historical economic data corresponding to the historical load data to obtain an economic growth factor;
correcting the reference load curve according to the economic growth factor to obtain a reference load curve;
and obtaining the day-by-day cooling electric quantity according to the second preset load reference day set and the reference load curve.
According to the description, historical economic data are considered on the basis of the reference load curve, the reference load curve is corrected according to the economic growth factor to obtain the reference load curve, the power load is combined with the economic condition to obtain a more accurate reference load curve, and therefore the accuracy of the obtained day-by-day cooling electric quantity is improved.
Further, the obtaining the daily cooling electric quantity according to the second preset load reference daily set and the reference load curve includes:
subtracting the reference load curve from the load curve corresponding to the second preset load reference day set to obtain a cooling load curve;
and integrating the cooling load curve to obtain the day-by-day cooling electric quantity.
According to the above description, the load curve corresponding to the second preset load reference day set is subtracted from the reference load curve to obtain the cooling load curve, and the cooling load curve is further integrated to obtain the day-by-day cooling electric quantity data, so that the accurate cooling electric quantity data per day can be obtained.
Further, the reference load curves include a weekday reference load curve and a weekend reference load curve.
According to the above description, the reference load curve is divided into the working day reference load curve and the weekend reference load curve according to different date attributes, so that the working day and weekend loads can be respectively processed in a targeted manner, and the accuracy of predicting the cooling electric quantity on different dates is improved.
Further, training the neural network by using a training set which is formed by taking meteorological factors corresponding to the second preset load reference day set as input and taking the day-by-day cooling electric quantity as output comprises the following steps of:
the weather factors comprise date types, highest air temperature, lowest air temperature, high-temperature duration days and weather; different ones of the date types and the weather have different coefficient factors;
and inputting the date type, the highest air temperature, the lowest air temperature, the high temperature duration days and the weather generation meteorological factor matrix into the neural network.
As can be seen from the above description, by setting a plurality of different weather factors and setting different coefficient factors for different types of dates and weather, accurate predictions can be made for different types of dates and weather.
Further, obtaining the predicted cooling power amount then includes:
and determining the electricity prices of different time periods according to the predicted cooling electricity quantity.
According to the description, the electricity prices in different time periods are determined according to the obtained predicted cooling electricity quantity, so that the operation management of the power system is optimized.
Further, the determining the electricity prices of different time periods according to the predicted cooling electric quantity comprises:
determining the electricity prices of different time periods according to a temporary step electricity price formula:
in the formula: p (t) is the predicted time period electricity price, p 0 E (t) is the temperature-reducing electric quantity in the prediction time interval,
to predict the hourly average daily cooling capacity, q 0 To predict the total electricity in a time period, ε is a price elastic coefficient.
According to the above description, the electricity price of the forecast date is subjected to gradient management according to the temporary step electricity price formula and the forecast cooling electricity quantity, so that the electricity price of the cooling electricity quantity is effectively managed.
Further, still include:
acquiring a historical peak-valley difference electricity price according to the historical load data;
obtaining a price elastic coefficient according to the historical peak-valley difference electricity price:
in the formula: q. q.s 0 、p 0 The initial power demand and the initial power price before the peak-valley difference power price is executed respectively, and the electric quantity and the power price variable quantity after the peak-valley difference power price is executed respectively are delta q and delta p.
According to the above description, the price elasticity coefficient is controlled by acquiring the historical peak-valley difference electricity price through the historical load data, so that the electricity price elasticity control of the electricity consumption peak and the electricity consumption valley can be pertinently performed.
Another embodiment of the present invention provides a cooling power prediction and demand side management apparatus based on meteorological factors, which includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein when the processor executes the computer program, each step of the cooling power prediction and demand side management method based on meteorological factors is implemented.
The cooling electric quantity prediction management method and device based on meteorological factors can be suitable for predicting cooling electric quantity on different types of dates and provide an effective demand side management method, and the following description is provided through a specific implementation mode:
example one
Referring to fig. 1, a method for predicting cooling power and managing demand side based on meteorological factors includes the steps:
s1, obtaining historical load data to obtain a first preset load reference day set and a second preset load reference day set; if the weather in spring is clear and the highest temperature in the day is not more than 28 ℃, selecting 10-15 days as a first preset condition, and the load data before the temperature is not increased is the first preset load reference day set; if the area is the southern area, the load data (the air temperature is increased and the cooling electric quantity demand is increased) is taken as the load data after the air temperature is increased, namely the second preset load reference day set, according to the 5-10 monthly history load data in the last three years; the load difference before and after temperature rise, namely the temperature rise electric quantity can be known through the load curve corresponding to the first preset load reference day set and the load curve corresponding to the second preset load reference day set;
s2, clustering the first preset load reference day set to obtain a reference load curve, specifically:
using membership matrix W = (W) ik ) 24×n L representing that the integral point load of the first preset load reference day belongs to each cluster center ref ＝[l 1 ,l 2 ,...，l 24 ]Degree;
optimizing clustering by minimizing an objective function J to obtain the reference load curve L ref (t)；
Wherein n is the number of samples to be clustered, w ik The membership degree of the kth sample belonging to the ith cluster; d is a radical of ik The distance from the kth sample to the ith class center;
in an alternative embodiment, the reference load curves include a weekday reference load curve and a weekend reference load curve; namely, the selected 10 to 15 days comprise weekdays and weekends, the weekday load and the weekend load are respectively used as a weekday load curve set and a weekend reference load curve set to respectively obtain a weekend reference load curve
And weekend reference load curve
S3, obtaining the day-by-day cooling electric quantity according to the second preset load reference day set and the reference load curve comprises the following steps:
s31, obtaining historical economic data corresponding to the historical load data to obtain an economic growth factor; wherein the economic growth factor is an economic monthly growth coefficient; the method is obtained by evaluating economic growth conditions of a cooling electric quantity measuring month and a reference month (such as 4 months), and combining the economic situation analysis of the month-by-month in three years and the natural growth condition;
s32, correcting the reference load curve according to the economic growth factor to obtain a reference load curve:
wherein,
in order to measure the reference load curve of the month working day,
in order to measure the standard load curve of the weekend,
the reference load curve is a reference load curve for the working day,
c is an economic growth coefficient of the measured month compared with the reference month;
s33, obtaining the day-by-day cooling electric quantity according to the second preset load reference day set and the reference load curve, specifically:
subtracting the reference load curve from the load curve corresponding to the second preset load reference day set to obtain a cooling load curve; if the load data of the 5-10 month history in the last three years is obtained, the corresponding load curve L is obtained i (t)；
Wherein L is i (t)、
Respectively is a load curve and a cooling load curve of the ith day;
integrating the cooling load curve to obtain the day-by-day cooling electric quantity, namely:
s4, training a neural network by using a training set formed by taking meteorological factors corresponding to the second preset load reference day set as input and taking the day-by-day cooling electric quantity as output to obtain a cooling electric quantity measuring and calculating model; the weather factors comprise date type, highest air temperature, lowest air temperature, high temperature lasting days (lasting days with the highest temperature of the day being more than 33 ℃) and weather; different date types and weather have different coefficient factors, for example, in the date types, the working day coefficient is 0.5, the weekend coefficient is 1, the holiday coefficient day is 2, and in the weather, the no-rain coefficient is 0.1, the light rain coefficient is 0.2, and the medium-to-heavy rain coefficient is 0.3; inputting the date type, the highest air temperature, the lowest air temperature, the high-temperature continuous days and the weather generation meteorological factor matrix into the neural network, wherein the coefficient is only used for identifying the date type and the weather type and can take any value; specifically, the method comprises the following steps:
s41, initializing weight v ik ，ω kj And a threshold value theta k ，θ j (i =1, 2.. D, k =1, 2.. Q, j =1, 2.. P, wherein d, q, p are the number of neurons of the input layer, the output layer and the hidden layer respectively;
s42, taking the meteorological factor matrix as input, and setting the meteorological factor matrix as x i (i =1,2, \8230;, d); using the cooling electric quantity day by day as output y j (j＝1,2,...,l)；
S43, calculating actual output of the BP neural network:
s431, calculating hidden layer output
Wherein f (x) is the activation function of the hidden layer, taken
S432, calculating output layer output
S44, correcting the weight value
ω kj (t+1)＝ω kj (t)+δ j (t)b k ，
v ik (t+1)＝v ik (t)+δ k (t)x i ；
In the formula,
δ k (t)＝b k (1-b k )δ kj (t)ω kj ；
in the training process, when the error reaches the precision or the maximum cycle number, outputting the result, otherwise, turning to the step S42;
s5, inputting meteorological factors corresponding to a day set to be predicted into the cooling electric quantity measuring and calculating model to obtain predicted cooling electric quantity; the final cooling electric quantity data is obtained by multiplying the cooling electric quantity obtained by measurement and calculation by an annual growth factor; the annual growth factor can be obtained by combining the air conditioner retention amount in the statistical yearbook, the economic development index and the industrial structure data.
Example two
The difference between the embodiment and the first embodiment is that an electric quantity management plan is made in advance aiming at the predicted day-by-day cooling electric quantity, and early warning and demand side management work is carried out; if the date that cooling electric quantity is great, release peak power consumption early warning in advance to according to cooling electric quantity peak period characteristics, carry out demand side management plan, including carrying out interim ladder price of electricity policy, implement the round of control plan means to the air conditioner polymer, realize demand side classification management, wherein stage price of electricity makes the policy and is:
determining the electricity prices of different time periods according to a temporary step electricity price formula
In the formula: p (t) is the predicted time interval electricity price, p 0 E (t) is the temperature reduction electric quantity in the prediction period,
to predict the hourly average daily cooling capacity, q 0 Epsilon is a price elasticity coefficient for predicting the total electric quantity in a time interval;
the price elasticity coefficient is calculated according to a historical peak-valley difference electricity price execution result:
in the formula: q. q of 0 、p 0 Respectively an initial power demand and an initial power price before executing the peak-valley difference power price, and respectively electric quantity and power price variable quantity after executing the peak-valley difference power price; meanwhile, the price ratio of the management time interval to the non-management time interval does not exceed 5 times.
EXAMPLE III
Referring to fig. 2, a cooling power prediction and demand side management apparatus based on meteorological factors includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement each step of the cooling power prediction and demand side management method based on meteorological factors according to any one of the first or second embodiments;
referring to fig. 3, the cooling electric quantity prediction management device based on meteorological factors can calculate and input the meteorological factor matrix and the cooling electric quantity into the BP neural network, predict the cooling electric quantity day by day and month, and realize the functions of day early warning of high cooling electric quantity and advising of demand side management time interval;
taking the prediction of the cooling electric quantity in 2021 year 7 month in a certain southern province as an example, a specific implementation mode is described as follows:
s1, selecting a working day and a weekend with clear weather of 4 months and a day maximum temperature of no more than 28 ℃, and respectively forming a working day load reference day set and a weekend load reference day set;
s2, clustering a working day load reference day set and a weekend load reference day set respectively to obtain a working day reference curve and a weekend reference load curve;
s3, referring to FIG. 4 (taking working days as an example), on the basis of the reference load curve, considering an economic growth factor, such as a reference load curve obtained by taking a growth coefficient of 1.02 according to the natural growth rate between 5 months and 3 months every year in the last three years and the economic condition in the current month; calculating to obtain the cooling electric quantity day by day in 5-10 months in three years based on the reference load curve; subtracting the annual reference load curve from the actual load curve of 5-10 months in the last three years to obtain a daily cooling load curve, and integrating the cooling load curve to obtain daily cooling electric quantity;
s4, forecasting daily cooling electric quantity and monthly total cooling electric quantity by adopting a BP neural network algorithm; acquiring day-by-day meteorological data, acquiring a meteorological factor matrix as input, outputting cooling electric quantity data, and taking data of a predicted month close to the month and the same period of the last year as a training set, for example, predicting the cooling electric quantity of 2021 year 7 month, and inputting data of 2020 year 6-8 month and 2021 year 6 month as training samples; the listed training set samples for 10 days are shown in table 1; wherein the cooling electric quantity data in 6 months in 2020 is obtained by multiplying the cooling electric quantity data measured in the previous step by an annual growth factor, and the annual growth factor is taken as 1.03 by combining the nearly five-year-old average air conditioner retention quantity data and an annual GDP index;
TABLE 1 training set samples
Referring to fig. 5, after the training set is normalized, the normalized training set is input into an electric quantity prediction device based on a BP neural network for training, and then a meteorological factor matrix based on meteorological forecast data of 7 months in 2021 is used as prediction input to obtain a daily cooling electric quantity prediction value and a monthly cooling electric quantity prediction value of 64.3 hundred million kilowatt-hours; comparing the measured value with an actual measured value to obtain a prediction error of 3.2%;
according to the prediction result of the cooling electric quantity, carrying out demand side management on the time period with higher cooling electric quantity; if the high cooling electricity quantity day early warning is generated according to the prediction result, outputting early warning days of 7 months, 8 days-10 days, 7 months, 12 days-16 days, 7 months, 19 days-7 months, 21 days, 7 months, 28 days and 7 months, 31 days; according to the characteristic of a typical cooling load curve of 7 months (as shown in fig. 6), carrying out demand side management on the early warning days; the output demand side management recommended time period is 9-00, assuming that the basic electricity price is 0.5 yuan and the price elasticity coefficient is 0.04, the temporary peak electricity price level in the time period is shown in fig. 7, and after a temporary step electricity price policy is implemented in the time period, the cooling electricity quantity curve is shown in fig. 8, and the cooling electricity quantity is effectively reduced by 3%; meanwhile, a wheel control implementation plan can be adopted for part of air conditioner polymers according to needs, so that cooling electric quantity management is realized.
In summary, according to the method and the device for predicting cooling electric quantity and managing the demand side based on the meteorological factor, the first preset load reference day set and the second preset load reference day set are obtained through historical data based on different preset conditions, the first preset load reference day set is clustered to obtain the reference load curve, the day-by-day cooling electric quantity is obtained according to the second preset load reference day set and the reference load curve, namely, the historical cooling electric quantity data can be accurately measured and calculated through a reference load method, the neural network is trained through the accurate historical cooling electric quantity data and the meteorological factor, and the coupling relation between factors such as the meteorological factors and the cooling electric quantity is established, so that the accurate cooling electric quantity data can be obtained when the subsequent days to be predicted, influence factors of the meteorological factors are integrated in the process of predicting the cooling electric quantity, technicians can execute the demand side management plan based on the prediction result, including executing a temporary step electricity price policy, and implementing a rotation control plan means on the air conditioning aggregate, thereby realizing the demand side classification management, optimizing the operation management of the electric power system, and being more beneficial to improving the operation efficiency of the electric power system and preventing the load loss risk.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all equivalent changes made by using the contents of the present specification and the drawings, or applied directly or indirectly to the related technical fields, are included in the scope of the present invention.
Claims (10)
1. A cooling electric quantity prediction and demand side management method based on meteorological factors is characterized by comprising the following steps:
acquiring historical load data to obtain a first preset load reference day set and a second preset load reference day set;
clustering the first preset load reference day set to obtain a reference load curve;
obtaining the day-by-day cooling electric quantity according to the second preset load reference day set and the reference load curve;
training a neural network by using a training set formed by taking meteorological factors corresponding to the second preset load reference day set as input and taking the day-by-day cooling electric quantity as output to obtain a cooling electric quantity measuring and calculating model;
and inputting the meteorological factors corresponding to the day set to be predicted into the cooling electric quantity measuring and calculating model to obtain the predicted cooling electric quantity.
2. The method for forecasting the cooling electric quantity and managing the demand side based on the meteorological factor according to claim 1, wherein the clustering the first preset load reference day set to obtain a reference load curve includes:
using membership matrix W = (W) ik ) 24×n L representing that the integral point load of the first preset load reference day belongs to each cluster center ref ＝[l 1 ,l 2 ,...，l 24 ]Degree;
optimizing clustering by minimizing an objective function J to obtain the reference load curve L ref (t)；
Wherein n is the number of samples to be clustered, w ik The membership degree of the kth sample belonging to the ith cluster; d ik The distance from the kth sample to the ith class center.
3. The method according to claim 1, wherein the obtaining of the cooling power on a daily basis according to the second predetermined load reference daily set and the reference load curve comprises:
obtaining historical economic data corresponding to the historical load data to obtain an economic growth factor;
correcting the reference load curve according to the economic growth factor to obtain a reference load curve;
and obtaining the day-by-day cooling electric quantity according to the second preset load reference day set and the reference load curve.
4. The method according to claim 3, wherein the obtaining the daily cooling power according to the second predetermined load reference daily set and the reference load curve comprises:
subtracting the reference load curve from the load curve corresponding to the second preset load reference day set to obtain a cooling load curve;
and integrating the cooling load curve to obtain the day-by-day cooling electric quantity.
5. The method for forecasting the cooling electric quantity and managing the demand side based on the meteorological factors according to claim 1,2 or 3, wherein the reference load curves comprise a working day reference load curve and a weekend reference load curve.
6. The method as claimed in claim 1, wherein the training of the neural network with the meteorological factor corresponding to the second predetermined load reference day set as an input and the daily cooling power as an output comprises:
the weather factors comprise date types, highest air temperature, lowest air temperature, high-temperature duration days and weather; different ones of the date types and the weather have different coefficient factors;
and inputting the date type, the highest air temperature, the lowest air temperature, the high temperature lasting days and the weather generation meteorological factor matrix into the neural network.
7. The method of claim 1, wherein the obtaining the predicted cooling power comprises:
and determining the electricity prices in different time periods according to the predicted cooling electricity quantity.
8. The method of claim 7, wherein the determining the electricity prices at different time periods according to the predicted cooling electric quantity comprises:
determining the electricity prices of different time periods according to a temporary step electricity price formula:
in the formula: p (t) is the predicted time period electricity price, p 0 E (t) is the temperature-reducing electric quantity in the prediction time interval,
to predict the hourly average daily cooling capacity, q 0 To predict the total electricity in a time period, ε is a price elastic coefficient.
9. The method for forecasting the cooling power and managing the demand side based on the meteorological factor according to claim 8, further comprising:
acquiring a historical peak-valley difference electricity price according to the historical load data;
obtaining a price elastic coefficient according to the historical peak-valley difference electricity price:
in the formula: q. q of 0 、p 0 The initial power demand and the initial power price before the peak-valley difference power price is executed respectively, and the electric quantity and the power price variable quantity after the peak-valley difference power price is executed respectively are delta q and delta p.
10. A meteorological-factor-based cooling power prediction and demand-side management apparatus comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to perform the steps of the meteorological-factor-based cooling power prediction and demand-side management method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210924740.9A CN115392546A (en) | 2022-08-02 | 2022-08-02 | Method and device for predicting cooling electric quantity and managing demand side based on meteorological factors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210924740.9A CN115392546A (en) | 2022-08-02 | 2022-08-02 | Method and device for predicting cooling electric quantity and managing demand side based on meteorological factors |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115392546A true CN115392546A (en) | 2022-11-25 |
Family
ID=84119145
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210924740.9A Pending CN115392546A (en) | 2022-08-02 | 2022-08-02 | Method and device for predicting cooling electric quantity and managing demand side based on meteorological factors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115392546A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117333219A (en) * | 2023-12-01 | 2024-01-02 | 国网浙江省电力有限公司 | Transaction electric quantity prediction method, device, equipment and storage medium |
-
2022
- 2022-08-02 CN CN202210924740.9A patent/CN115392546A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117333219A (en) * | 2023-12-01 | 2024-01-02 | 国网浙江省电力有限公司 | Transaction electric quantity prediction method, device, equipment and storage medium |
| CN117333219B (en) * | 2023-12-01 | 2024-03-08 | 国网浙江省电力有限公司 | Transaction electric quantity prediction method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111860979B (en) | Short-term load prediction method based on TCN and IPSO-LSSVM combined model | |
| CN111932402B (en) | Short-term power load bidirectional combination prediction method based on similar day and LSTM | |
| CN108593990B (en) | Electricity stealing detection method based on electricity consumption behavior mode of electric energy user and application | |
| CN110097297A (en) | A kind of various dimensions stealing situation Intellisense method, system, equipment and medium | |
| US10175709B2 (en) | Consumer electric power control system and consumer electric power control method | |
| CN109934386B (en) | Heat load prediction method for combined heat and power system | |
| CN105117810A (en) | Residential electricity consumption mid-term load prediction method under multistep electricity price mechanism | |
| CN105069525A (en) | All-weather 96-point daily load curve prediction and optimization correction system | |
| CN106779129A (en) | A kind of Short-Term Load Forecasting Method for considering meteorologic factor | |
| CN113255973A (en) | Power load prediction method, power load prediction device, computer equipment and storage medium | |
| CN113743673B (en) | Power load prediction method during typhoon | |
| CN110033134A (en) | A kind of short-term load forecasting algorithm of segmentation day by day considering meteorologic factor | |
| CN102426674A (en) | Power system load prediction method based on Markov chain | |
| CN114862139A (en) | Data-driven transformer area line loss rate abnormity diagnosis method | |
| CN114372360A (en) | Method, terminal and storage medium for power load prediction | |
| CN107977898B (en) | Generated energy insurance pricing evaluation method of photovoltaic power station | |
| CN110880055A (en) | Building intelligent ammeter system | |
| CN115392546A (en) | Method and device for predicting cooling electric quantity and managing demand side based on meteorological factors | |
| CN112149309A (en) | Temperature-adjusting load curve fitting method considering temperature-load correlation | |
| CN117851908B (en) | Improved on-line low-voltage transformer area electric energy meter misalignment monitoring method and device | |
| CN109858668B (en) | Coordination prediction method for power load region in thunder and lightning climate | |
| CN105761163A (en) | Temperature adjustment load quantifying method based on correlation analysis and meteorological factor clustering | |
| CN114186733A (en) | Short-term load prediction method and device | |
| CN113902181A (en) | Short-term prediction method and equipment for common variable heavy overload | |
| CN113537578A (en) | Power consumer behavior prediction method |
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 |