CN116845878B - Electric load prediction method for micro-grid - Google Patents
Electric load prediction method for micro-grid Download PDFInfo
- Publication number
- CN116845878B CN116845878B CN202310838578.3A CN202310838578A CN116845878B CN 116845878 B CN116845878 B CN 116845878B CN 202310838578 A CN202310838578 A CN 202310838578A CN 116845878 B CN116845878 B CN 116845878B
- Authority
- CN
- China
- Prior art keywords
- data points
- time scale
- degree
- scale
- fluctuation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 79
- 230000007704 transition Effects 0.000 claims abstract description 97
- 239000011159 matrix material Substances 0.000 claims abstract description 72
- 230000008859 change Effects 0.000 claims abstract description 36
- 230000005611 electricity Effects 0.000 claims description 80
- 230000008569 process Effects 0.000 claims description 29
- 238000010606 normalization Methods 0.000 claims description 22
- 230000035772 mutation Effects 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000009826 distribution Methods 0.000 abstract description 29
- 238000004590 computer program Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 9
- 238000003860 storage Methods 0.000 description 8
- 238000005457 optimization Methods 0.000 description 6
- 238000004146 energy storage Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 230000001105 regulatory effect Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic 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—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
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Public Health (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Algebra (AREA)
- Data Mining & Analysis (AREA)
- Computational Mathematics (AREA)
- Evolutionary Computation (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention provides a method for predicting the power consumption load of a micro-grid, which is used for acquiring power consumption load data of a multi-element time scale of a current data point under the condition of accessing a new power consumption load or a power consumption load with an emergency; and predicting the power consumption load data of the next data point according to a pre-established prediction model and the power consumption load data of the current data point in a multi-time scale, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of an emergency, the distribution weights of the state probability transition matrices in different time scales are adjusted through the trend change of the data points in the multi-time scale, so that a prediction model is obtained, the power consumption load data is predicted through the prediction model, and the prediction accuracy is improved.
Description
Technical Field
The invention relates to the technical field of intelligent electricity utilization, in particular to an electricity utilization load prediction method for a micro-grid.
Background
In the process of predicting the electric load of the micro-grid, the electric load of the micro-grid is increased sharply due to weather factors, such as sudden severe weather conditions of high temperature, low temperature, storm and the like.
In the prior art, when a hidden Markov model is established to predict the electric load of a micro-grid, when a new load is accessed or an emergency occurs in the micro-grid, the hidden state of the hidden Markov model is greatly changed due to the large change of the sudden electric load, and abnormal values occur in the time sequence data of the electric load, so that the precision of the hidden Markov model is reduced, and the situation of over fitting or under fitting occurs, so that the electric load cannot be accurately predicted.
Disclosure of Invention
In view of the foregoing, embodiments of the present invention have been made to provide a method, apparatus, terminal device, and readable storage medium for electric load prediction for a micro grid that overcomes or at least partially solves the foregoing problems.
In a first aspect, an embodiment of the present invention provides a method for predicting an electrical load of a micro-grid, the method including:
under the condition of accessing a new electricity load or an electricity load with an emergency, acquiring electricity load data of a multi-element time scale of a current data point;
and predicting the electricity load data of the multi-time scale of the next data point according to a pre-established prediction model and the electricity load data of the multi-time scale of the current data point, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of the emergency.
Optionally, the electricity load data of the multiple time scales at least comprises time series data of electricity load amounts of five time scales of minutes, hours, days, weeks and months.
Optionally, the state probability transition matrix for eliminating the influence of the emergency is obtained by the following steps:
under the condition of accessing a new electricity load or an electricity load with an emergency, acquiring sample electricity load data of multiple time scales under different time spans;
and optimizing the state probability transition matrix of the multi-element time scale according to the influence difference of the sample electricity load data to obtain the state probability transition matrix for eliminating the influence of the emergency.
Optionally, the optimizing the state probability transition matrix of the multiple time scales according to the influence difference of the sample electrical load data to obtain the state probability transition matrix for eliminating the influence of the emergency event includes:
acquiring small time scale electricity load data points in the sample electricity load data;
determining the fluctuation degree corresponding to the electric load data points of different time scales according to the outlier factors of the electric load data points of the small time scales;
determining the burst degree of the power utilization load data points with different scales according to the fluctuation degree;
And optimizing the weight of the multi-element time scale corresponding to the burst degree according to the change trend of the power consumption load data point to obtain the state probability transition matrix for eliminating the influence of the burst event.
Optionally, the determining the fluctuation degree corresponding to the electrical load data points of different time scales according to the outlier factor of the electrical load data points of the small time scales includes:
calculating outlier factors of all user load data points corresponding to the small time scale according to the user load data points of the small time scale;
and determining the fluctuation degree corresponding to the power utilization load data points of different time scales according to the outlier factors of the user load data points and the data point variances of the different data points in the preset distance neighborhood.
Optionally, the determining the fluctuation degree corresponding to the electricity load data point of different time scales according to the outlier factor of each user load data point and the data point variance of different data points of different time scales in the preset distance neighborhood includes:
measuring the fluctuation degree of the data point by the data point deviation measured by the average link distance of the data point in the K distance neighborhood and the overall fluctuation of the data point in the K distance;
Degree of fluctuation beta for the jth data point at the ith time scale ij ：
β i,j ＝Norm(COF ij ·σ ij )
Wherein: norm represents that for a data point in the ith time scale, max-min normalization is performed by the COF outlier factor for all data points in that scale; sigma (sigma) ij Representing the variance of the data points in the K-distance neighborhood for the jth data point in the ith time scale.
Optionally, determining the burst degree of the electrical load data points with different scales according to the fluctuation degree includes:
and determining the mutation degree of the data points in the minimum time scale in the multi-time scale according to the fluctuation degree corresponding to the power consumption load data points in different time scales, the minimum scale in the multi-time scale and the maximum scale in the multi-time scale.
Optionally, determining the mutation degree of the data point in the minimum time scale in the multiple time scales according to the fluctuation degree corresponding to the power consumption load data points in different time scales, the minimum scale in the multiple time scales and the maximum scale in the multiple time scales includes:
degree of mutation for data points in the smallest time scale in the multivariate time scaleThe calculation method is as follows:
wherein: beta i,j : representing the extent of fluctuation of the jth data point in the ith time scale;
mini: representing the smallest scale of the multiple time scales;
maxi: representing the largest scale of the multiple time scales;
norms: the normalization process in the above equation is a normalization process with the smallest data point in the smallest time scale.
Optionally, the optimizing the weight of the multiple time scales corresponding to the emergency degree according to the variation trend of the electrical load data point to obtain the state probability transition matrix for eliminating the influence of the emergency, including:
for the burst degree of each data point, acquiring the fluctuation degree from the previous data point to the current data point;
according to the trend change of the fluctuation degree, determining trend influence factors under different time scales;
optimizing trend influence factors under different scales according to fluctuation differences of each data point on each time scale to obtain scale weights of state probability transition matrixes of different data points under different time scales;
and carrying out normalization processing on the scale weights of the state probability transition matrixes of different data points under different time scales to obtain the state probability transition matrix for eliminating the influence of the emergency.
Optionally, the trend influencing factor ε ij The method comprises the following steps:
wherein: beta i,j : indicating the extent of fluctuation of the jth data point at the ith time scale.
In a second aspect, an embodiment of the present invention provides an electrical load prediction apparatus for a micro-grid, the apparatus including:
the acquisition module is used for acquiring the power utilization load data of the multi-element time scale of the current data point under the condition of accessing a new power utilization load or a power utilization load with an emergency;
and the prediction module is used for predicting the electricity load data of the multi-time scale of the next data point according to a pre-established prediction model and the electricity load data of the multi-time scale of the current data point, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of the emergency.
Optionally, the electricity load data of the multiple time scales at least comprises time series data of electricity load amounts of five time scales of minutes, hours, days, weeks and months.
Optionally, the apparatus further comprises a setup module for:
under the condition of accessing a new electricity load or an electricity load with an emergency, acquiring sample electricity load data of multiple time scales under different time spans;
And optimizing the state probability transition matrix of the multi-element time scale according to the influence difference of the sample electricity load data to obtain the state probability transition matrix for eliminating the influence of the emergency.
Optionally, the establishing module is configured to:
acquiring small time scale electricity load data points in the sample electricity load data;
determining the fluctuation degree corresponding to the electric load data points of different time scales according to the outlier factors of the electric load data points of the small time scales;
determining the burst degree of the power utilization load data points with different scales according to the fluctuation degree;
and optimizing the weight of the multi-element time scale corresponding to the burst degree according to the change trend of the power consumption load data point to obtain the state probability transition matrix for eliminating the influence of the burst event.
Optionally, the establishing module is configured to:
calculating outlier factors of all user load data points corresponding to the small time scale according to the user load data points of the small time scale;
and determining the fluctuation degree corresponding to the power utilization load data points of different time scales according to the outlier factors of the user load data points and the data point variances of the different data points in the preset distance neighborhood.
Optionally, the establishing module is configured to:
measuring the fluctuation degree of the data point by the data point deviation measured by the average link distance of the data point in the K distance neighborhood and the overall fluctuation of the data point in the K distance;
degree of fluctuation beta for the jth data point at the ith time scale ij ：
β i,j ＝Norm(COF ij ·σ ij )
Wherein: norm represents that for a data point in the ith time scale, max-min normalization is performed by the COF outlier factor for all data points in that scale; sigma (sigma) ij Representing the variance of the data points in the K-distance neighborhood for the jth data point in the ith time scale.
Optionally, the establishing module is configured to:
and determining the mutation degree of the data points in the minimum time scale in the multi-time scale according to the fluctuation degree corresponding to the power consumption load data points in different time scales, the minimum scale in the multi-time scale and the maximum scale in the multi-time scale.
Optionally, the establishing module is configured to:
degree of mutation for data points in the smallest time scale in the multivariate time scaleThe calculation method is as follows:
wherein: beta i,j : representing the extent of fluctuation of the jth data point in the ith time scale;
mini: representing the smallest scale of the multiple time scales;
maxi: representing the largest scale of the multiple time scales;
norms: the normalization process in the above equation is a normalization process with the smallest data point in the smallest time scale.
Optionally, the establishing module is configured to:
for the burst degree of each data point, acquiring the fluctuation degree from the previous data point to the current data point;
according to the trend change of the fluctuation degree, determining trend influence factors under different time scales;
optimizing trend influence factors under different scales according to fluctuation differences of each data point on each time scale to obtain scale weights of state probability transition matrixes of different data points under different time scales;
and carrying out normalization processing on the scale weights of the state probability transition matrixes of different data points under different time scales to obtain the state probability transition matrix for eliminating the influence of the emergency.
Optionally, the trend influencing factor ε ij The method comprises the following steps:
wherein: beta i,j : indicating the extent of fluctuation of the jth data point at the ith time scale.
In a third aspect, an embodiment of the present invention provides a terminal device, including: at least one processor and memory;
The memory stores a computer program; the at least one processor executes the computer program stored by the memory to implement the method for microgrid electrical load prediction provided in the first aspect.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having stored therein a computer program that when executed implements the electrical load prediction method for a micro grid provided in the first aspect.
The embodiment of the invention has the following advantages:
the embodiment of the invention provides a method, a device, a terminal device and a readable storage medium for predicting the power consumption load of a micro-grid, which are used for acquiring power consumption load data of a multi-element time scale of a current data point under the condition of accessing a new power consumption load or a power consumption load with an emergency; and predicting the power consumption load data of the next data point according to a pre-established prediction model and the power consumption load data of the current data point in a multi-time scale, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of an emergency, the distribution weights of the state probability transition matrices in different time scales are adjusted through the trend change of the data points in the multi-time scale, so that a prediction model is obtained, the power consumption load data is predicted through the prediction model, and the prediction accuracy is improved.
Drawings
FIG. 1 is a flow chart of steps of an embodiment of a method for microgrid electrical load prediction in accordance with the present invention;
FIG. 2 is a schematic diagram of the different time scales of the present invention;
FIG. 3 is a block diagram of an embodiment of an electrical load prediction apparatus for a micro-grid according to the present invention;
fig. 4 is a schematic structural view of a terminal device of the present invention.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Micro-Grid (Micro-Grid) is also a Micro-Grid, and refers to a small power generation and distribution system composed of a distributed power supply, an energy storage device, an energy conversion device, a load, a monitoring and protecting device and the like.
The proposal of the micro-grid aims to realize flexible and efficient application of the distributed power supply and solve the problem of grid connection of the distributed power supply with huge quantity and various forms. The development and extension of the micro-grid can fully promote the large-scale access of the distributed power supply and the renewable energy sources, realize the high-reliability supply of various energy forms of loads, and be an effective way for realizing an active power distribution network, so that the traditional power grid is transited to the intelligent power grid.
The micro-grid at least comprises a direct-current micro-grid, an alternating-current and direct-current hybrid micro-grid, a medium-voltage distribution branch micro-grid and a low-voltage micro-grid, wherein:
the direct-current micro-grid comprises a distributed power supply, an energy storage device, a load and the like which are all connected to a direct-current bus, and the direct-current network is connected to an external alternating-current power grid through a power electronic inversion device. The direct-current micro-grid can provide electric energy for alternating-current and direct-current loads with different voltage levels through the power electronic conversion device, and the fluctuation of the distributed power supply and the loads can be regulated on the direct-current side through the energy storage device.
The ac microgrid comprises a distributed power source, an energy storage device and the like, which are all connected to an ac busbar through a power electronic device. Ac microgrids remain the primary form of microgrid. Through the control of the switch at the PCS, the conversion between the grid-connected operation of the micro-grid and the island mode can be realized.
The AC/DC hybrid micro-grid comprises an AC bus and a DC bus, and can directly supply power to an AC load and a DC load.
The medium-voltage distribution branch micro-grid effectively integrates distributed power sources and loads on the basis of the medium-voltage distribution branch and is suitable for supplying power to user areas with medium capacity, higher power supply reliability requirements and more concentration.
The low-voltage micro-grid is formed by properly integrating the distributed power supply and the load of the user on the low-voltage level, and most of the micro-grid is owned by the power or energy user, so that the scale is relatively small.
The electric load of the micro-grid is increased sharply due to weather factors (such as sudden high temperature, low temperature, storm and other severe weather conditions) in the process of predicting the electric load of the micro-grid. When the hidden Markov model is built to predict the power consumption load of the micro-grid, sudden large changes of the power consumption load can cause severe changes of the hidden state of the model, abnormal values appear in time sequence data of the power consumption load, and over fitting or under fitting conditions which cause the accuracy of the model to be reduced appear.
In the existing method, the state probability transition matrix in the original single time scale can be optimized through the state probability transition matrix of the multi-time scale, and the state probability transition matrix of the single time point is measured through a plurality of time scales in the hidden Markov model, so that the influence of sudden change in a certain time scale is reduced.
After the state probability transition in the multi-element time scale is introduced into the hidden Markov model, the probability transition matrix in the small time scale can be corrected through the state probability transition matrix in the large time scale, so that the influence of sudden and severe changes is reduced. However, when a new load is accessed into the micro-grid, a sudden increase of the load can occur, and the influence in the multi-element time scale can be like a sudden and severe increase, and in the multi-element time scale optimization, the sudden increase of the part can be reduced, so that the prediction model of the electric load of the micro-grid is inaccurate.
In view of the above, an embodiment of the present invention provides a method for predicting an electrical load of a micro grid, for predicting the electrical load. The execution main body of the embodiment is an electric load prediction device for a micro-grid, and the electric load prediction device is arranged on terminal equipment, wherein the terminal equipment at least comprises a computer, a tablet terminal and the like.
Referring to fig. 1, there is shown a flowchart of steps of an embodiment of a method for predicting electrical loads for a micro-grid according to the present invention, the method may specifically include the steps of:
s101, under the condition of accessing a new electricity load or an electricity load with an emergency, acquiring electricity load data of a multi-element time scale of a current data point;
specifically, under the condition of accessing a new electricity load or an electricity load with an emergency, the terminal equipment acquires the electricity load data of the current data point under the current condition, wherein the electricity load data is in a multi-time scale, for example, the time sequence data of the electricity load under five time scales of minutes, hours, days, weeks and months is extracted.
S102, predicting the electricity load data of the multi-time scale of the next data point according to a pre-established prediction model and the electricity load data of the multi-time scale of the current data point, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of the emergency.
Specifically, the terminal equipment collects sample electricity load data of multiple time scales through a micro-grid electricity load data detection system, optimizes a state probability transition matrix of the multiple time scales through influence differences of new access loads on electricity loads of emergencies under different time spans, acquires a state probability transition matrix for eliminating the influence of the emergencies, and establishes a hidden Markov model, namely a prediction model, through the optimized state probability transition matrix.
After the hidden Markov model is built, the terminal equipment acquires the monitored micro-grid power consumption load data point in real time, and adopts the prediction model to predict the power consumption load data of the next data point.
According to the embodiment of the invention, the multi-element time scale weighting is introduced through the load fluctuation trend difference of the new load access and the emergency influence in the time scales of different lengths, and the influence of the large time scale on the probability transition matrix of the small time scale is adaptively adjusted according to the trend difference of the new load access and the emergency in the long time span so as to avoid erroneous judgment.
According to the method for predicting the electric load of the micro-grid, provided by the embodiment of the invention, under the condition of accessing a new electric load or an electric load with an emergency, the electric load data of the current data point in a multi-time scale are obtained; and predicting the power consumption load data of the next data point according to a pre-established prediction model and the power consumption load data of the current data point in a multi-time scale, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of an emergency, the distribution weights of the state probability transition matrices in different time scales are adjusted through the trend change of the data points in the multi-time scale, so that a prediction model is obtained, the power consumption load data is predicted through the prediction model, and the prediction accuracy is improved.
Optionally, the state probability transition matrix for eliminating the influence of the emergency is obtained by the following way:
under the condition of accessing a new electricity load or an electricity load with an emergency, acquiring sample electricity load data of multiple time scales under different time spans;
and optimizing the state probability transition matrix of the multi-element time scale according to the influence difference of the sample electricity load data to obtain the state probability transition matrix for eliminating the influence of the emergency.
Specifically, the electric load time sequence data of a plurality of time scales are collected through a micro-grid electric load data detection system. In the grid management system, for the micro-grid needing to establish a prediction model, the time sequence data of the electricity load quantity under five time scales of minutes, hours, days, weeks and months are extracted, as shown in fig. 2.
And optimizing the state probability transition matrix of the multi-element time scale through the influence difference of the power load of the new access load under different time spans of the emergency, and obtaining the state probability transition matrix for eliminating the influence of the emergency.
Optionally, optimizing the state probability transition matrix of the multiple time scales according to the influence difference of the sample electricity load data to obtain a state probability transition matrix for eliminating the influence of the emergency, including:
Acquiring small time scale electricity load data points in sample electricity load data;
determining the fluctuation degree corresponding to the electricity load data points of different time scales according to the outlier factors of the electricity load data points of the small time scales;
determining the burst degree of the power utilization load data points with different scales according to the fluctuation degree;
and optimizing the weight of the multi-element time scale corresponding to the burst degree according to the change trend of the power consumption load data points to obtain a state probability transition matrix for eliminating the influence of the burst event.
Specifically, after historical data of the micro-grid power consumption load under a multi-time scale, which is required to be established by a prediction model, is acquired, firstly, fluctuation degrees of data points are required to be acquired through outlier factors of micro-grid power consumption load monitoring data under a small time scale.
After the fluctuation degree of the data points is obtained, the burst degree of the data points in the small time scale can be obtained through probability state transition of the fluctuation degree of the data points in the small time scale and the data points in the large time scale.
After the burst degree of the data point is obtained, the respective weights of the multiple time scales can be obtained through the burst degree, but in the process of eliminating the influence on the emergency, the influence on the data point information accessed by the new load similar to the emergency is reduced through a state probability transition matrix under a large time scale, so that the real-time weight distribution adjustment is required to be carried out through the change trend of the data point under a subsequent small time scale on the basis of the evaluation of the current data point in the process of weight distribution. Therefore, the prediction model can keep the load change information of the new load access power grid on the basis of eliminating the influence of the emergency, and correct prediction is carried out.
The method comprises the following specific steps:
a. the fluctuation degree of the data points is obtained through the outlier factors and the local change amplitude of the power utilization load data points under the small time scale.
b. The burst degree of the data points is obtained through the difference of the fluctuation degree of the data points under the multi-element time scale.
c. And carrying out self-adaptive optimization on the multi-element time scale matching weight under the burst degree through the trend change of the data points, and obtaining a state probability transition matrix after the optimization on the small time scale.
Optionally, determining the fluctuation degree corresponding to the electrical load data points of different time scales according to the outlier factor of the electrical load data points of the small time scales includes:
calculating outlier factors of all user load data points corresponding to the small time scale according to the user load data points of the small time scale;
and determining the fluctuation degree corresponding to the power utilization load data points of different time scales according to the outlier factors of the user load data points and the data point variances of the different data points in the preset distance neighborhood.
Specifically, the fluctuation degree of the data point is obtained through the outlier factor and the local change amplitude of the power utilization load data point under a small time scale, which specifically comprises the following steps:
After the historical time sequence data of the micro-grid electricity load under the multi-element time scale, namely the sample electricity load data, is acquired, a multi-element time scale optimization method applied to the time sequence data is acquired through the historical time sequence data, so that future prediction of a hidden Markov model acquired through the historical time sequence data is optimized.
In the multi-time scale micro-grid electricity load time sequence data, the time stamp interval of time sequence data points under a small time scale is small and is sensitive to the electricity load change of an emergency, so that the initial fluctuation assessment of the data points needs to be carried out in the small time scale, and the fluctuation degree is assessed through the fluctuation condition difference under the corresponding large time scale.
In electrical load timing data, fluctuations in the timing data can be assessed by the COF outlier factor of the data points. COF outliers for data points at each time scale are measured by their correspondence. When the outlier degree obtained by normalization of the COF outlier factors under different scales is used as the fluctuation degree of the data points, the K distance neighborhood is selected as k=10 data points closest to the data point to be judged in the COF calculation process. Degree of fluctuation beta for the jth data point at the ith time scale ij ：
β i,j ＝Norm(COF ij ·σ ij )
Wherein: norm represents that for a data point in the ith time scale, max-min normalization is performed by the COF outlier factor for all data points in that scale.
σ ij Representing the variance of the data points in the K-distance neighborhood for the jth data point in the ith time scale. In the above formula, the fluctuation degree of the data point is measured by the deviation of the data point measured by the average link distance of the data point in the adjacent area of the K distance and the integral fluctuation of the data point in the K distance.
Optionally, determining the burst degree of the electrical load data points of different scales according to the fluctuation degree comprises:
and determining the mutation degree of the data points in the minimum time scale in the multi-time scale according to the fluctuation degree corresponding to the power consumption load data points in different time scales, the minimum scale in the multi-time scale and the maximum scale in the multi-time scale.
Specifically, the burst degree of the data point is obtained through the difference of the fluctuation degree of the data point under the multi-element time scale, and the method specifically comprises the following steps:
after the fluctuation degree of the data points is obtained, the fluctuation degree of the data points in the small time scale and the fluctuation degree of the data points under the corresponding large time scale can be measured according to the difference, so that the burst degree of the data points in the small time scale is obtained, for the burst degree, when the electricity load of the micro-grid is influenced by an emergency, large fluctuation occurs, but the fluctuation caused by the emergency is gradually reduced in the data points in the large time scale, and for the data points (the fluctuation of the electricity load is larger) which are caused by the normal load of the micro-grid, the fluctuation degree of the data points is correspondingly increased, so that the fluctuation degree of the data points can be measured through the fluctuation difference degree under the multi-element time scale, and the fluctuation degree of the data points is measured from the small time scale to the large time scale because the fluctuation of the data of the small time scale is most obvious.
Degree of mutation for data points in the smallest time scale in the multivariate time scaleThe calculation method is as follows:
wherein: beta i,j : representing the extent of fluctuation of the jth data point in the ith time scale.
mini: representing the smallest scale, i.e. the smallest time unit, of the multiple time scales.
maxi: representing the largest scale, i.e. the largest time unit, of the multiple time scales.
Norms: the normalization process in the above equation is normalized by the smallest data point in the smallest time scale.
The sum of fluctuation degree differences of data points in different time scales is used as a measurement basis of mutation degree. Since the degree of fluctuation corresponding to the data points in each time scale is a result after normalization. Therefore, the sum of fluctuation differences of the data points corresponding to each other in different time scales can be used for indicating whether the fluctuation degree of the data points in the minimum time scale has the same fluctuation degree in the large time scale, and when the difference of the fluctuation degrees is larger, the higher the mutation degree of the data points in the small time scale is (the more likely that the power consumption load of the micro-grid is severely fluctuated due to an emergency).
Optionally, optimizing the weight of the multiple time scales corresponding to the emergency degree according to the variation trend of the power consumption load data points to obtain a state probability transition matrix for eliminating the influence of the emergency, including:
For the burst degree of each data point, acquiring the fluctuation degree from the previous data point to the current data point;
according to the trend change of the fluctuation degree, determining trend influence factors under different time scales;
optimizing trend influence factors under different scales according to fluctuation differences of each data point on each time scale to obtain scale weights of state probability transition matrixes of different data points under different time scales;
and carrying out normalization processing on the scale weights of the state probability transition matrixes of different data points under different time scales to obtain the state probability transition matrix for eliminating the influence of the emergency.
Specifically, the adaptive optimization is performed on the multi-element time scale matching weight under the burst degree through the trend change of the data points, and the state probability transition matrix after the optimization of the small time scale is obtained, which specifically comprises the following steps:
after the mutation degree of the data points is obtained in the process, the weight distribution of the multi-element time scale state probability transition matrix can be carried out through the fluctuation degree difference of different scales in the mutation degree.
In the process of performing multi-time scale weight allocation according to the burst degree acquired in the process, because the fluctuation of the data point when the burst event occurs is consistent with the fluctuation of the sudden access of the new load at the beginning of the fluctuation, that is, the burst degree of the data point cannot distinguish the data point from the fluctuation of the sudden access of the new load.
And carrying out self-adaptive adjustment on the data point distribution weight on the fluctuation degree of the data points in continuous data point changes through the trend difference of the emergency and the new load access.
For the weight distribution of the state probability transition matrix of the multi-element time scale, the sudden degree is adjusted at the position of starting abnormal fluctuation of the data points, but with the addition of new data points, the adjustment is needed through the numerical change trend of the data points, so for the sudden degree of each data point, the weight distribution of the state probability transition matrix of the current data point through the fluctuation degree difference between the multi-element time scale is needed through the fluctuation degree change of the previous data point to the current data point. For each time scale, a decrease in the degree of fluctuation indicates that the weight of the state probability transition matrix at that scale should be lower, thereby highlighting the state probability transition matrix with high difference.
Trend impact factor epsilon as described for this process ij The method comprises the following steps:
wherein: beta i,j : indicating the extent of fluctuation of the jth data point at the ith time scale.
For trend influence factors in the weight distribution process of the state probability transition matrix of the data points in the small time scale, judgment needs to be carried out through trend change between the current data point and the previous data point, and the fluctuation degree difference of all time scales is measured. And normalizing all time scales by softmax so that the sum of trend impact factors at all time scales is 1. It is ensured that the sum can be made to be one in the weight allocation of the subsequent state probability transition matrix.
After the trend influence factor epsilon is obtained i,j Then, carrying out normalized distribution of weights through fluctuation difference of the data points on each time scale and optimizing through trend influence factors, and carrying out scale weight xi of a state probability transition matrix of the jth data point under the ith time scale i,j ：
The above equation optimizes the base scale weights by trend impact factors by normalizing the softmax for the extent of burst of each data point at each time scale as the base scale weight. The sum of all scales is guaranteed to be one by normalization.
Based on the above, the distribution weights of the state probability transition matrixes of different time scales are adjusted through the trend change of the data points under the multi-time scales, and compared with the distribution weights of the local fluctuation conditions of the data points in the different time scales in the above process, the distribution weights can be adaptively regulated and controlled through the fluctuation degree change trend of the data points, so that the power consumption load fluctuation of an emergency event is distinguished from the power consumption load fluctuation caused by new load access in the actual prediction process, and the accuracy of the power consumption load prediction of the micro-grid is improved.
After the scale weights for the multiple time scales of data points are obtained, then the computation of the state probability transition matrix at each time scale may begin. By the scale weight ζ of each data point at each time scale i,j And (5) carrying out weighted fusion: the weighted probability of transition from state i to state j is calculated, namely: p' (j|i) =ε 1,j ·P 1 (a|b)+ε 2,j ·P 2 (a|b)+…+ε maxi,j ·P n (a|b)
Wherein: p (P) 1 (a|b)、P 2 (a|b)…P n (a|b) represents the state transition probabilities, ε, respectively at different time scales i,j Is the weight of each time scale.
Normalizing the fused state probability transition matrix again: since the sum of each row of the state probability transition matrix may no longer be equal to 1 after the weighted fusion, the normalization process needs to be performed again.
In the embodiment of the invention, in the process of predicting the electric load of the micro-grid by using the hidden Markov model, the influence of emergency on the model is eliminated through information fusion of a state probability transition matrix of a multi-time scale, the load change information of a new load access grid is reserved on the basis, the fluctuation degree of a data point is obtained through the outlier factor and the local change amplitude of an electric load data point under a small time scale, the burst degree of the data point is obtained through the difference of the fluctuation degree of the data point under the multi-time scale, the multi-time scale matching weight under the burst degree is adaptively optimized through the trend change of the data point, and the state probability transition matrix optimized for the small time scale is obtained.
According to the embodiment of the invention, the weight distribution among the state probability transition matrixes of each time scale is carried out through the time sequence data fluctuation difference under the multiple scales, compared with the prior weight distribution through the power consumption load historical data, the self-adaptive weight acquisition can be carried out through the fluctuation of the data points, the influence of the power consumption load sharp fluctuation caused by the emergency can be reduced in the prediction model, so that the influence of the emergency can be reduced when the hidden Markov prediction model encounters the emergency, and the prediction of the next state can be accurately carried out.
According to the embodiment of the invention, the distribution weights of the state probability transition matrixes in different time scales are adjusted through trend changes of the data points in the multiple time scales, and compared with the distribution weights of the local fluctuation situations of the data points in the different time scales in the process, the distribution weights can be adaptively regulated and controlled through fluctuation degree change trends of the data points, so that the power consumption load fluctuation of an emergency event is distinguished from the power consumption load fluctuation caused by new load access in the actual prediction process, and the accuracy of the power consumption load prediction of the micro-grid is improved.
It should be noted that, for simplicity of description, the method embodiments are shown as a series of acts, but it should be understood by those skilled in the art that the embodiments are not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred embodiments, and that the acts are not necessarily required by the embodiments of the invention.
According to the method for predicting the electric load of the micro-grid, provided by the embodiment of the invention, under the condition of accessing a new electric load or an electric load with an emergency, the electric load data of the current data point in a multi-time scale are obtained; and predicting the power consumption load data of the next data point according to a pre-established prediction model and the power consumption load data of the current data point in a multi-time scale, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of an emergency, the distribution weights of the state probability transition matrices in different time scales are adjusted through the trend change of the data points in the multi-time scale, so that a prediction model is obtained, the power consumption load data is predicted through the prediction model, and the prediction accuracy is improved.
Another embodiment of the present invention provides an electrical load prediction apparatus for a micro-grid, configured to perform the electrical load prediction method for a micro-grid provided in the foregoing embodiment.
Referring to fig. 3, there is shown a block diagram of an embodiment of an electrical load prediction apparatus for a micro-grid according to the present invention, which may specifically include the following modules: an acquisition module 301 and a prediction module 302, wherein:
the acquisition module 301 is configured to acquire power consumption load data of a multiple time scale of a current data point under the condition of accessing a new power consumption load or a power consumption load with an emergency;
the prediction module 302 is configured to predict the electrical load data of the multiple time scales of the next data point according to a pre-established prediction model and the electrical load data of the multiple time scales of the current data point, where the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of the emergency.
According to the power consumption load prediction device for the micro-grid, provided by the embodiment of the invention, under the condition of accessing a new power consumption load or a power consumption load with an emergency event, power consumption load data of a multi-element time scale of a current data point are obtained; and predicting the power consumption load data of the next data point according to a pre-established prediction model and the power consumption load data of the current data point in a multi-time scale, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of an emergency, the distribution weights of the state probability transition matrices in different time scales are adjusted through the trend change of the data points in the multi-time scale, so that a prediction model is obtained, the power consumption load data is predicted through the prediction model, and the prediction accuracy is improved.
The invention further provides a device for predicting the electric load of the micro-grid, which is provided by the embodiment.
Optionally, the electricity load data of the multiple time scales at least comprises time series data of electricity load amounts of five time scales of minutes, hours, days, weeks and months.
Optionally, the apparatus further comprises a setup module for:
under the condition of accessing a new electricity load or an electricity load with an emergency, acquiring sample electricity load data of multiple time scales under different time spans;
and optimizing the state probability transition matrix of the multi-element time scale according to the influence difference of the sample electricity load data to obtain the state probability transition matrix for eliminating the influence of the emergency.
Optionally, the establishing module is configured to:
acquiring small time scale electricity load data points in sample electricity load data;
determining the fluctuation degree corresponding to the electricity load data points of different time scales according to the outlier factors of the electricity load data points of the small time scales;
determining the burst degree of the power utilization load data points with different scales according to the fluctuation degree;
and optimizing the weight of the multi-element time scale corresponding to the burst degree according to the change trend of the power consumption load data points to obtain a state probability transition matrix for eliminating the influence of the burst event.
Optionally, the establishing module is configured to:
calculating outlier factors of all user load data points corresponding to the small time scale according to the user load data points of the small time scale;
and determining the fluctuation degree corresponding to the power utilization load data points of different time scales according to the outlier factors of the user load data points and the data point variances of the different data points in the preset distance neighborhood.
Optionally, the establishing module is configured to:
measuring the fluctuation degree of the data point by the data point deviation measured by the average link distance of the data point in the K distance neighborhood and the overall fluctuation of the data point in the K distance;
degree of fluctuation beta for the jth data point at the ith time scale ij ：
β i,j ＝Norm(COF ij ·σ ij )
Wherein: norm represents that for a data point in the ith time scale, max-min normalization is performed by the COF outlier factor for all data points in that scale; sigma (sigma) ij Representing the variance of the data points in the K-distance neighborhood for the jth data point in the ith time scale.
Optionally, the establishing module is configured to:
and determining the mutation degree of the data points in the minimum time scale in the multi-time scale according to the fluctuation degree corresponding to the power consumption load data points in different time scales, the minimum scale in the multi-time scale and the maximum scale in the multi-time scale.
Optionally, the establishing module is configured to:
degree of mutation for data points in the smallest time scale in the multivariate time scaleThe calculation method is as follows:
wherein: beta i,j : representing the extent of fluctuation of the jth data point in the ith time scale;
mini: representing the smallest scale of the multiple time scales;
maxi: representing the largest scale of the multiple time scales;
norms: the normalization process in the above equation is a normalization process with the smallest data point in the smallest time scale.
Optionally, the establishing module is configured to:
for the burst degree of each data point, acquiring the fluctuation degree from the previous data point to the current data point;
according to the trend change of the fluctuation degree, determining trend influence factors under different time scales;
optimizing trend influence factors under different scales according to fluctuation differences of each data point on each time scale to obtain scale weights of state probability transition matrixes of different data points under different time scales;
and carrying out normalization processing on the scale weights of the state probability transition matrixes of different data points under different time scales to obtain the state probability transition matrix for eliminating the influence of the emergency.
Optionally, the trend influencing factor ε ij The method comprises the following steps:
wherein: beta i,j : indicating the extent of fluctuation of the jth data point at the ith time scale.
It should be noted that, in this embodiment, each of the possible embodiments may be implemented separately, or may be implemented in any combination without conflict, which is not limited to the implementation of the present application.
For the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points.
According to the power consumption load prediction device for the micro-grid, provided by the embodiment of the invention, under the condition of accessing a new power consumption load or a power consumption load with an emergency event, power consumption load data of a multi-element time scale of a current data point are obtained; and predicting the power consumption load data of the next data point according to a pre-established prediction model and the power consumption load data of the current data point in a multi-time scale, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of an emergency, the distribution weights of the state probability transition matrices in different time scales are adjusted through the trend change of the data points in the multi-time scale, so that a prediction model is obtained, the power consumption load data is predicted through the prediction model, and the prediction accuracy is improved.
An embodiment of the present invention provides a terminal device, configured to execute the electrical load prediction method for a micro-grid provided in the embodiment.
Fig. 4 is a schematic structural view of a terminal device of the present invention, as shown in fig. 4, the terminal device includes: at least one processor 401 and a memory 402;
the memory stores a computer program; the at least one processor executes the computer program stored in the memory to implement the electrical load prediction method for the micro grid provided by the above embodiment.
The terminal equipment provided by the embodiment obtains the power consumption load data of the current data point in a multi-time scale under the condition of accessing a new power consumption load or a power consumption load with an emergency; and predicting the power consumption load data of the next data point according to a pre-established prediction model and the power consumption load data of the current data point in a multi-time scale, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of an emergency, the distribution weights of the state probability transition matrices in different time scales are adjusted through the trend change of the data points in the multi-time scale, so that a prediction model is obtained, the power consumption load data is predicted through the prediction model, and the prediction accuracy is improved.
Yet another embodiment of the present application provides a computer readable storage medium having a computer program stored therein, which when executed implements the electrical load prediction method for a micro grid provided in any one of the above embodiments.
According to the computer readable storage medium of the present embodiment, by accessing a new electricity load or an electricity load with an emergency, electricity load data of a multiple time scale of a current data point is acquired; and predicting the power consumption load data of the next data point according to a pre-established prediction model and the power consumption load data of the current data point in a multi-time scale, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of an emergency, the distribution weights of the state probability transition matrices in different time scales are adjusted through the trend change of the data points in the multi-time scale, so that a prediction model is obtained, the power consumption load data is predicted through the prediction model, and the prediction accuracy is improved.
In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.
It will be apparent to those skilled in the art that embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the invention may take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, electronic devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing electronic device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing electronic device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or electronic device that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or electronic device. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or electronic device that comprises the element.
The above description of the present invention provides a method for predicting electrical loads for micro-grids and a device for predicting electrical loads for micro-grids, and specific examples are applied to illustrate the principles and embodiments of the present invention, and the above description of the examples is only used to help understand the method and core ideas of the present invention; meanwhile, as those skilled in the art will vary in the specific embodiments and application scope according to the idea of the present invention, the present disclosure should not be construed as limiting the present invention in summary.
Claims (8)
1. A method for electrical load prediction for a microgrid, the method comprising:
under the condition of accessing a new electricity load or an electricity load with an emergency, acquiring electricity load data of a multi-element time scale of a current data point;
predicting the electricity load data of the multi-time scale of the next data point according to a pre-established prediction model and the electricity load data of the multi-time scale of the current data point, wherein the pre-established prediction model is determined according to a state probability transition matrix for eliminating the influence of an emergency;
the state probability transition matrix for eliminating the influence of the emergency is obtained by the following steps:
under the condition of accessing a new electricity load or an electricity load with an emergency, acquiring sample electricity load data of multiple time scales under different time spans;
optimizing a state probability transition matrix of a multi-element time scale according to the influence difference of the sample electricity load data to obtain the state probability transition matrix for eliminating the influence of the emergency;
optimizing the state probability transition matrix of the multi-time scale according to the influence difference of the sample electricity load data to obtain the state probability transition matrix for eliminating the influence of the emergency, wherein the method comprises the following steps:
Acquiring small time scale electricity load data points in the sample electricity load data;
determining the fluctuation degree corresponding to the electric load data points of different time scales according to the outlier factors of the electric load data points of the small time scales;
determining the burst degree of the power utilization load data points with different scales according to the fluctuation degree;
and optimizing the weight of the multi-element time scale corresponding to the burst degree according to the change trend of the power consumption load data point to obtain the state probability transition matrix for eliminating the influence of the burst event.
2. The electrical load prediction method for a micro-grid according to claim 1, wherein the multi-component time scale electrical load data comprises at least five time scale electrical load amount time series data of minutes, hours, days, weeks, and months.
3. The method for micro-grid electrical load prediction according to claim 1, wherein determining the degree of fluctuation corresponding to the electrical load data points of different time scales according to the outlier factor of the electrical load data points of the small time scales comprises:
calculating outlier factors of all user load data points corresponding to the small time scale according to the user load data points of the small time scale;
And determining the fluctuation degree corresponding to the power utilization load data points of different time scales according to the outlier factors of the user load data points and the data point variances of the different data points in the preset distance neighborhood.
4. A method for microgrid electrical load prediction according to claim 3, wherein said determining the degree of fluctuation corresponding to the electrical load data points of different time scales based on the outlier factor of each user load data point and the variance of the data points of different data points in the preset distance neighborhood comprises:
by data points atData point deviation from +.>Measuring the fluctuation degree of the data points by the overall fluctuation of the data points in the distance;
for the first∈1 on the individual time scale>Degree of fluctuation of data points->：
Wherein the method comprises the steps ofIndicate +.>Data points in a time scale by +.>Outlier factor in->Normalizing; />Indicate->The>Data points are +.>Variance of data points in the distance neighborhood.
5. The electrical load prediction method for a micro-grid according to claim 1, wherein determining the burst degree of electrical load data points of different scales according to the fluctuation degree comprises:
And determining the mutation degree of the data points in the minimum time scale in the multi-time scale according to the fluctuation degree corresponding to the power consumption load data points in different time scales, the minimum scale in the multi-time scale and the maximum scale in the multi-time scale.
6. The method for predicting electrical loads for a micro grid according to claim 5, wherein determining the mutation degree of the data points in the minimum time scale in the multi-time scale according to the fluctuation degree corresponding to the electrical load data points in the different time scales, the minimum scale in the multi-time scale and the maximum scale in the multi-time scale comprises:
degree of mutation for data points in the smallest time scale in the multivariate time scaleThe calculation method is as follows:
wherein:: indicate->The>The degree of fluctuation of the data points;
: representing the smallest scale of the multiple time scales;
: representing the largest scale of the multiple time scales;
: the normalization process in the above equation is a normalization process with the smallest data point in the smallest time scale.
7. The method for predicting electrical loads of a micro-grid according to claim 1, wherein optimizing the weights of the multiple time scales corresponding to the burst degree according to the trend of the electrical load data points to obtain the state probability transition matrix for eliminating the influence of the burst event comprises:
For the burst degree of each data point, acquiring the fluctuation degree from the previous data point to the current data point;
according to the trend change of the fluctuation degree, determining trend influence factors under different time scales;
optimizing trend influence factors under different scales according to fluctuation differences of each data point on each time scale to obtain scale weights of state probability transition matrixes of different data points under different time scales;
and carrying out normalization processing on the scale weights of the state probability transition matrixes of different data points under different time scales to obtain the state probability transition matrix for eliminating the influence of the emergency.
8. The electrical load prediction method for a micro-grid according to claim 7, wherein the trend influencing factorThe method comprises the following steps:
wherein:: indicate->∈1 on the individual time scale>Degree of fluctuation of data points.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310838578.3A CN116845878B (en) | 2023-07-10 | 2023-07-10 | Electric load prediction method for micro-grid |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310838578.3A CN116845878B (en) | 2023-07-10 | 2023-07-10 | Electric load prediction method for micro-grid |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116845878A CN116845878A (en) | 2023-10-03 |
| CN116845878B true CN116845878B (en) | 2024-01-26 |
Family
ID=88161383
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310838578.3A Active CN116845878B (en) | 2023-07-10 | 2023-07-10 | Electric load prediction method for micro-grid |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116845878B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117648554B (en) * | 2024-01-29 | 2024-04-26 | 山东德源电力科技股份有限公司 | Intelligent data acquisition method for photovoltaic multifunctional circuit breaker |
| CN117878927B (en) * | 2024-03-11 | 2024-05-28 | 国网黑龙江省电力有限公司绥化供电公司 | Power system load trend analysis method based on time sequence analysis |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110503256A (en) * | 2019-08-14 | 2019-11-26 | 北京国网信通埃森哲信息技术有限公司 | Short-term load forecasting method and system based on big data technology |
| CN114444775A (en) * | 2021-12-30 | 2022-05-06 | 中国电力科学研究院有限公司 | User power load characteristic prediction method and device |
-
2023
- 2023-07-10 CN CN202310838578.3A patent/CN116845878B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110503256A (en) * | 2019-08-14 | 2019-11-26 | 北京国网信通埃森哲信息技术有限公司 | Short-term load forecasting method and system based on big data technology |
| CN114444775A (en) * | 2021-12-30 | 2022-05-06 | 中国电力科学研究院有限公司 | User power load characteristic prediction method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116845878A (en) | 2023-10-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116845878B (en) | Electric load prediction method for micro-grid | |
| Menke et al. | Distribution system monitoring for smart power grids with distributed generation using artificial neural networks | |
| CN107453396B (en) | Multi-objective optimization scheduling method for output of distributed photovoltaic power supply | |
| CN109840633B (en) | Photovoltaic output power prediction method, system and storage medium | |
| CN111049193B (en) | Standby demand dynamic evaluation method for multiple scheduling scenes of wind power system | |
| CN116125361B (en) | Voltage transformer error evaluation method, system, electronic equipment and storage medium | |
| JP2016067098A (en) | Power system monitoring device and power system monitoring system | |
| Vergura et al. | Inferential statistics for monitoring and fault forecasting of PV plants | |
| CN110969306A (en) | Power distribution low-voltage distribution area load prediction method and device based on deep learning | |
| CN117411011A (en) | Flexible power load regulating system of multiple power generation systems | |
| CN114915261A (en) | Fault monitoring method and device for photovoltaic power station | |
| CN116125204A (en) | Fault prediction system based on power grid digitization | |
| Memmel et al. | Forecast of renewable curtailment in distribution grids considering uncertainties | |
| CN112101646A (en) | Temperature rise fault early warning method for fan water cooling system | |
| CN113746425B (en) | Photovoltaic inverter parameter anomaly analysis method and system | |
| CN115001149A (en) | Energy storage control method and device and microgrid | |
| CN113689057A (en) | Photovoltaic power generation power prediction method and device | |
| JP6679980B2 (en) | Power demand forecasting apparatus and power demand forecasting method | |
| TW201727559A (en) | Management method and system of renewable energy power plant checking whether the power generation of a renewable energy power plant is normal according to the estimated power generation amount | |
| CN113657032B (en) | Low-frequency load shedding method and system for pre-centralized coordination real-time distribution control | |
| KR102480440B1 (en) | Management system of solar power plant | |
| CN115409644A (en) | Method and device for calculating power generation power of photovoltaic power station | |
| Menke | A comprehensive approach to implement monitoring and state estimation in distribution grids with a low number of measurements | |
| CN113255994A (en) | Optimal configuration method, device, equipment and medium for power grid CPS | |
| CN111697575B (en) | Flexibility-improved quantitative analysis method, device and 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A Method for Predicting Electricity Load in Microgrids Granted publication date: 20240126 Pledgee: Zhejiang Hangzhou Yuhang Rural Commercial Bank Co.,Ltd. Wuchang Branch Pledgor: Hangzhou Qizhi Energy Technology Co.,Ltd. Registration number: Y2024980021196 |