CN-121980204-A - Mine ventilator group collaborative energy-saving operation optimizing system based on data analysis

CN121980204ACN 121980204 ACN121980204 ACN 121980204ACN-121980204-A

Abstract

The invention discloses a mine ventilator group collaborative energy-saving operation optimization system based on data analysis, which comprises a data acquisition module, a correlation acquisition module and a ventilation path dynamic correlation matrix, wherein the data acquisition module acquires multi-source time sequence data of a ventilation system, builds a network topology undirected graph through space-time correlation coding preprocessing based on network topology, acquires dynamic data blocks with topology labels based on the network topology undirected graph, the correlation acquisition module extracts air volume time sequence data based on the dynamic data blocks and performs first-order difference operation to acquire air volume dynamic change rate, introduces physical connectivity constraint of ventilation network topology based on the air volume dynamic change rate to acquire a wind path dynamic correlation matrix, and quantifies the conduction intensity of air volume dynamic change of a wind path and a core correlation path between fans, so that the problem that static topology cannot reflect system dynamic response in the prior art is solved.

Inventors

LI HONGYE
CHEN RUI
Zhou Zhensong
ZHAO XIAOXI
Xi Mengbo
LI HAO
XIE FANGMING
LI YUDIAN

Assignees

诺文科风机(北京)有限公司

Dates

Publication Date: 20260505
Application Date: 20260408

Claims (8)

1. Mine ventilator group cooperation energy-saving operation optimizing system based on data analysis, which is characterized by comprising: The data calling module is used for collecting multi-source time sequence data of the ventilation system, constructing a network topology undirected graph through space-time associated coding preprocessing based on network topology, and obtaining dynamic data blocks with topology labels based on the network topology undirected graph; The association acquisition module is used for extracting air volume time sequence data based on the dynamic data block, performing first-order differential operation to obtain an air volume dynamic change rate, and introducing physical connectivity constraint of ventilation network topology based on the air volume dynamic change rate to obtain an air path dynamic association degree matrix; the intelligent analysis module is used for carrying out multi-to-multi topology association mapping of fans and the wind paths based on the wind path dynamic association matrix to obtain core dynamic association, establishing a fan association mapping function of the fans and the wind paths by the ventilation network topology undirected graph to obtain a fan group topology adjacency matrix, and constructing a fan group dynamic association matrix based on the combination of the core dynamic association and the fan group topology adjacency matrix; And the intelligent optimization module is used for carrying out efficiency optimization based on the fan group dynamic correlation matrix to obtain an optimal fan regulation sequence so as to realize energy-saving operation optimization.
2. The mine ventilator group collaborative energy-saving operation optimization system based on data analysis according to claim 1, wherein the process of constructing a network topology undirected graph is: multi-source time sequence original data set collected by sensors of monitoring points of mine ventilation system To multisource time sequence original data set Mapping in association with topological spatial features, assigning unique topological labels to each group of time sequence data, and abstracting network topology of the mine ventilation system into an undirected graph Wherein Is a roadway junction and fan installation node set, And (5) collecting all ventilation air paths.
3. The mine ventilator group collaborative energy efficient operation optimization system based on data analysis according to claim 2 wherein the process of obtaining topology tagged dynamic data blocks is: Identifying multisource temporal raw data sets The air path or the fan is Obtaining a topologically tagged dataset from spatial locations in a database ; Setting a sliding window length For a pair of Dividing time windows, integrating time sequence data with topological labels in each window into dynamic data blocks, and finally generating dynamic data blocks with topological labels 。
4. The data analysis-based mine ventilator group collaborative energy-saving operation optimization system according to claim 3, wherein the process of obtaining the dynamic change rate of the air quantity by performing a first-order difference operation is as follows: based on dynamic data blocks Let the total number of air paths in the ventilation system be Extracting the first time window Data blocks corresponding to the air paths are used as air quantity time sequence data ; For time sequence data of air quantity Zero-mean normalization is carried out to obtain normalized air volume time sequence data For the normalized air volume time sequence data Performing first-order differential operation to obtain a dynamic change rate sequence of the air quantity corresponding to the air path i and the air path j Sequence of dynamic change rate of air quantity 。
5. The data analysis-based mine ventilator group collaborative energy-saving operation optimization system according to claim 4, wherein the process of obtaining the wind path dynamic association matrix by introducing physical connectivity constraints of a ventilation network topology is as follows: Network topology undirected graph Judging wind path And air path The direct physical connection relation is used for the dynamic change rate sequence of the air quantity corresponding to the air path i and the air path j with direct physical connection And (3) with Based on the sequence of dynamic change rate of air quantity And (3) with Dynamic association degree of air path i and air path j is obtained by adopting cosine similarity ; When (when) When the air flow changes, the air flow changes in the same direction, and when When the air flow changes, the change trend of the air flow of the two air paths is reversed, and Time of day ; Finally between all air paths Construction of elements Wind path dynamic association degree matrix of order 。
6. The data analysis-based mine ventilator group collaborative energy-saving operation optimization system according to claim 5, wherein the process of obtaining the core dynamic association degree by performing the many-to-many topology association mapping of fans and air paths is as follows: define the global scope of the fan group, let the total number of mine ventilation machine groups be The serial numbers are in turn Wherein And (2) and ; According to the topology priority of the blower installation location, for a pair of The table fan is uniquely numbered and based on mine ventilation network topology undirected graph Establishing a fan association mapping function of a fan and an air path ; Wherein, the Is a set of fans of a fan group, For air-path collection in ventilation network A power set of (2); Dynamic relevance matrix based on wind path Any two fans are calculated Dynamic degree of association between : Extraction fan And fan Intersection wind path pairs corresponding to the wind path subsets and judging dynamic association degree of the intersection wind path pairs with the intersection as an empty set ; Dynamic association degree matrix of air path pairs with non-empty intersections In the method, the dynamic association degree of all the intersection wind paths to the corresponding wind paths is extracted And selecting the maximum value as the fan And fan Core dynamic association degree of (2) 。
7. The data analysis-based mine ventilator group collaborative energy saving operation optimization system according to claim 6, wherein the process of constructing a ventilator group topology adjacency matrix is: By groups of fans Platform fan is node, based on ventilation network topology undirected graph Mapping function associated with fan Definition of Order type 0-1 topological base adjacency matrix The matrix elements are noted as ; Wherein, if fan And fan With wind path conduction paths, i.e. Then ; If fan And fan Without any wind path conduction path ; When (when) When in use, the unified setting is performed 。
8. The mine ventilator group collaborative energy saving operation optimization system based on data analysis according to claim 7, wherein the process of constructing a ventilator group dynamic association matrix is: Adjacency matrix based on topology Based on the dynamic association degree of the cores of each pair of fans Embedding the fan group dynamic association matrix into a matrix frame in an element-by-element multiplication mode to obtain a fan group dynamic association matrix, wherein the standardized mathematical expression is as follows: ; Wherein, the Is the matrix Hadamard product operation symbol, To take the core dynamic association degree of all fan pairs as element The original correlation matrix of the order is obtained, For the number of groups of fans, And a dynamic association matrix for the fan group.

Description

Mine ventilator group collaborative energy-saving operation optimizing system based on data analysis Technical Field The invention relates to the technical field of new generation information, in particular to a mine ventilator group collaborative energy-saving operation optimizing system based on data analysis. Background Energy-saving optimization of mine ventilation systems has undergone a development phase from single-machine independent regulation to multi-machine coordinated regulation. Early technologies focused on the variable frequency speed regulation transformation of a single fan, and reduced single energy consumption by optimizing the individual operation curve of the fan, but did not consider the coupling relationship between the air path and the fan in the ventilation system. As mine exploitation depth increases and ventilation network complexity increases, a multi-fan cooperative regulation technology is gradually developed, and a fan regulation partition is divided by analyzing a static topological relation of an air path, so as to attempt to realize the air volume cooperative matching of a local air path. In recent years, along with the application of the internet of things and data analysis technology, a part of schemes introduce multi-source time sequence data monitoring, but still stay in the passive feedback of the running state of a fan, the quantitative mapping of the dynamic association of a wind path and the dynamic response of a fan group is not established, and the precise energy-saving regulation and control of a global ventilation system are difficult to support. The existing mine ventilator group energy-saving technology has three core defects: the wind path association depiction only depends on static topology, and the conduction intensity of the dynamic change of the wind quantity is not quantized, so that the wind quantity fluctuation of the associated wind path cannot be prejudged when a fan is regulated and controlled, the local ventilation unbalance is easy to be caused, the ventilation safety is influenced, and the energy-saving effect is weakened. The regulation and control priorities of the fan groups are divided only according to experience, and dynamic relevancy ordering of the key passages is not combined, so that the regulation and control of the core fans are delayed, the secondary fans are blindly regulated, and the energy consumption redundancy of the system is caused. The efficiency optimizing process is used for processing a single fan in an isolated manner, and the linkage influence of pre-judging and controlling based on a fan group dynamic association model is not adopted, so that the fan operation parameter optimization only meets the single efficiency optimization, the global optimization of the whole energy consumption of the fan group cannot be realized, and the dynamic operation requirement of a complex ventilation system is difficult to adapt. Therefore, a mine ventilator group collaborative energy-saving operation optimization system based on data analysis is provided to solve the problems. Disclosure of Invention In order to overcome the defects in the prior art, the invention provides the following technical scheme for achieving the purposes: mine ventilator group collaborative energy-saving operation optimizing system based on data analysis, comprising: The data calling module is used for collecting multi-source time sequence data of the ventilation system, constructing a network topology undirected graph through space-time associated coding preprocessing based on network topology, and obtaining dynamic data blocks with topology labels based on the network topology undirected graph; The association acquisition module is used for extracting air volume time sequence data based on the dynamic data block, performing first-order differential operation to obtain an air volume dynamic change rate, and introducing physical connectivity constraint of ventilation network topology based on the air volume dynamic change rate to obtain an air path dynamic association degree matrix; the intelligent analysis module is used for carrying out multi-to-multi topology association mapping of fans and the wind paths based on the wind path dynamic association matrix to obtain core dynamic association, establishing a fan association mapping function of the fans and the wind paths by the ventilation network topology undirected graph to obtain a fan group topology adjacency matrix, and constructing a fan group dynamic association matrix based on the combination of the core dynamic association and the fan group topology adjacency matrix; And the intelligent optimization module is used for carrying out efficiency optimization based on the fan group dynamic correlation matrix to obtain an optimal fan regulation sequence so as to realize energy-saving operation optimization. The process for constructing the network topology undirected graph comprises the following steps: mult