Enterprise-oriented electricity consumption intelligent monitoring method and system

Abstract

The invention relates to the technical field of enterprise electricity consumption monitoring, in particular to an enterprise-oriented electricity consumption intelligent monitoring method and system. The method comprises the following steps: firstly, acquiring a phase difference in power consumption data of each terminal of an enterprise; obtaining a relation topology network of all terminals; further in the target regulation and control period, according to the similar fluctuation characteristics of the phase difference of any two terminals in the adjacent regulation and control segments, a terminal relation graph is obtained in combination with the relation topology network; further dividing the terminal to obtain a plurality of relational communities based on a robust algorithm; and finally, adjusting the electric energy redundancy according to the association direction among all terminals in each relational community and the severe fluctuation characteristic of the phase difference in the newest regulation and control segment.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise electricity consumption monitoring, and in particular to an enterprise-oriented intelligent electricity consumption monitoring method and system. Background Art

[0002] Usually, electricity consumption monitoring is mainly aimed at large enterprises, which have multiple electricity terminals according to the production links and equipment terminals. Through real-time monitoring and data analysis, high-energy consumption links or terminals can be identified, and electricity coordination and scheduling can be carried out to ensure that the company can maintain normal output when the electricity consumption is relatively small at the same time, avoiding problems such as circuit fuse burning or load caused by excessive electricity consumption at the same time.

[0003] Too many devices started at the same time will lead to a surge in electricity consumption. If excessive power redundancy is set, the company's operating costs will increase significantly. When the redundancy is insufficient, switching large equipment such as motors will cause grid fluctuations, which may cause damage to other equipment. At the same time, there are complex production relationships between different equipment terminals, so it is necessary to set appropriate power redundancy, and then coordinate and dispatch electricity to ensure enterprise production. Summary of the invention

[0004] In order to solve the technical problem of unreasonable setting of power redundancy for enterprise equipment terminals, the purpose of the present invention is to provide an enterprise-oriented power consumption intelligent monitoring method and system, and the technical solution adopted is as follows: An intelligent monitoring method for electricity consumption of an enterprise, the method comprising: Obtain the phase difference in the power consumption data of each terminal of the enterprise; obtain the control segment with a preset time domain length, and use the latest two adjacent control segments as a target control cycle; obtain the relationship topology network of all the terminals, and set the weights between all the terminals under the same diversion device to the preset weights; In the target control cycle, according to the fluctuation similarity characteristics of the phase difference between any two terminals in adjacent control segments, the relationship strength and the association direction of the corresponding two terminals are obtained; based on the relationship topology network, the terminal relationship graph is obtained in combination with the relationship strength and the association direction; Dividing the terminals based on the Louvain algorithm to obtain multiple relationship communities; Adjusting the power redundancy according to the association direction between all the terminals in each of the relationship communities and the dramatic fluctuation characteristics of the phase difference in the latest regulation segment; The division process includes: in the target control cycle, according to the fluctuation difference characteristics of the phase difference between any two terminals, obtaining the difference parameters between the corresponding two terminals; before the first iteration of the Louvain algorithm, dividing the edge terminals according to the difference parameters between the edge terminals and the corresponding adjacent terminals in the terminal relationship graph; In each iteration, in any of the regulation segments in the target regulation cycle, the terminals are clustered according to the fluctuation similarity characteristics of the phase difference between the terminals in each community to obtain terminal clustering clusters; the terminal clustering clusters of any two communities are nearest neighbor matched, and the corresponding two communities are merged according to the similarity characteristics of the matched terminal clustering clusters to obtain the network structure of each iteration, and the Louvain algorithm is used for iteration again to finally obtain multiple relationship communities.

[0005] Furthermore, according to the fluctuation similarity characteristics of the phase difference between any two terminals in adjacent control segments, the method for obtaining the relationship strength and the correlation direction of the corresponding two terminals includes: Select any two of the terminals as the first terminal and the second terminal respectively, and use the two control segments in the target control cycle as the first segment and the second segment respectively in time sequence; Obtaining a correlation coefficient between the phase difference of the first segment of the first terminal and the phase difference of the second segment of the second terminal as a first correlation coefficient; obtaining a correlation coefficient between the phase difference of the second segment of the first terminal and the phase difference of the first segment of the second terminal as a second correlation coefficient; The maximum value of the first correlation coefficient and the second correlation coefficient is selected as the relationship strength between the corresponding two terminals; the association direction is: from the terminal of the first segment in the relationship strength to the terminal of the second segment.

[0006] Furthermore, the method for obtaining the terminal relationship diagram includes: When the relationship strength is lower than a preset strength threshold, the relationship strength is reset to zero; and the sum of the relationship strength and the preset weight is used as the relationship weight corresponding to the two terminals; Based on the relationship topology network, in combination with the relationship weight and the association direction, a terminal relationship graph is obtained.

[0007] Furthermore, the method for obtaining the difference parameter includes: In the target control cycle, the phase difference is divided into a phase difference sequence based on the maximum value of each phase difference; each maximum value in the phase difference corresponds to the phase difference sequence with the maximum value itself as the starting point; The phase difference sequences in any two of the terminals are randomly paired to form a binary group; according to the difference characteristics of the two maxima corresponding to each binary group and the DTW distance of the two phase difference sequences corresponding to each binary group, a difference sub-parameter of each binary group is obtained; the difference characteristics of the maxima and the DTW distance are both positively correlated with the difference sub-parameter; The minimum value of the difference sub-parameters corresponding to any two of the terminals is used as the difference parameter between the corresponding two terminals.

[0008] Furthermore, the method for dividing the edge terminals according to the difference parameters between the edge terminals and the corresponding adjacent terminals in the terminal relationship graph includes: When the difference parameter between the edge terminal and the corresponding adjacent terminal is greater than a preset difference threshold, the corresponding edge terminal is deleted; when the difference parameter between the edge terminal and the corresponding adjacent terminal is less than or equal to the preset difference threshold, the corresponding edge terminal is divided into an exclusive community.

[0009] Furthermore, the method for obtaining the terminal clustering includes: In any community, the variance and mean of the phase difference of the terminal in the first regulation segment are used to construct a two-dimensional space; density clustering is performed in the two-dimensional space to obtain terminal clusters.

[0010] Furthermore, the method for obtaining the network structure of each iteration includes: Select any two communities as a first community and a second community respectively; obtain a matching pair of the terminal clusters of the first community and the second community; In the matching pair, obtaining similarity parameters of the matching pair according to the difference parameters between the terminals of the two terminal clusters; taking the average of all the similar parameters corresponding to the first community and the second community as a merging parameter of the first community and the second community; when the merging parameter is greater than a preset merging threshold, merging the corresponding two communities; All the communities are merged to obtain the network structure of each iteration.

[0011] Furthermore, the method for adjusting the power redundancy includes: In any of the relationship communities, the starting terminal of the association direction is recorded as the cause terminal, and the end point is recorded as the result terminal; the variance of the phase difference of each cause terminal in the latest regulation segment is normalized and used as the power redundancy coefficient of the corresponding result terminal; and the power redundancy amount is adjusted based on the power redundancy coefficient.

[0012] Furthermore, the method for adjusting the power redundancy amount based on the power redundancy coefficient includes: The rated power of the result terminal is set as the basic power consumption, and the product of the basic power consumption and the power redundancy coefficient is used as the adjusted power redundancy.

[0013] The present invention also proposes an intelligent electricity consumption monitoring system for enterprises, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements any one of the steps of the intelligent electricity consumption monitoring method for enterprises when executing the computer program.

[0014] The present invention has the following beneficial effects: The present invention first obtains the phase difference in the electricity consumption data of each terminal of the enterprise, and uses the phase difference to capture the production status of the terminal equipment, providing a basis for subsequent analysis; further obtains the control segment with a preset time domain length, and uses the latest two adjacent control segments as a target control cycle, which is convenient for analyzing the transmission characteristics of power consumption; further obtains the relationship topology network of all terminals, sets the weights between all terminals under the same shunt device as preset weights, and facilitates the analysis of the connection characteristics between the terminals with the help of the topological relationship between the terminals, and prepares for the subsequent acquisition of the terminal relationship diagram; further in the target control cycle, according to the fluctuation similarity characteristics of the phase difference between any two terminals in adjacent control segments, and combined with the relationship topology network, the terminal relationship diagram is obtained, which reflects the transmission characteristics of electric energy from the perspective of phase difference fluctuation, and shows the relationship between the terminals. The association direction and relationship strength of the terminals are obtained to provide a basis for finally setting a suitable power redundancy; further, based on the Louvain algorithm, the terminals are divided into multiple relationship communities, which further highlights the connection between terminal relationships and the transmission between terminal relationships and relationships, and is suitable for network division with relationships; finally, according to the association direction between all terminals in each relationship community and the drastic fluctuation characteristics of the phase difference in the latest regulation segment, the power redundancy is adjusted to eliminate the relationship interference between devices that do not belong to the same relationship community, improve the accuracy of power regulation, and ensure the safety of enterprise production; the present invention obtains the power transmission relationship between terminals by analyzing the phase difference of the terminals, and divides the relationship communities with the help of the Louvain algorithm, reduces the interference between devices in different relationship communities, and accurately regulates the power redundancy, providing a guarantee for the safe production of enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A flow chart of an enterprise-oriented intelligent power consumption monitoring method provided by an embodiment of the present invention; Figure 2 A flowchart of dividing a relationship community provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of an intelligent power consumption monitoring method and system for enterprises proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The specific scheme of the intelligent power consumption monitoring method and system for enterprises provided by the present invention is described in detail below with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a flow chart of an enterprise-oriented power consumption intelligent monitoring method provided by an embodiment of the present invention, specifically comprising: Step S1: Obtain the phase difference in the power consumption data of each terminal of the enterprise; obtain the control segment with a preset time domain length, and take the latest two adjacent control segments as a target control cycle; obtain the relationship topology network of all terminals, and set the weights between all terminals under the same diversion device to the preset weights.

[0021] The change in the phase difference of the enterprise's equipment load has a great impact on the overall power grid, and can more sensitively reflect the production status of each terminal equipment in the enterprise. Therefore, this application chooses to analyze the change in phase difference, and first obtains the phase difference in the electricity consumption data of each terminal of the enterprise. Considering that the product is transmitted at different equipment terminals during the production process, the energy consumption will also change over time, so the control segment is obtained with a preset time domain length, and the latest two adjacent control segments are used as a target control cycle to facilitate the analysis of the transmission characteristics of energy consumption.

[0022] The different equipment terminals in an enterprise are similar to the network topology, that is, starting from one or more transformers, electricity is directly transmitted to the power terminal or shunt device, and then the shunt device transmits the electric energy to each terminal device through other shunt devices; then in the production process, if the load of any of the devices changes, it can affect more all the devices under the current shunt device, and then the device terminals under the upper-level shunt device, so the graph structure constructed when constructing the relationship topology network of the terminal is a weighted graph, and the weights between all terminal devices under the same shunt device are initially set to preset weights.

[0023] As an example, a voltage and current monitoring module is installed on each terminal to obtain phase difference data, with a preset time domain length of 1 hour and a preset weight of 0.5.

[0024] It should be noted that the method of constructing a relational topology network is already a prior art. In other embodiments of the present invention, the implementer may also set other preset time domain lengths and other preset weights, and the value range of the preset weights is 0-1.

[0025] Step S2: In the target control cycle, according to the fluctuation similarity characteristics of the phase difference between any two terminals in adjacent control segments, the relationship strength and correlation direction of the corresponding two terminals are obtained; based on the relationship topology network, the relationship strength and correlation direction are combined to obtain the terminal relationship diagram.

[0026] Taking into account that when the terminal equipment starts to process the product, the phase difference increases due to the increase of the inductive load, and when the production ends, the phase difference decreases, the fluctuation characteristics of the phase difference reflect the power consumption of the terminal. The more similar the phase difference fluctuations of different terminals in adjacent time periods are, the more obvious the power transmission characteristics are, and the stronger the correlation between the terminals. At the same time, the sequence implies the correlation direction of the terminals. Therefore, in the target control cycle, according to the similar fluctuation characteristics of the phase difference between any two terminals in adjacent control segments, the relationship strength and correlation direction of the corresponding two terminals are obtained, and further combined with the relationship topology network, the terminal relationship diagram is obtained, which shows the connection relationship and influence relationship between the terminals, providing a basis for finally setting the appropriate power redundancy.

[0027] Preferably, in one embodiment of the present invention, any two terminals are selected as the first terminal and the second terminal respectively, and two control segments in the target control cycle are respectively used as the first segment and the second segment in time sequence; Considering that the larger the correlation coefficient between the data, the stronger the correlation between the data and the more obvious the similarity characteristics, the fluctuation similarity characteristics of the phase difference are expressed by the correlation coefficient; in order to determine the load transfer order between the first terminal and the second terminal, both transfer directions are analyzed, and finally the direction with the largest correlation coefficient is selected. Based on this: Obtaining a correlation coefficient between a phase difference of a first segment of the first terminal and a phase difference of a second segment of the second terminal as a first correlation coefficient; obtaining a correlation coefficient between a phase difference of a second segment of the first terminal and a phase difference of a first segment of the second terminal as a second correlation coefficient; The maximum value between the first correlation coefficient and the second correlation coefficient is selected as the relationship strength between the corresponding two terminals; the association direction is: from the terminal of the first segment in the relationship strength to the terminal of the second segment.

[0028] It should be noted that, in one embodiment of the present invention, the correlation coefficient is the Pearson correlation coefficient; in other embodiments of the present invention, the implementer may also obtain the DTW similarity to represent the fluctuation similarity characteristics of the phase difference, and use the maximum DTW similarity of the two terminals as the correlation strength. The Pearson correlation coefficient and DTW similarity are both technical means well known to those skilled in the art, and will not be described in detail here.

[0029] Preferably, in one embodiment of the present invention, considering that when the relationship strength between two terminals is low, it indicates that the possibility of association between the two terminals is low, so when the relationship strength is lower than a preset strength threshold, the relationship strength is reset to zero; the sum of the relationship strength and the preset weight is used as the relationship weight corresponding to the two terminals; Based on the relationship topology network, the terminal relationship graph is obtained by combining the relationship weight and association direction.

[0030] It should be noted that there is no preset weight assigned between terminals that are not in the same shunt device. The relationship strength is used as the relationship weight between terminals. For example, the preset strength threshold is 0.3, AB is in the same shunt device, the initial preset weight is 0.5, and the calculated relationship strength is 0.8, then the relationship weight is 1.3; AC is not in the same shunt device, the calculated relationship strength is 0.6, then the relationship weight is 0.6; BC is not in the same shunt device, the calculated relationship strength is 0.2, then it is corrected to 0. In the terminal relationship diagram, the connection between devices with a relationship weight of zero needs to be discarded to simplify the terminal relationship diagram and reduce interference.

[0031] Step S3: Divide the terminals based on the Louvain algorithm to obtain multiple relationship communities.

[0032] In the embodiment of the present invention, after obtaining the terminal relationship graph, the Louvain algorithm is used to divide the terminal relationship graph into communities. The relationship community represents a whole with relationships, which further highlights the connection between terminal relationships and the transmission between terminal relationships. It is more suitable for network division with relationships. Therefore, multiple relationship communities are obtained by dividing the terminals based on the Louvain algorithm. Please refer to Figure 2, which shows a flow chart of dividing a relationship community provided by an embodiment of the present invention, and the dividing process includes: Step S301: In the target control cycle, according to the fluctuation difference characteristics of the phase difference between any two terminals, the difference parameters between the corresponding two terminals are obtained; before the first iteration of the Louvain algorithm, the edge terminals are divided according to the difference parameters between the edge terminals and the corresponding adjacent terminals in the terminal relationship diagram.

[0033] Considering that the Leuven algorithm first treats each terminal as a separate community, and calculates the size of the modularity increment of the terminal added to other communities based on the idea of ​​hierarchical clustering to merge the communities, at this time, for the edge terminals in the terminal relationship graph, that is, there is only one adjacent connected terminal, since the community of the edge terminal is more obvious, the modularity increment will be calculated in the Leuven algorithm, resulting in increased computational complexity; at the same time, some terminals belong to the same link of the same production line, that is, some terminals have the same changes in the current regulation segment, and the transmitted terminals are also the same, then these terminals cannot be merged as a whole, but will be merged in sequence according to the idea of ​​hierarchical clustering, and the computational efficiency is low, so it is necessary to improve the Leuven algorithm to achieve rapid division of terminals.

[0034] Therefore, in the target control cycle, the difference parameters between the two terminals are obtained according to the fluctuation difference characteristics of the phase difference between any two terminals, and the difference between the terminals is quantified to provide a basis for rapid terminal division. In order to eliminate the influence of edge terminals on the Leuven algorithm, before the first iteration of the Leuven algorithm, the edge terminals are divided according to the difference parameters between the edge terminals and the corresponding adjacent terminals in the terminal relationship diagram.

[0035] Preferably, in one embodiment of the present invention, considering that the time range of the target regulation cycle is relatively long and there may be multiple repeated production processes, firstly, in the target regulation cycle, the phase difference is divided into a phase difference sequence based on the maximum value of each phase difference; each maximum value in the phase difference corresponds to a phase difference sequence starting from the maximum value itself; Considering that in the production process of an enterprise, there may be multiple terminals in the same production link, that is, the equipment is the same and the products are the same, then the changes in phase difference are also highly similar; considering that there may be multiple production lines, the time of terminals in the same production link in different production lines may not be synchronized, and considering that it takes time for products to flow between different terminals, in order to analyze the characteristics of power transfer and thus analyze the correlation between terminals, the phase difference sequences must be matched and compared arbitrarily; considering that the maximum value of the phase difference sequence represents the power consumption characteristics of the terminal, the DTW distance between the phase difference sequences reflects the differences in the phase difference sequences, based on this: The phase difference sequences in any two terminals are randomly paired to form a binary group; the difference sub-parameter of each binary group is obtained according to the difference characteristics of the two maxima corresponding to each binary group and the DTW distance of the two phase difference sequences corresponding to each binary group; the difference characteristics of the maxima and the DTW distance are both positively correlated with the difference sub-parameter; The minimum value of the difference sub-parameters corresponding to any two terminals is used as the difference parameter between the corresponding two terminals.

[0036] As an example, the calculation formula of the difference parameter includes: ; Wherein, P is the serial number of the terminal; Q is the serial number of the terminal; ; represents the difference parameter between the Pth terminal and the Qth terminal; p is the sequence number of the phase difference sequence of the Pth terminal, , represents the number of phase difference sequences of the Pth terminal; q is the sequence number of the phase difference sequence of the Qth terminal, , represents the number of phase difference sequences of the Qth terminal; represents the maximum value in the pth phase difference sequence; represents the maximum value in the qth phase difference sequence; represents the DTW distance between the p-th phase difference sequence and the q-th phase difference sequence; represents the difference sub-parameter between the p-th phase difference sequence and the q-th phase difference sequence; A set of sub-parameters representing the difference between the P-th terminal and the Q-th terminal; Indicates taking the minimum value; represents the linear normalization function.

[0037] In the calculation formula of the difference parameter, the difference characteristics between the maximum values ​​of the phase difference sequence are expressed by the absolute value of the difference. The larger the absolute value of the difference, the more obvious the difference characteristics and the greater the difference. The difference sub-parameters are obtained by combining the difference characteristics of the maximum value and the DTW distance by multiplication, and a set of difference sub-parameters is formed. Finally, the smallest difference sub-parameter in the set of difference sub-parameters is selected to represent the minimum difference between different terminals, reflecting the correlation between the terminals, quantifying the differences between the terminals, and providing a basis for rapid terminal division.

[0038] It should be noted that the method of obtaining the DTW distance is already an existing technology, and the unit of the phase difference is degree; in other embodiments of the present invention, the implementer can also obtain the difference parameter by fusing the difference features of the maximum value and the DTW distance through positive correlation methods such as addition or weighted summation, or set the unit of the phase difference to radians, which will not be repeated here.

[0039] Preferably, in one embodiment of the present invention, considering that when the difference parameter between the edge terminal and the corresponding adjacent terminal is too large, it reflects that the connection between the edge terminal and the adjacent terminal is too weak and is not affected by the work of other terminals, so when the difference parameter between the edge terminal and the corresponding adjacent terminal is greater than a preset difference threshold, the corresponding edge terminal is deleted, which helps to simplify the network and avoid invalid calculations; When the difference parameter is small, it means that the characteristics of the edge terminal are relatively unique but still significantly correlated with the adjacent terminals. Classifying it as an exclusive community can retain its characteristics and avoid misclassification or information loss. Therefore, when the difference parameter between the edge terminal and the corresponding adjacent terminal is less than or equal to the preset difference threshold, the corresponding edge terminal is classified into an exclusive community.

[0040] It should be noted that, in one embodiment of the present invention, the preset difference threshold is 0.7, and the edge terminals that are not deleted do not participate in the iterative division and do not participate in the power regulation.

[0041] Step S302: In each iteration, in any control segment in the target control cycle, the terminals are clustered according to the fluctuation similarity characteristics of the phase difference between the terminals in each community to obtain terminal clustering clusters; the terminal clustering clusters of any two communities are nearest neighbor matched, and the corresponding two communities are merged according to the similar characteristics of the matched terminal clustering clusters to obtain the network structure of each iteration, and then the Louvain algorithm is used for iteration to finally obtain multiple relationship communities.

[0042] Considering that the terminal relationship graph is constructed based on the relationship between adjacent regulation segments of the target regulation cycle, the terminal relationship graph cannot reflect the relationship between terminals in the same regulation segment, which leads to the possibility that only the first regulation segment is divided into communities for the second regulation segment when dividing communities based on relationships, that is, terminals that appear in pairs are divided, while pairs of terminals with the same pair relationship have no relationship, resulting in no division; Therefore, it is also necessary to merge communities and adjust the network structure based on the phase difference fluctuations between terminals in the same control segment. That is, in each iteration, in any control segment in the target control cycle, the terminals are clustered according to the similar fluctuation characteristics of the phase difference between the terminals in each community to obtain terminal clustering clusters; the terminal clustering clusters of any two communities are matched with the nearest neighbors, and the corresponding two communities are merged according to the similar characteristics of the matched terminal clustering clusters to obtain the network structure of each iteration, and then the Louvain algorithm is used to merge the terminal pairs showing the same relationship into the same community, and finally multiple relationship communities are obtained to facilitate subsequent power regulation.

[0043] For example, a change in A leads to a change in C, and a change in B leads to a change in D. However, the terminal relationship diagram is constructed based on the relationship between AC and BD and cannot reflect the relationship between AB and CD. The community cannot be completely divided based on the terminal relationship diagram alone. Therefore, clustering is used to make terminals similar to those in the same regulatory segment in the same community a terminal cluster. The terminal relationship diagram is continuously adjusted to obtain the network structure of each iteration.

[0044] Preferably, in one embodiment of the present invention, the first regulation segment in the target regulation cycle is selected for analysis; considering that the variance and mean of the phase difference represent the severity of the overall fluctuation of the phase difference and the overall concentration characteristics, in any community, the variance and mean of the phase difference of the terminal in the first regulation segment are used to construct a two-dimensional space; density clustering is performed in the two-dimensional space to obtain terminal clustering clusters.

[0045] It should be noted that, in one embodiment of the present invention, the algorithm used for density clustering is the DBSCAN algorithm. In other embodiments of the present invention, the implementer may also select the second regulation segment in the target regulation cycle for analysis, and select other density clustering algorithms such as the OPTICS algorithm, etc., which are all technical means well known to those skilled in the art and will not be described in detail.

[0046] Preferably, in one embodiment of the present invention, any two communities are selected as the first community and the second community respectively; a matching pair of terminal clusters of the first community and the second community is obtained; wherein, when the first community and the second community are selected, communities of edge terminals are not included; when terminal clusters are matched, the same terminal cluster may correspond to multiple terminal clusters; by obtaining the coordinates of the cluster centers of the terminal clusters of the first community and the second community in two-dimensional space, the Euclidean distance between the cluster centers of different terminal clusters is calculated, and the nearest neighbor matching is performed with the minimum Euclidean distance; Considering that the difference parameters between terminals represent the difference characteristics between terminals, the similarity characteristics of the matching pairs of terminal clusters are analyzed by the difference parameters; if the similarity parameters of all matching pairs of the first community and the second community are larger, it means that the similarity relationship between the terminals of the first community and the second community is more obvious, and the terminals of the first community and the second community need to be merged more. Based on this, in the matching pair, the similarity parameters of the matching pair are obtained according to the difference parameters between the terminals of the two terminal clustering clusters; The average of all similar parameters corresponding to the first community and the second community is used as a merging parameter of the first community and the second community; when the merging parameter is greater than a preset merging threshold, the corresponding two communities are merged; All communities are merged to obtain the network structure of each iteration.

[0047] As an example, the calculation formula for similarity parameters includes: ; in, The serial number of the matching pair of the terminal clusters representing the first community and the second community; Indicates the first and second communities The similarity parameter of the matching pair; u represents the The serial number of the terminal in the first community of the matching pair; Indicates The number of terminals in the first community of the matching pairs; Indicates The serial number of the terminal in the second community of the matching pair; Indicates The number of terminals in the second community of the matching pairs of terminals clustered; It represents the exponential function with the natural constant e as the base; Indicates The difference parameter corresponding to the u-th terminal in the terminal cluster of the first community and the v-th terminal in the terminal cluster of the second community for a matching pair.

[0048] In the calculation formula of similar parameters, the negative correlation mapping function The difference parameters are negatively correlated and mapped, x represents the independent variable, thereby indicating the similarity characteristics between the terminals; the overall similarity characteristics of all terminals between the two terminal clusters are expressed by the mean as the similarity parameter of the matching pair.

[0049] As an example, the preset merging threshold is 0.5. In other embodiments of the present invention, implementers can adjust it by themselves.

[0050] After obtaining the network structure of each iteration, the Louvain algorithm is used to iterate the network structure and update the community, and finally several relationship communities are obtained to provide a basis for the final power regulation.

[0051] Step S4: Adjust the power redundancy according to the association direction between all terminals in each relationship community and the dramatic fluctuation characteristics of the phase difference in the latest regulation segment.

[0052] After obtaining multiple relationship communities, the relationship interference between devices that do not belong to the same relationship community is eliminated to improve the accuracy of power regulation. There are multiple terminals in each relationship community, and the terminals contain load transfer relationships, which are specifically manifested in the association direction between the terminals. The dramatic fluctuation characteristics of the phase difference of some terminals in the latest regulation segment can be used, combined with the transfer characteristics of the load, to adaptively adjust the power redundancy of another part of the terminals to ensure the safety of enterprise production.

[0053] Preferably, in one embodiment of the present invention, in any relationship community, the starting terminal of the association direction is recorded as the cause terminal, and the end point is recorded as the result terminal, and the cause-effect terminal pair is first determined; Considering that the larger the variance of the current phase difference of the cause terminal in a certain relationship community and the more drastic the fluctuation, the phase difference of the result terminal in the same community will also change accordingly in the future, and a higher power redundancy needs to be set to ensure production safety. Therefore, the variance of the phase difference of each cause terminal in the latest regulation segment is normalized as the power redundancy coefficient of the corresponding result terminal; the power redundancy is adjusted based on the power redundancy coefficient.

[0054] As an example, the calculation formula of the power redundancy factor includes: ; Wherein, T represents the power redundancy coefficient of the result terminal; It represents the variance of the phase difference of the cause terminal corresponding to the result terminal device in the latest control segment. The variance is normalized by the sigmoid function. Because the variance is a non-negative number, it takes a value in [0.5,1] after being mapped by the sigmoid function. It is also necessary to further adjust the value range, that is, subtract 0.5 from the mapped value and then multiply it by 2, so that the variance of the phase difference is normalized.

[0055] In one embodiment of the present invention, the rated power of the result terminal is set as the basic power consumption, and the product of the basic power consumption and the power redundancy coefficient is used as the adjusted power redundancy. The rated power can be obtained from the equipment nameplate or manual, and will not be described in detail here.

[0056] It should be noted that, in one embodiment of the present invention, the relationship community and the power redundancy are updated once after each control segment ends; in other embodiments of the present invention, considering that the relationship between equipment during enterprise production is usually relatively stable, that is, the relationship community is relatively stable, the implementer can set a longer update interval for the relationship community, such as updating the relationship community once every 100 control segments, to save computing resources.

[0057] An embodiment of the present invention also provides an intelligent electricity consumption monitoring system for enterprises, the system includes a memory, a processor and a computer program, wherein the memory is used to store the corresponding computer program, the processor is used to run the corresponding computer program, and when the computer program runs in the processor, it can implement an intelligent electricity consumption monitoring method for enterprises described in steps S1-S4.

[0058] In summary, in response to the technical problem of unreasonable setting of power redundancy for enterprise equipment terminals, the present invention proposes an intelligent power consumption monitoring method and system for enterprises. The present invention first obtains the phase difference in the power consumption data of each terminal of the enterprise; obtains the relationship topology network of all terminals; further, in the target control cycle, according to the similar fluctuation characteristics of the phase difference between any two terminals in adjacent control segments, combined with the relationship topology network, obtains the terminal relationship graph; further based on the Louvain algorithm, the terminals are divided to obtain multiple relationship communities; finally, according to the association direction between all terminals in each relationship community, and the drastic fluctuation characteristics of the phase difference in the latest control segment, the power redundancy is adjusted. The present invention analyzes the phase difference of the terminal to obtain the power transfer relationship between the terminals, and divides the relationship community with the help of the Louvain algorithm, while reducing the interference between devices in different relationship communities, accurately regulating the power redundancy, and providing protection for enterprise safety production.

[0059] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0060] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. An intelligent monitoring method for electricity consumption for enterprises, characterized in that: The method comprises: Obtain the phase difference in the power consumption data of each terminal of the enterprise; obtain the control segment with a preset time domain length, and use the latest two adjacent control segments as a target control cycle; obtain the relationship topology network of all the terminals, and set the weights between all the terminals under the same diversion device to the preset weights; In the target control cycle, according to the fluctuation similarity characteristics of the phase difference between any two terminals in adjacent control segments, the relationship strength and the association direction of the corresponding two terminals are obtained; based on the relationship topology network, the terminal relationship graph is obtained in combination with the relationship strength and the association direction; Dividing the terminals based on the Louvain algorithm to obtain multiple relationship communities; Adjusting the power redundancy according to the association direction between all the terminals in each of the relationship communities and the dramatic fluctuation characteristics of the phase difference in the latest regulation segment; The division process includes: in the target control cycle, according to the fluctuation difference characteristics of the phase difference between any two terminals, obtaining the difference parameters between the corresponding two terminals; before the first iteration of the Louvain algorithm, dividing the edge terminals according to the difference parameters between the edge terminals and the corresponding adjacent terminals in the terminal relationship graph; In each iteration, in any of the regulation segments in the target regulation cycle, the terminals are clustered according to the fluctuation similarity characteristics of the phase difference between the terminals in each community to obtain terminal clustering clusters; the terminal clustering clusters of any two communities are nearest neighbor matched, and the corresponding two communities are merged according to the similarity characteristics of the matched terminal clustering clusters to obtain the network structure of each iteration, and the Louvain algorithm is used for iteration again to finally obtain multiple relationship communities.

2. According to claim 1, a method for intelligent monitoring of electricity consumption for enterprises is characterized in that: According to the fluctuation similarity characteristics of the phase difference between any two terminals in adjacent control segments, the method for obtaining the relationship strength and the correlation direction corresponding to the two terminals includes: Select any two of the terminals as the first terminal and the second terminal respectively, and use the two control segments in the target control cycle as the first segment and the second segment respectively in time sequence; Obtaining a correlation coefficient between the phase difference of the first segment of the first terminal and the phase difference of the second segment of the second terminal as a first correlation coefficient; obtaining a correlation coefficient between the phase difference of the second segment of the first terminal and the phase difference of the first segment of the second terminal as a second correlation coefficient; The maximum value of the first correlation coefficient and the second correlation coefficient is selected as the relationship strength between the corresponding two terminals; the association direction is: from the terminal of the first segment in the relationship strength to the terminal of the second segment.

3. According to claim 2, a method for intelligent monitoring of electricity consumption for enterprises is characterized in that: The method for obtaining the terminal relationship diagram includes: When the relationship strength is lower than a preset strength threshold, the relationship strength is reset to zero; and the sum of the relationship strength and the preset weight is used as the relationship weight corresponding to the two terminals; Based on the relationship topology network, in combination with the relationship weight and the association direction, a terminal relationship graph is obtained.

4. The method for intelligent monitoring of electricity consumption for enterprises according to claim 1 is characterized in that: The method for obtaining the difference parameter includes: In the target control cycle, the phase difference is divided into a phase difference sequence based on the maximum value of each phase difference; each maximum value in the phase difference corresponds to the phase difference sequence with the maximum value itself as the starting point; The phase difference sequences in any two of the terminals are randomly paired to form a binary group; according to the difference characteristics of the two maxima corresponding to each binary group and the DTW distance of the two phase difference sequences corresponding to each binary group, a difference sub-parameter of each binary group is obtained; the difference characteristics of the maxima and the DTW distance are both positively correlated with the difference sub-parameter; The minimum value of the difference sub-parameters corresponding to any two of the terminals is used as the difference parameter between the corresponding two terminals.

5. The method for intelligent monitoring of electricity consumption for enterprises according to claim 4 is characterized in that: The method for dividing the edge terminals according to the difference parameters between the edge terminals and the corresponding adjacent terminals in the terminal relationship graph comprises: When the difference parameter between the edge terminal and the corresponding adjacent terminal is greater than a preset difference threshold, the corresponding edge terminal is deleted; when the difference parameter between the edge terminal and the corresponding adjacent terminal is less than or equal to the preset difference threshold, the corresponding edge terminal is divided into an exclusive community.

6. The method for intelligent monitoring of electricity consumption for enterprises according to claim 1, characterized in that: The method for obtaining the terminal cluster includes: In any community, the variance and mean of the phase difference of the terminal in the first regulation segment are used to construct a two-dimensional space; density clustering is performed in the two-dimensional space to obtain terminal clusters.

7. The method for intelligent monitoring of electricity consumption for enterprises according to claim 1, characterized in that: The method for obtaining the network structure of each iteration includes: Select any two communities as a first community and a second community respectively; obtain a matching pair of the terminal clusters of the first community and the second community; In the matching pair, obtaining similarity parameters of the matching pair according to the difference parameters between the terminals of the two terminal clusters; taking the average of all the similar parameters corresponding to the first community and the second community as a merging parameter of the first community and the second community; when the merging parameter is greater than a preset merging threshold, merging the corresponding two communities; All the communities are merged to obtain the network structure of each iteration.

8. The method for intelligent monitoring of electricity consumption for enterprises according to claim 2 is characterized in that: The method for adjusting the electric energy redundancy comprises: In any of the relationship communities, the starting terminal of the association direction is recorded as the cause terminal, and the end point is recorded as the result terminal; the variance of the phase difference of each cause terminal in the latest regulation segment is normalized and used as the power redundancy coefficient of the corresponding result terminal; and the power redundancy amount is adjusted based on the power redundancy coefficient.

9. The method for intelligent monitoring of electricity consumption for enterprises according to claim 8, characterized in that: The method for adjusting the power redundancy amount based on the power redundancy coefficient includes: The rated power of the result terminal is set as the basic power consumption, and the product of the basic power consumption and the power redundancy coefficient is used as the adjusted power redundancy.

10. An intelligent power consumption monitoring system for enterprises, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the enterprise-oriented intelligent electricity consumption monitoring method as described in any one of claims 1 to 9 are implemented.

Images