Disclosure of Invention
The purpose of the invention is: the accurate power utilization management method for the fishery power consumers is provided, and the power utilization of the fishery power consumers and the safety of the power grids where the fishery power consumers are located are guaranteed.
The technical scheme of the invention is as follows: the invention discloses an accurate power utilization management method for fishery power users, which comprises the following steps:
collecting time-sharing data of n fishery power users including dissolved oxygen content, water temperature, PH value and load power in summer cloudy days and rainy days as bottom data, and correspondingly obtaining n data curves of the dissolved oxygen content, the water temperature, the PH value and the load power; n is an integer of not less than 2;
respectively aggregating n data curves of the dissolved oxygen content, the water temperature, the PH value and the load power into 1 class and 4 classes in total to respectively obtain characteristic curves of the dissolved oxygen content, the water temperature, the PH value and the load power, and taking each class center as a characteristic curve of a fishery power user;
thirdly, acquiring bottom layer data of the fishery power users according with the data characteristics according to the characteristic curves of the fishery power users, adopting Euclidean distance to compare the acquired curve similarity between the fishery load user data curves and the fishery load user characteristic curves, and screening out curve sections with high similarity as fishery power user sample data;
acquiring grid-connected data of L new energy power supplies including wind power and solar power generation to obtain L data curves of the new energy power supplies for grid connection;
respectively clustering the L data curves into 1 class according to the difference of wind power and photoelectricity, correspondingly obtaining characteristic curves of wind power access and photoelectricity access, and taking each class center as a characteristic curve of new energy access data;
acquiring new energy access data of the line where the fishery power user is located according to the characteristic curve of the new energy access data, wherein the new energy access data is in accordance with the data characteristics and serves as new energy access sample data;
collecting line data and transformer data of a line where the fishery power consumer is located, and removing load data in fishery power consumer sample data from the line data to obtain the line data with the fishery power consumer load data removed;
and calculating data relation among fishery power user sample data, new energy access sample data, line data without fishery power user load data and transformer data, judging whether fishery power users to be optimized meet power utilization management precision requirements or not according to the data relation, and if not, performing power utilization management precision optimization on the users.
The further scheme is as follows: the step II comprises the following specific steps:
i, carrying out standardized processing on bottom layer data of n fishery power users by adopting a formula (1);
wherein x is
ij' jth underlying data, x, for ith fishery power consumer
ijIs x
ij' the value after the normalization process is performed,
is the average value of all the bottom-layer data of the ith fishery power consumer,
the maximum value of the ith fishery power consumer bottom layer data is k, the bottom layer data volume of each fishery power consumer is k, for example, the value of k can be 24, and n is an integer not less than 2;
II, clustering the bottom data of the n standardized fishery power users by adopting a fuzzy C mean algorithm, wherein the fuzzy C mean algorithm adopts a clustering number parameter of 1;
in the step, the objective function of the fuzzy C-means algorithm is an equation (2);
wherein n is the number of samples to be classified, namely the number of fishery power users, c is a clustering number parameter, c is 1, d
ijIs a sample point x
jI.e. the jth underlying data and the cluster center p of the ith class
iEuclidean distance between, d
ij=||x
j-p
iAnd | l, m is a weighting index and is used for controlling the sharing degree of the samples among the mode classes, the larger the value is, the better the noise suppression effect is, and u is
ijIs a sample point x
jMembership to class i and satisfies
In the step, the fuzzy index m is 2, the maximum iteration number N is 100, and the initial clustering center p is determined(0)Then, adjusting the value of the clustering center through the formula (3), calculating a membership matrix through the formula (4) to perform iteration, and stopping the iteration when the iteration number is equal to the set maximum iteration number to obtain the clustering center of the characteristic curves of the dissolved oxygen content, the water temperature, the PH value and the load power as the characteristic curve of the fishery power user;
the further scheme is as follows: the fifth step comprises the following specific steps:
III, standardizing access data of the L new energy sources by adopting a formula (5);
wherein, y
ij' is the j-th access data (such as grid-connected power) in the ith new energy resource, y
ijIs y
ij' the value after the normalization process is performed,
is the average value of the access data of the ith new energy source,
the maximum value of the access data of the ith new energy is k, the value of k is an integer of 24, and L is not less than 2;
IV, clustering the access data of the L new energy sources subjected to the standardized processing by adopting a fuzzy C mean algorithm, wherein the fuzzy C mean algorithm adopts a clustering number parameter of 1;
in the step, the objective function of the fuzzy C-means algorithm is an equation (6);
wherein, L is the number of samples to be classified (i.e. the number of new energy sources in this step), c is a parameter of the number of clusters, c is 1, d
ijIs a sample point y
j(in this step, the jth new energy access data) and the ith cluster center p
iHas an Euclidean distance d between
ij=||y
j-p
iAnd | l, m is a weighting index and is used for controlling the sharing degree of the samples among the mode classes, the larger the value is, the better the noise suppression effect is, and u is
ijIs a sample point y
jMembership to class i and satisfies
In the step, the fuzzy index m is 2, the maximum iteration number N is 100, and the initial clustering center p is determined(0)Then, adjusting the value of the clustering center through a formula (7), calculating a membership matrix through a formula (8) to perform iteration, and stopping iteration when the iteration number is equal to the set maximum iteration number to obtain the clustering center of the characteristic curve of the new energy access data as the characteristic curve of the new energy access data;
the further scheme is as follows: the step eight comprises the following specific steps:
acquiring line data and transformer data of a line where a fishery power consumer is located, wherein fishery power consumer load data is to be eliminated from the line data, and obtaining fishery power consumer sample data, new energy access sample data, line data with fishery power consumer load data eliminated and data relation between the transformer data by adopting an equation (9);
setting transformer capacity as S, fishery power user sample data PaThe total load of the line except the load data of fishery power users is P, the power factor is cos phi, and the output of new energy is PneThen, it needs to satisfy:
judging whether the fishery user to be optimized meets the power utilization management precision requirement or not according to the data relation, if the user meets the data relation, the power utilization management of the user meets the precision requirement, and if not, performing power utilization management precision optimization on the user;
VI, optimizing the users which do not meet the precision requirement by adopting the formulas (10) to (13);
a. when S is<(P+Pa-Pne) When/cos phi, the electricity consumption of fishery power users is adjusted, and the P is reduceda;
b. When 0 is present>(P+Pa-Pne) When cos phi is exceeded, the power consumption of fishery power consumer is adjusted and P is increaseda;
c. When P is presentaWhen the peak P occurs simultaneously, the power consumption of fishery power users is adjusted to realize peak staggering;
wherein: when case a occurs, note:
λ=(P+Pa-Pne)/cosφ-S (10);
setting a margin coefficient to be 1.2 for reserving a certain margin, and then adjusting the fishery power user power at the moment as follows:
Pa′=(Pa/cosφ-1.2·λ)·cosφ (11);
when case b occurs, note:
γ=(Pne-P+Pa)/cosφ (12);
setting a margin coefficient to be 1.2 for reserving a certain margin, and then adjusting the fishery power user power at the moment as follows:
Pa′=(Pa/cosφ+1.2·λ)·cosφ (13);
when the condition is satisfied
When the condition c occurs, the electricity consumption of fishery power consumers is adjusted to P
aThe peak is adjusted to realize peak staggering.
The invention has the positive effects that: the fishery power consumer electricity utilization accurate management method comprises the steps of obtaining bottom data of fishery power consumers related to normal work of loads, based on the bottom data, adopting a fuzzy clustering algorithm to conduct clustering analysis on fishery power consumer characteristics, achieving fishery power consumer electricity utilization characteristic extraction, conducting accurate electricity utilization management on fishery power consumers by introducing seasonal factors and new energy access factors, facilitating achieving accurate load management, and achieving accurate electricity utilization management of fishery power consumers from the aspects of economy, safety, user satisfaction and the like according to user use habits and load characteristic identification results.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
(example 1)
Referring to fig. 1, the precise power consumption management method for fishery power consumers in the embodiment is mainly implemented by the following steps:
collecting time-sharing data of n fishery power users including dissolved oxygen content, water temperature, PH value and load power in summer cloudy days and rainy days as bottom data, and correspondingly obtaining n data curves of the dissolved oxygen content, the water temperature, the PH value and the load power; n is an integer of not less than 2;
for each selected fishery power consumer, the dissolved oxygen content, the water temperature, the PH value and the load power data of the date with the weather condition of cloudy days and rainy days in summer are adjusted, and by taking 1 hour as a span, each fishery power consumer can respectively call 24 data of the dissolved oxygen content, the water temperature, the PH value and the load power to respectively obtain n data curves of the dissolved oxygen content, the water temperature, the PH value and the load power.
Respectively aggregating n data curves of the dissolved oxygen content, the water temperature, the PH value and the load power into 1 class and 4 classes in total to respectively obtain characteristic curves of the dissolved oxygen content, the water temperature, the PH value and the load power, and taking each class center as a characteristic curve of a fishery power user, wherein the method specifically comprises the following steps:
i, carrying out standardized processing on bottom layer data of n fishery power users by adopting a formula (1);
wherein x is
ij' jth underlying data, x, for ith fishery power consumer
ijIs x
ij' the value after the normalization process is performed,
is the average value of all the bottom-layer data of the ith fishery power consumer,
the maximum value of the ith fishery power consumer bottom layer data is k, the bottom layer data volume of each fishery power consumer is k, for example, the value of k can be 24, and n is an integer not less than 2;
II, clustering the bottom data of the n standardized fishery power users by adopting a fuzzy C mean algorithm, wherein the fuzzy C mean algorithm adopts a clustering number parameter of 1;
in the step, the objective function of the fuzzy C-means algorithm is an equation (2);
wherein n is the number of samples to be classified, namely the number of fishery power users, c is a clustering number parameter, c is 1, d
ijIs a sample point x
jI.e. the jth underlying data and the cluster center p of the ith class
iEuclidean distance between, d
ij=||x
j-p
iAnd | l, m is a weighting index and is used for controlling the sharing degree of the samples among the mode classes, the larger the value is, the better the noise suppression effect is, and u is
ijIs a sample point x
jMembership to class i and satisfies
In the step, the fuzzy index m is 2, the maximum iteration number N is 100, and the initial clustering center p is determined(0)Then, adjusting the value of the clustering center through the formula (3), calculating a membership matrix through the formula (4) to perform iteration, and stopping the iteration when the iteration number is equal to the set maximum iteration number to obtain the clustering center of the characteristic curves of the dissolved oxygen content, the water temperature, the PH value and the load power as the characteristic curve of the fishery power user;
and thirdly, acquiring bottom layer data of the fishery power users according with the data characteristics according to the characteristic curves of the fishery power users, adopting Euclidean distance to compare the acquired curve similarity between the fishery load user data curve and the fishery load user characteristic curve, and screening out a curve section with high similarity as fishery power user sample data.
Acquiring grid-connected data of L new energy power supplies including wind power and solar power generation to obtain L data curves of the new energy power supplies for grid connection;
respectively clustering the L data curves into 1 class according to the difference of wind power and photoelectricity, correspondingly obtaining characteristic curves of wind power access and photoelectricity access, and taking each class center as a characteristic curve of new energy access data;
in the step, a fuzzy clustering algorithm is adopted, a new energy access characteristic curve is extracted according to the called data information, the description of the new energy access data characteristics is realized, specifically, a fuzzy C mean value algorithm is adopted as the clustering algorithm, the access data of L new energies are clustered into 1 class, and are copolymerized into 2 classes according to the difference between wind power and photoelectricity, so that the characteristic curves of wind power access and photoelectricity access are respectively obtained and are used as the characteristic curves of the new energy access data, and the method specifically comprises the steps III and IV;
III, standardizing access data of the L new energy sources by adopting a formula (5);
wherein, y
ij' is the j-th access data (such as grid-connected power) in the ith new energy resource, y
ijIs y
ij' the value after the normalization process is performed,
is the average value of the access data of the ith new energy source,
the maximum value of the access data of the ith new energy is k, the value of k is an integer of 24, and L is not less than 2;
IV, clustering the access data of the L new energy sources subjected to the standardized processing by adopting a fuzzy C mean algorithm, wherein the fuzzy C mean algorithm adopts a clustering number parameter of 1;
in the step, the objective function of the fuzzy C-means algorithm is an equation (6);
wherein, L is the number of samples to be classified (i.e. the number of new energy sources in this step), c is a parameter of the number of clusters, c is 1, d
ijIs a sample point y
j(in this step, the jth new energy access data) and the ith cluster center p
iHas an Euclidean distance d between
ij=||y
j-p
iI, m is plusThe weight index is used for controlling the sharing degree of the samples among the mode classes, the larger the value is, the better the noise suppression effect is, and u is
ijIs a sample point y
jMembership to class i and satisfies
In the step, the fuzzy index m is 2, the maximum iteration number N is 100, and the initial clustering center p is determined(0)Then, adjusting the value of the clustering center through a formula (7), calculating a membership matrix through a formula (8) to perform iteration, and stopping iteration when the iteration number is equal to the set maximum iteration number to obtain the clustering center of the characteristic curve of the new energy access data as the characteristic curve of the new energy access data;
and sixthly, acquiring new energy access data of the line where the fishery power user is located according with the data characteristics according to the characteristic curve of the new energy access data, and taking the new energy access data as new energy access sample data.
Collecting line data and transformer data of a line where the fishery power consumer is located, and removing load data in fishery power consumer sample data from the line data to obtain the line data with the fishery power consumer load data removed;
calculating fishery power user sample data, new energy access sample data, line data without fishery power user load data and data relation between transformer data (related data are processed according to seasonal factors), judging whether fishery users to be optimized meet power consumption management precision requirements or not according to the data relation, if the users meet the data relation, the user power consumption management meets the precision requirements, if not, the users are optimized precisely in power consumption management, and the method specifically comprises the following steps:
acquiring line data and transformer data of a line where a fishery power consumer is located, wherein fishery power consumer load data is to be eliminated from the line data, and obtaining fishery power consumer sample data, new energy access sample data, line data with fishery power consumer load data eliminated and data relation between the transformer data by adopting an equation (9);
setting transformer capacity as S, fishery power user sample data PaThe total load of the line except the load data of fishery power users is P, the power factor is cos phi, and the output of new energy is PneThen, it needs to satisfy:
judging whether the fishery user to be optimized meets the power utilization management precision requirement or not according to the data relation, if the user meets the data relation, the power utilization management of the user meets the precision requirement, and if not, performing power utilization management precision optimization on the user;
VI, optimizing the users which do not meet the precision requirement by adopting the formulas (10) to (13);
a. when S is<(P+Pa-Pne) When/cos phi, the electricity consumption of fishery power users is adjusted, and the P is reduceda;
b. When 0 is present>(P+Pa-Pne) When cos phi is exceeded, the power consumption of fishery power consumer is adjusted and P is increaseda;
c. When P is presentaWhen the peak P occurs simultaneously, the power consumption of fishery power users is adjusted to realize peak staggering;
wherein: when case a occurs, note:
λ=(P+Pa-Pne)/cosφ-S (10);
setting a margin coefficient to be 1.2 for reserving a certain margin, and then adjusting the fishery power user power at the moment as follows:
Pa′=(Pa/cosφ-1.2·λ)·cosφ (11);
when case b occurs, note:
γ=(Pne-P+Pa)/cosφ (12);
setting a margin coefficient to be 1.2 for reserving a certain margin, and then adjusting the fishery power user power at the moment as follows:
Pa′=(Pa/cosφ+1.2·λ)·cosφ (13);
when the condition is satisfied
When the condition c occurs, the electricity consumption of fishery power consumers is adjusted to P
aThe peak is adjusted to realize peak staggering.
The fishery power consumer electricity utilization accurate management method of the embodiment acquires bottom data which relates to normal work of loads of fishery power consumers, based on the bottom data, the fishery power consumer characteristics are subjected to cluster analysis by adopting a fuzzy clustering algorithm, fishery power consumer electricity utilization characteristic extraction is realized, accurate electricity utilization management is carried out on fishery power consumers by introducing seasonal factors and new energy access factors, accurate load management is favorably realized, the fishery power consumer electricity utilization accurate management can be realized according to user use habits and load characteristic identification results, and the fishery power consumer electricity utilization management accuracy is realized from the aspects of economy, safety, user satisfaction and the like.
The above embodiments are illustrative of specific embodiments of the present invention, and are not restrictive of the present invention, and those skilled in the relevant art can make various changes and modifications without departing from the spirit and scope of the present invention to obtain corresponding equivalent technical solutions, and therefore all equivalent technical solutions should be included in the scope of the present invention.