CN115482125B

CN115482125B - Water conservancy panoramic information sensing method and device

Info

Publication number: CN115482125B
Application number: CN202211296994.7A
Authority: CN
Inventors: 智勇鸣; 吴国颖; 赵玉忠; 陈红生; 高琳; 何鑑坤; 夏永丽; 曾秀英; 郑涌; 陈骞; 梁剧文; 黄文龙; 张海丽; 陈志运; 李淑英
Original assignee: China Water Resources Pearl River Planning Surverying & Designing Co ltd
Current assignee: China Water Resources Pearl River Planning Surverying & Designing Co ltd
Priority date: 2022-10-21
Filing date: 2022-10-21
Publication date: 2023-09-08
Anticipated expiration: 2042-10-21
Also published as: CN115482125A

Abstract

The invention provides a water conservancy panoramic information sensing method and a water conservancy panoramic information sensing device, which are characterized in that a TSMP algorithm is adopted to respectively transmit original water conservancy data acquired by different sensors at each target moment in a preset time period to target terminal equipment of a target water conservancy system, a neighbor propagation clustering algorithm is adopted to respectively clean the original water conservancy data corresponding to each target moment received by the target terminal equipment, a Huffman compression algorithm is adopted to compress all data points corresponding to each target moment obtained after data cleaning, and a pre-established distributed data fusion algorithm is adopted to fuse the data compression data corresponding to each target moment obtained after data compression, so that the sensing data of the target water conservancy system in the preset time period is obtained. The invention can improve the efficiency and accuracy of the water conservancy system information perception, thereby improving the refinement, timeliness and intelligentization level of the water conservancy information management work.

Description

Water conservancy panoramic information sensing method and device

Technical Field

The invention relates to the technical field of data processing, in particular to a water conservancy panoramic information sensing method and device.

Background

The information perception provides information sources for the application of the Internet of things and is the basis of the application of the Internet of things. The most basic form of information perception is data collection, namely, the perception data acquired by each partial node is transmitted to the sink node through a network. With the continuous development of information sensing technology, the information sensing technology can be applied to a water conservancy system to manage water conservancy information. The purpose of water conservancy system information perception is to acquire the water conservancy information that the user is interested, does not need to collect all perception data of water conservancy system under most circumstances, and will cause the network load too big with all data transmission to the sink node of water conservancy system, is difficult to guarantee the timeliness of water conservancy information management work. The efficiency and accuracy of the current water conservancy system information sensing scheme are low, and the application requirements of water conservancy information management are difficult to meet.

Disclosure of Invention

Therefore, the invention aims to provide a water conservancy panoramic information sensing method and device so as to improve the efficiency and accuracy of water conservancy system information sensing, thereby improving the refinement, timeliness and intelligentization level of water conservancy information management work.

In a first aspect, an embodiment of the present invention provides a water conservancy panorama information sensing method, where the method includes: acquiring original water conservancy data of a target area in real time through a plurality of sensors of a target water conservancy system; each original water conservancy data has respective data acquisition time and respective data size; the method comprises the steps of respectively transmitting original water conservancy data acquired at each target moment in a preset period of time by adopting a TSMP algorithm to target terminal equipment of a target water conservancy system; adopting a neighbor propagation clustering algorithm to respectively carry out data cleaning on the original water conservancy data corresponding to each target moment received by the target terminal equipment, and obtaining a plurality of classes corresponding to each target moment; wherein each class comprises a plurality of data points, the clustering center of each class is a data point, and the clustering centers of different classes are different; carrying out data compression on all data points corresponding to each target moment by adopting a Huffman compression algorithm to obtain water conservancy compression data corresponding to each target moment; carrying out data fusion on the water conservancy compression data corresponding to each target moment by adopting a pre-established distributed data fusion algorithm to obtain water conservancy fusion data corresponding to each target moment; the distributed data fusion algorithm is established based on the data transmission relation among the sensors.

In a second aspect, an embodiment of the present invention further provides a water conservancy panorama information sensing apparatus, where the apparatus includes: the data acquisition module is used for acquiring original water conservancy data of a target area in real time through a plurality of sensors of the target water conservancy system; each original water conservancy data has respective data acquisition time and respective data size; the data transmission module is used for respectively transmitting the original water conservancy data acquired at each target moment in a preset period of time by using a TSMP algorithm to target terminal equipment of the target water conservancy system; the data cleaning module is used for respectively carrying out data cleaning on the original water conservancy data corresponding to each target moment received by the target terminal equipment by adopting a neighbor propagation clustering algorithm to obtain a plurality of classes corresponding to each target moment; wherein each class comprises a plurality of data points, the clustering center of each class is a data point, and the clustering centers of different classes are different; the data compression module is used for carrying out data compression on all data points corresponding to each target moment by adopting a Huffman compression algorithm to obtain water conservancy compression data corresponding to each target moment; the data fusion module is used for carrying out data fusion on the water conservancy compression data corresponding to each target moment by adopting a pre-established distributed data fusion algorithm to obtain water conservancy fusion data corresponding to each target moment; the distributed data fusion algorithm is established based on the data transmission relation among the sensors.

According to the water conservancy panoramic information sensing method and device provided by the embodiment of the invention, the original water conservancy data of the target area are acquired in real time through the plurality of sensors of the target water conservancy system, the original water conservancy data acquired by the different sensors at each target moment in a preset period are respectively transmitted to the target terminal equipment of the target water conservancy system by adopting a TSMP algorithm, and the original water conservancy data corresponding to each target moment received by the target terminal equipment is respectively subjected to data cleaning by adopting a neighbor propagation clustering algorithm, so that a plurality of classes corresponding to each target moment are obtained, and each class comprises a plurality of data points; carrying out data compression on all data points corresponding to each target moment by adopting a Huffman compression algorithm, thereby obtaining water conservancy compression data corresponding to each target moment; and carrying out data fusion on the water conservancy compression data corresponding to each target moment by adopting a pre-established distributed data fusion algorithm, so as to obtain the water conservancy fusion data corresponding to each target moment. By adopting the technology, the efficiency and accuracy of water conservancy system information perception can be improved, so that the refinement, timeliness and intelligent level of water conservancy information management work are improved.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a water conservancy panorama information sensing method according to an embodiment of the present invention;

FIG. 2 is a flowchart illustrating a TSMP algorithm according to an embodiment of the present invention;

FIG. 3 is a diagram showing an exemplary structure of a REQUEST in an embodiment of the present invention;

Fig. 4 is a diagram illustrating an exemplary structure of a huffman binary tree according to an embodiment of the present invention;

fig. 5 is a diagram illustrating a structure of a sensor network according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram of a water conservancy panorama information sensing device according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described in conjunction with the embodiments, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

With the continuous development of information sensing technology, the information sensing technology can be applied to a water conservancy system to manage water conservancy information. However, the efficiency and accuracy of the current water conservancy system information sensing scheme are low, and the application requirements of water conservancy information management are difficult to meet. Based on the method and the device for sensing the water conservancy panoramic information, provided by the embodiment of the invention, the efficiency and the accuracy of sensing the water conservancy system information can be improved, so that the refinement, timeliness and intelligent level of water conservancy information management work are improved.

For the convenience of understanding the present embodiment, first, a detailed description will be given of a water conservancy panorama information sensing method disclosed in the present embodiment, referring to a schematic flow chart of a water conservancy panorama information sensing method shown in fig. 1, the method may include the following steps:

step S102, acquiring original water conservancy data of a target area in real time through a plurality of sensors of a target water conservancy system; each original water conservancy data has a respective data acquisition time and a respective data size.

The original water conservancy data can adopt water level data, rainfall data, flow velocity data, water quality data and the like, can be customized according to actual water conservancy demands, and is not limited. Accordingly, the above-mentioned sensor can specifically be a liquid level sensor, a rainfall sensor, a flow rate sensor, etc., so that the number and types of the above-mentioned sensors can be customized according to the actual water conservancy requirement, and this is not limited.

And step S104, transmitting the original water conservancy data acquired at each target moment in a preset period by adopting a TSMP algorithm to target terminal equipment of a target water conservancy system.

Step S106, adopting a neighbor propagation clustering algorithm to respectively carry out data cleaning on the original water conservancy data corresponding to each target moment received by the target terminal equipment, and obtaining a plurality of classes corresponding to each target moment; wherein each class contains a plurality of data points, the cluster center of each class is a data point, and the cluster centers of different classes are different.

The data points may be monobasic, dibasic, tribasic or higher, and specifically may be customized according to actual requirements, which is not limited. Each data point has a respective data acquisition time and a respective data size.

And S108, carrying out data compression on all data points corresponding to each target moment by adopting a Huffman compression algorithm to obtain water conservancy compression data corresponding to each target moment.

Step S110, carrying out data fusion on the water conservancy compression data corresponding to each target moment by adopting a pre-established distributed data fusion algorithm to obtain water conservancy fusion data corresponding to each target moment; the distributed data fusion algorithm is established based on the data transmission relation among the sensors.

According to the water conservancy panoramic information sensing method provided by the embodiment of the invention, the original water conservancy data of the target area are acquired in real time through a plurality of sensors of the target water conservancy system, the original water conservancy data acquired by different sensors at each target moment in a preset period are respectively transmitted to target terminal equipment of the target water conservancy system by adopting a TSMP algorithm, and the original water conservancy data corresponding to each target moment received by the target terminal equipment is respectively subjected to data cleaning by adopting a neighbor propagation clustering algorithm, so that a plurality of classes corresponding to each target moment are obtained, and each class comprises a plurality of data points; carrying out data compression on all data points corresponding to each target moment by adopting a Huffman compression algorithm, thereby obtaining water conservancy compression data corresponding to each target moment; and carrying out data fusion on the water conservancy compression data corresponding to each target moment by adopting a pre-established distributed data fusion algorithm, so as to obtain the water conservancy fusion data corresponding to each target moment. By adopting the technology, the efficiency and accuracy of water conservancy system information perception can be improved, so that the refinement, timeliness and intelligent level of water conservancy information management work are improved.

As a possible implementation manner, the step S104 (that is, the TSMP algorithm is used to transmit the raw water conservancy data collected by the different sensors at each target moment in the preset period to the target terminal device of the target water conservancy system) may include the following operation manners:

(11) For the original water conservancy data corresponding to each target moment, respectively generating respective request messages through each sensor; the request message comprises a request synchronous message, a request data message, a time stamp for starting data transmission and a data transmission time delay; the request synchronization message has a synchronization flag indicating whether the data to be transmitted is time synchronized, and the request data message has flag data indicating whether the data to be transmitted is valid.

(12) For each generated request message, setting a synchronization flag of the request message according to whether a sensor generating the request message has communicated with a target terminal device before generating the request message and whether data to be transmitted of the sensor has time synchronization, setting flag data of the request message according to whether the target terminal device receives the data to be transmitted of the sensor, and sending the request message with the synchronization flag and the flag data set to the target terminal device.

(13) And transmitting the original water conservancy data acquired by different sensors at each target moment in a preset period to target terminal equipment of a target water conservancy system according to each request message received by the target terminal equipment.

Illustratively, the step of setting the synchronization flag of the request message according to whether the sensor generating the request message has communicated with the target terminal device before generating the request message and whether time synchronization of data to be transmitted of the sensor occurs in (12) above may include the following operation modes:

(121) If the sensor generating the request message has communicated with the target terminal device before generating the request message and the data to be transmitted of the sensor is time-synchronized beyond a preset time period before the sensor generates the request message, or the sensor generating the request message has not communicated with the target terminal device before generating the request message, setting the synchronization flag of the request message to 1;

(122) And if the data to be transmitted of the sensor is not time-synchronized beyond a preset time period before the sensor generates the request message and the sensor is not communicated with the target terminal equipment after the time synchronization of the data to be transmitted, setting the synchronization flag of the request message to 0.

The step of setting the flag data of the request message according to whether the target terminal device receives the data to be transmitted of the sensor in (12) may include the following operation modes:

(123) If the target terminal equipment does not receive the data to be transmitted of the sensor, the flag data of the request message is set to 0;

(124) If the target terminal device receives the data to be transmitted of the sensor, the flag data of the request message is set to 1.

Illustratively, the step of (13) above may include the following manner of operation:

(131) For each request message received by the target terminal equipment, if the synchronization mark of the request message is 1 and the mark data of the request message is 0, carrying out time synchronization on the data to be transmitted corresponding to the request message, feeding back the response message of the request message to the sensor through the target terminal equipment, and not transmitting the data to be transmitted from the sensor for sending the request message to the target terminal equipment;

(132) For each request message received by the target terminal device, if the synchronization flag of the request message is 1 and the flag data of the request message is 1, time synchronization is performed on the target data corresponding to the request message, the response message of the request message is fed back to the sensor through the target terminal device, and then the data to be transmitted is transmitted from the sensor sending the request message to the target terminal device.

For ease of understanding, the operation of the above (11) to (13) will be exemplarily described herein with reference to fig. 2 and 3 as follows:

referring to fig. 2, the TSMP algorithm consists of three main parts, control frame (i.e., REQUEST, i.e., REQUEST message described above) analysis, time synchronization, and data frame transmission. The control frames are generated and transmitted by the sensor and then received and analyzed by the target terminal device. In the target water conservancy system, if the original water conservancy data acquired by a certain sensor at a certain target moment is required to be transmitted to target terminal equipment, a REQUEST can be sent to the target terminal equipment through the sensor, so that the subsequent time synchronization and data frame transmission can be carried out according to the analysis result of the REQUEST by the target terminal equipment.

Referring to fig. 3, the REQUEST may include two messages, a REQUEST synchronization message (i.e., "Sync" in fig. 3) having a synchronization flag (i.e., sync in fig. 2) indicating whether or not Data to be transmitted is time-synchronized, and a REQUEST Data message (i.e., "Data" in fig. 3) having flag Data (i.e., data in fig. 2) indicating whether or not Data to be transmitted is valid. The REQUEST may also contain a destination address and a source address to characterize the direction of data transfer. The REQUEST may also contain a check code to ensure the security of the data transmission. In addition, the REQUEST may further include a timestamp of the start of data transmission (i.e. "timestamp" in fig. 3) and a data transmission delay (i.e. "transmission delay" in fig. 3), so as to ensure that the target terminal may receive the corresponding original hydraulic data according to the start time of transmission and the time of data transmission after receiving the REQUEST.

Referring to fig. 2, for a certain sensor, when it is required to transmit the raw water conservancy Data collected by the sensor at a certain target moment to a target terminal device, a corresponding REQUEST (Sync and Data of the REQUEST are all 0 by default) may be generated by the sensor, and a Data transmission history of the sensor may be queried to determine whether the sensor communicates with the target terminal device: if the sensor has communicated with the target terminal device before the REQUEST was generated and the data to be transmitted is time synchronized long ago (e.g., more than three months before the time of generation of the REQUEST), or the sensor has not communicated with the target terminal device before the REQUEST was generated, then Sync of the REQUEST may be set to 1 to characterize the data to be transmitted as historical data for time synchronization; if the data to be transmitted is time synchronized shortly before (e.g., within three months before the time of generation of the REQUEST) and the sensor is not in communication with the target terminal device after the time synchronization of the data to be transmitted, then the Sync of the REQUEST may be set to 0 to characterize the data to be transmitted as new data that is not time synchronized. It is also necessary to determine whether the data to be transmitted is transmitted: if the target terminal equipment does not receive the Data to be transmitted, the Data of the REQUEST can be set to 0 so as to represent that the Data to be transmitted is invalid Data; if the target terminal device receives the Data to be transmitted, the Data of the REQUEST may be set to 1 to characterize the Data to be transmitted as valid Data. After the Sync and Data of the REQUEST are set, the REQUEST with the Sync and Data set is transmitted to the target terminal device. After receiving the REQUEST with the set Sync and Data, the target terminal device analyzes the REQUEST to determine the Sync and Data of the REQUEST, and performs subsequent time synchronization and Data frame transmission according to the analysis result: if the Sync of the REQUEST is 1 and the Data of the REQUEST is 0, the Data to be transmitted may be time-synchronized, and the REQUEST result (i.e., the response message) of the REQUEST message is fed back to the sensor, and then the flow is ended; if the Sync and Data of the REQUEST are both 1, time synchronization can be performed on the Data to be transmitted, and the REQUEST result (i.e., the response message) of the REQUEST message is fed back to the sensor, and then Data frame transmission (i.e., the Data to be transmitted is transmitted from the sensor to the target terminal device) is performed; if the Sync and the Data of the REQUEST are both 0, the Data to be transmitted can be determined as invalid Data, and then the flow is ended; if the Sync of the REQUEST is 0 and the Data of the REQUEST is 1, the REQUEST result (i.e., a response message) of the REQUEST message may be fed back to the sensor, followed by Data frame transmission (i.e., transmission of Data to be transmitted from the sensor to the target terminal device). Through the process, the reliable transmission of various water conservancy data which needs to be collected by the water conservancy system can be ensured, and the efficiency and accuracy of the water conservancy data transmission are improved.

As a possible operation manner, the above step S106 (that is, performing data cleaning on the raw water conservancy data corresponding to each target time received by the target terminal device by using a neighbor propagation clustering algorithm to obtain a plurality of classes corresponding to each target time) may include the following operation manners:

(21) And for the original water conservancy data corresponding to each target moment received by the target terminal equipment, taking each original water conservancy data corresponding to the target moment as a data point, respectively calculating the similarity between data points and the reference degree of each data point, and respectively calculating the attraction between data points, the attribution between data points and the self attraction of each data point based on the similarity between data points and the reference degree of each data point.

(22) For the original water conservancy data corresponding to each target moment, respectively carrying out multiple iterations on the attraction degree between each data point corresponding to the target moment and the attribution degree between each data point corresponding to the target moment based on a preset attenuation coefficient; the result of each iteration comprises a corresponding clustering center; stopping iteration until the clustering centers in the results of continuous multiple iterations are identical or reach the preset iteration times, taking each clustering center corresponding to the stopped iteration as a clustering center of one class, and aggregating all data points corresponding to the target moment into multiple classes.

The similarity may be defined by using a euclidean distance, a martensii distance, a minkowski distance, a hamming distance, a Tanimoto coefficient, a Jaccard coefficient, a pearson correlation coefficient, a cosine similarity, or by other defining methods, which may be specifically determined according to actual requirements, and is not limited. The similarity is for two data points, and the reference is for one data point, and for a certain data point, the reference of the data point may be the similarity between the data point and itself.

For a data point and an other data point other than the data point, the degree of attraction (i.e., the representative degree) between the data point and the other data point may be used to characterize the degree of class in which the other data point represents the data point, and the degree of attribution (i.e., the suitable degree) between the data point and the other data point may be used to characterize the suitable degree of class in which the data point selects the other data point as the representative point (i.e., the cluster center).

The basic idea of the above-mentioned neighbor propagation clustering algorithm is to take each obtained data point as a potential clustering center, then pair each pair of data points and calculate the similarity between each data point and the respective reference degree of each data point respectively, so that the similarity between each data point and the respective reference degree of each data point form a network (i.e. a similarity matrix), and then the clustering center is calculated by transmitting the similarity value corresponding to the element position with different line numbers and column numbers in the similarity matrix and the reference value corresponding to the element position with the same line number and column number.

In the clustering operation process of the neighbor propagation clustering algorithm, each found clustering center is a data point which is actually existing and used for representing each class, namely, one clustering center can be used for uniquely representing one class. The attribution and attraction are iteratively transferred between the data points (including the original water conservancy data and the time stamp of the original water conservancy data) until the iterative process converges, the clusters (i.e. the class obtained by aggregation) and the center of each cluster (i.e. the cluster center) are also fixed, the cluster center is the data point which exists actually, and each data point which is not the cluster center is respectively attributed to the corresponding cluster.

In the above (21), after the original water conservancy data corresponding to a certain target time is received by the target terminal device, for each original water conservancy data corresponding to the target time, a data point may be formed by the original water conservancy data and a timestamp of the original water conservancy data, a negative euclidean distance between the data point and each other data point is calculated as a similarity between the data point and each other data point, a negative euclidean distance between the data point and itself is calculated as a reference degree of the data point (i.e. a similarity between the data point and itself), and the similarity may be ensured to be symmetrical by adopting the negative euclidean distance as the similarity, so that a subsequent calculation process is facilitated by using the similarity, thereby improving the calculation efficiency.

In the above (21), for all the data points corresponding to a certain target time, the attraction degree between the data points, the attribution degree between the data points and the self-attraction degree of each data point may be calculated according to the following formula:

wherein s (i, k) is the similarity between the point i and the point k, and elements with different row numbers and column numbers are used for representing the degree that the point k is suitable as the clustering center of the class where the point i is located in the similarity matrix, and the larger the degree is, the more suitable the expression is; s (k, k) is the reference degree of the point k, and elements with the same row serial number and column serial number are used for representing the possible degree of the point k serving as a clustering center in the similarity matrix;for points i and +.>Similarity between; />For point k and point->Similarity between; r (i, k) is the attraction degree between the point i and the point k, and is used for representing the degree of the clustering center of which the point k is suitable as the point i after other data points are considered as potential clustering centers; r (k, k) is the attraction of the point k itself; a (i, k) is the attribution degree between the point i and the point k, and is used for representing the proper degree of selecting the point k as the clustering center by the point i after taking other data points into consideration to form the support of the clustering center by the point k; />For point k and point->Degree of attribution between- >For points i and +.>Degree of attribution between->Is taken as a pointAnd the degree of attraction between points k. The number of final classes is affected by the initial reference (i.e. the reference corresponding to before the first iteration), typically the larger the initial reference the greater the number of final cluster centers. Assuming that all data points are equally likely to be the center of the cluster before the iteration begins, the initial reference level may be set to a fixed value (e.g., the average of all elements in the similarity matrix).

For example, in the above (22), an attenuation coefficient may be introduced for the original water conservancy data corresponding to a certain target timeIteration of the attraction degree (i.e. representative degree) and the attribution degree (i.e. proper degree) may be specifically performed for a plurality of iterations according to the following formula, where the attraction degree between each data point corresponding to the target time and the attribution degree between each data point corresponding to the target time are respectively:

wherein t is the iteration number,for the attenuation coefficient +.>r _t+1 (i, k) is the attraction between point i and point k after the t+1st iteration, a _t+1 (iK) is the degree of attribution between point i and point k after the t+1st iteration, r _t (i, k) is the attraction between point i and point k after the t-th iteration, a _t (i, k) is the degree of attribution between point i and point k after the t-th iteration.

The result of each iteration includes a corresponding determined cluster center, and according to the above formula (3) and the above formula (4), the larger r (i, k) and a (i, k) after the iteration, the more suitable the point k is as the cluster center of the point i, that is, the greater the degree that the point k represents the point i. Repeating the iterative process until the clustering centers in the results of continuous multiple iterations are identical or the preset iteration times are reached, stopping iteration, taking each clustering center corresponding to the stopping iteration as a clustering center of one class, and aggregating all data points corresponding to the target moment into multiple classes.

For ease of understanding, the operation of the above neighbor propagation clustering algorithm is exemplarily described as follows:

firstly, setting input data into a similarity matrix S, iteration times maxit and attenuation coefficientsThe output data is set to a set C of clusters (i.e., classes) C, followed by steps 1 to Step5 below:

step1, initializing r (i, k) =a (i, k) =0;

step2, iterating r (i, k) and a (i, k) according to formula (1) and formula (2);

step3, eliminating oscillations in updating r (i, k) and a (i, k) according to formula (3) and formula (4);

step4, repeating Step2 and Step3, and stopping calculation when the iteration times exceed the maximum value maxit or the number of continuous iterations of the clustering center are not changed;

Step5, forming the data points with the same clustering center into a class, and finally obtaining a set C of C classes.

After the set C of C classes is obtained by steps 1 to 5 described above, isolated data points that do not belong to any class are deleted as unnecessary data points, so that all data points of each class are retained as valid data points for which subsequent operations can be performed.

As a possible implementation manner, the step S108 (i.e. performing data compression on all data points corresponding to each target time by using a huffman compression algorithm to obtain the hydraulic compression number corresponding to each target time) may include the following operation manners:

(31) And counting the occurrence times of each data point corresponding to each target time for each target time, and constructing a variable length coding table corresponding to the target time according to the counting result corresponding to the target time.

(32) And respectively encoding all data points corresponding to each target moment based on a variable length encoding table corresponding to each target moment to obtain water conservancy compression data corresponding to each target moment.

For ease of understanding, the operation of (31) and (32) above is described herein by way of example with particular reference to the following:

For a certain target time, the Huffman compression algorithm encodes the data point corresponding to the target time through a variable length encoding table, wherein the variable length encoding table is obtained by counting the occurrence times of each data point, the data point with more occurrence times is encoded through shorter encoding, and the data point with less occurrence times is encoded through longer encoding, so that the average length and expected value of the character string obtained after encoding are reduced, and the data compression of the data point is realized.

Suppose we have such fifteen data points to encode-62, 65, 65, 70, 20, 62,6F, 70, 20, 62, 65, 65, 72 and 21. The coding steps are as follows:

(1.1) counting the number of occurrences of each data point, and then ranking all data points in order of the number of occurrences from large to small, thereby creating a priority list (as shown in Table 1).

Table 1 priority list

Data points	Number of occurrences
		“62”	3
“65”	4
		“70”	2
“20”	2
		“6F”	2
“72”	1
		“21”	1

(1.2) constructing a Huffman binary tree by taking each data point as a node according to the priority list. When constructing the Huffman binary tree, two data points with the least occurrence number are listed, the two data points are used as child nodes, the father nodes of the two child nodes are generated upwards, the generated father nodes can be represented by auxiliary symbols, the occurrence number of each of the two data points is added to obtain the occurrence number of the auxiliary symbols, then the two data points are deleted from the priority list, and the auxiliary symbols and the occurrence number of the auxiliary symbols are added to the last-second row of the priority list.

(1.3) repeating the operation of the previous step until only one auxiliary symbol "a" in the priority list is completed, and completing the construction of the Huffman binary tree.

(1.4) respectively assigning each left branch on the Huffman binary tree to 0, respectively assigning each right branch on the binary tree to 1, thereby obtaining an assigned Huffman binary tree (shown in fig. 4), and determining the corresponding coding result of each data point through the Huffman binary tree, thereby establishing a variable length coding table (shown in table 2).

Table 2 variable length coding table

Data points	Encoding
		“62”	00
“65”	11
		“70”	101
“20”	011
		“6F”	010
“72”	1000
		“21”	1001

The operation process of the Huffman compression algorithm firstly carries out statistics on the occurrence times of the data points, then codes the data points, and dynamically adjusts the Huffman binary tree while compressing the data points, thereby improving the efficiency of data compression.

As a possible implementation manner, the step of establishing the above distributed data fusion algorithm may include the following operation modes:

(41) Based on the data transmission relation among the sensors, a sensor network of the target water conservancy system is constructed; the sensor network uses nodes to represent sensors, the sensor network uses directional edges among the nodes to represent data transmission relations, and the sensor network uses the direction of the directional edges to represent data transmission directions.

(42) The method comprises the steps of setting trust degree among sensors and self-trust degree of each sensor respectively, setting adjacency degree among the sensors and self-adjacency degree of each sensor respectively according to distribution positions of the sensors relative to a target area, and determining signal sending sensors of each sensor and signal receiving sensors of each sensor respectively according to adjacency degree among the sensors.

(43) And setting a data fusion initial value of each sensor according to whether the sensor acquires original water conservancy data at the first target moment and whether the sensor receives the water conservancy compression data sent by the corresponding signal sending sensor at the first target moment, and setting the data fusion times of the sensor according to the number of the target moments in a preset period.

(44) And calculating the relative trust degree among the sensors according to the trust degree among the sensors and the adjacency degree among the sensors.

(45) And respectively constructing corresponding fusion functions for each sensor according to the relative trust degree among the sensors, the data fusion initial value of each sensor and the data fusion times of each sensor, and forming the fusion functions corresponding to all the sensors into the distributed data fusion algorithm.

For ease of understanding, the above (41) to (45) are exemplarily described herein with reference to fig. 5 as follows:

assume that the target water conservancy system has n sensors, and one sensor set is denoted by v= {1,2, …, n }, where each sensor performs directional information transfer with at least one other sensor, so as to form a sensor network (as shown in fig. 5). For any sensor i, j e V, if sensor i can receive information from sensor j, then there is a directed edge from sensor j to sensor i and vice versa. The structure of the sensor network is represented by g= (V, E), where G is the sensor network, V is the sensor set, and E is the directed edge set. The directed edge from sensor j to sensor i, denoted as (i, j) ∈e, is referred to as the "send" neighbor of sensor i (i.e., the send sensor described above) and sensor i is referred to as the "receive" neighbor of sensor j (i.e., the receive sensor). The "send" neighbor and the "receive" neighbor of sensor i are collectively referred to as the neighbors of sensor i.

For any sensor i, j e V, preset a _ij ∈(0，1]Representing the trust degree of a sensor i to a sensor j, and presetting a _ii ∈(0，1]Representing the degree of confidence of sensor i, constructing matrix a= (a) _ij )∈R ^n×n As a trust matrix of the sensor network G; preset b _ij To represent the adjacency of sensor i to sensor j with 0 or 1, a matrix b= (B) is constructed _ij )∈R ^n×n As an adjacency matrix for the sensor network G; wherein for any sensor i, j E V, if (i, j) E E, b _ij =1, otherwise b _ij =0. For sensor i e V, define a setTo represent a set of "send" neighbors of sensor i, define a setTo represent the "addressee" neighbor set of sensor i.

Defining the preset time period asOne at t ₀ A sequence of sequentially concatenated bounded time domains with starting times, each bounded time domain in the sequence having a starting time as a target time, each bounded time domain in the sequence being represented as T _m ＝[t _m ，t _m+1 ) M=0, 1,2, …. For each bounded time domain T _m Each sensor is at the T _m And acquiring original water conservancy data of the target area through respective data processors and processing the acquired original water conservancy data into corresponding water conservancy compression data.

The T can also be combined with _m Further divided into another at tau ₀ A sequence of sequentially concatenated bounded time domains with starting times, each bounded time domain in the sequence having a starting time as a target time, each bounded time domain in the sequence being represented as q _m’ ＝[τ _m’ ，τ _m’+1 ) M=0, 1,2, …. For T _m If a certain sensor is at each q _m’ Initial water conservancy data of the target area can not be acquired in the sensor, the sensor can be determined as the T _m Corresponding "blind" sensors (e.g., sensor s in fig. 5).

For each bounded time domain T _m Each sensor is at the T _m Collecting original water conservancy data of a target area by respective data processors, processing the collected original water conservancy data into corresponding water conservancy compression data, and determining the position of each sensor in the T _m Respectively carrying out information transfer with each neighbor of the sensor, and carrying out local water conservancy compression data z _i (m) data fusion with data from "send" neighbors to obtain each sensor at the T _m And (5) the fusion result corresponding to the inner part.

For each bounded time domain T _m Assuming that there are no uncertainty factors such as sensor network delay, communication noise, information loss, etc., a corresponding fusion function formula can be defined for each sensor i as follows:

wherein, i is E V,i _l for the first "send" neighbor of sensor i, l is the number of "send" neighbors contained in the "send" neighbor set of sensor i,/A>x _i (m) represents the result of the mth data fusion of the sensor i to the target trace in the fusion process, and x _i (0)＝z _i (m). Z in practical use _i (m) and x _i (m) may be a multidimensional vector, such as containing the coordinates of the target position, the attitude angle and the velocity thereof. For simplicity, the embodiments of the present invention only illustrate z _i (m) and x _i (m) is a one-dimensional case.

Based on the above formula (4.1), each sensor will derive z from its respective data processor _i (m) if sensor i is at T as the initial value of the data fusion for the sensor _m Inside is a "blind" sensor, let x _i (0) The = infinity is used for indicating that the sensor i does not acquire the original water conservancy data, transmitting the result corresponding to each data fusion to the 'receiving' neighbor of the sensor i, receiving the result corresponding to the last data fusion of the 'sending' neighbor of the sensor i, and transmitting z _i (m) and x _i (m) performing next data fusion on the target trace information.

Based on the above formula (4.1), the following formula can be defined even if there are "blind" sensors in the sensor network:

and then, according to the trust degree among the sensors and the adjacency degree among the sensors, calculating the relative trust degree among the sensors according to the following formula:wherein c _ij For the relative trust of sensor I to sensor j (i.e., the trust of sensor I to sensor j accounts for its "send" neighbor I ⁱⁿ _i Ratio of total confidence) of a), a) _ij For confidence of sensor i to sensor j, b _ij For the adjacency of sensor i to sensor j, n is the number of sensors, +.>/>

The following formula was then used to build a data analysis model as follows:

wherein m is the data fusion times of the sensor i in a preset period, i, j epsilon V, c _ij ∈[0，1]；

The following formula may also be used to extend the above formula (4.3) into the expression form of the matrix:

x(m+1)＝Cx(m)，k＝0，1，2，…(4.4)；

wherein x (m) = (x) ₁ (m)，x ₂ (m)，…，x _n (m)) ^T ，C＝(c _ij )∈R ^n×n Referred to as the relative trust matrix of the sensor network G. Obviously, matrix C is a row random matrix. It is considered that the degree of trust does not change with time, and therefore the matrix C does not change with time either.

After the respective fusion function of each sensor is defined according to the definition mode, the fusion functions corresponding to all the sensors can be formed into the distributed data fusion algorithm.

Based on the above-mentioned establishment procedure of the above-mentioned distributed data fusion algorithm, the above-mentioned step S110 (i.e. performing data fusion on the water conservancy compressed data corresponding to each target moment by adopting the pre-established distributed data fusion algorithm to obtain the water conservancy fusion data corresponding to each target moment) may include the following operation modes: for each sensor, performing first data fusion on the data fusion initial value of the sensor by adopting a fusion function corresponding to the sensor, sending the result of the first data fusion of the sensor to a receiving sensor of the sensor, and then performing next data fusion on the result of the last data fusion of the sensor and the water conservancy data to be fused corresponding to the next data fusion of the sensor by adopting the fusion function until the result of the last data fusion of the sensor is obtained; the data fusion times of the sensor are equal to the number of target moments in a preset period; the water conservancy data to be fused corresponding to the next data fusion of the sensor comprises: and the sensor acquires the water conservancy compression data corresponding to the original water conservancy data at the next target moment and/or the last data fusion result of the corresponding signaling sensor received by the sensor.

Illustratively, following the previous example, the data fusion process for sensor i may include the following steps according to equations (4.1) through (4.4) above:

in step 1, when m=0, the sensor i obtains x through its own data processor i _i (0) And x is taken as _i (0) A fusion processor i sent to the sensor i itself;

step 2, the sensor i uses the data processor i of the sensor i to make x _i (0) The 'receiving' neighbor sent to the sensor i and receiving the 'sending' neighbor j sent by the fusion processor j

Step 3, the sensor i uses the formula (4.1) to pair x through the own fusion processor i _j (0) And x locally _i (0) Data fusion is carried out, and x is calculated by using a formula (4.3) _i (m+1)；

Step 4, the sensor i uses the fusion processor i of the sensor i to make x _i (m+1) the "addressee" neighbor sent to sensor i;

step 5, if for any i, j e V, there is x _i (m+1)＝x _j (m+1) the fusion is ended; otherwise, letm=m+1, returning to step 3.

If the sensor network takes a certain sensor as the center, other sensors are all neighbors of the sensor, and after the data fusion process, the water conservancy fusion data of the other sensors are the same as the water conservancy fusion data of the sensor.

Based on the above-mentioned water conservancy panoramic information sensing method, the embodiment of the invention also provides a water conservancy panoramic information sensing device, as shown in fig. 6, which can include the following modules:

The data acquisition module 602 is configured to acquire raw water conservancy data of a target area in real time through a plurality of sensors of the target water conservancy system; each piece of original water conservancy data has respective data acquisition time and respective data size.

And the data transmission module 604 is used for respectively transmitting the original water conservancy data acquired by the different sensors at each target moment in a preset period to target terminal equipment of the target water conservancy system by adopting a TSMP algorithm.

The data cleaning module 606 is configured to perform data cleaning on the raw water conservancy data corresponding to each target time received by the target terminal device by using a neighbor propagation clustering algorithm, so as to obtain multiple classes corresponding to each target time; wherein each class contains a plurality of data points, the cluster center of each class is a data point, and the cluster centers of different classes are different.

The data compression module 608 is configured to perform data compression on all data points corresponding to each target time by using a huffman compression algorithm, so as to obtain hydraulic compression data corresponding to each target time.

The data fusion module 610 is configured to perform data fusion on the hydraulic compression data corresponding to each target moment by using a pre-established distributed data fusion algorithm, so as to obtain hydraulic fusion data corresponding to each target moment; the distributed data fusion algorithm is established based on the data transmission relation among the sensors.

According to the water conservancy panoramic information sensing device provided by the embodiment of the invention, the original water conservancy data of a target area are acquired in real time through a plurality of sensors of a target water conservancy system, the original water conservancy data acquired at each target moment in a preset period of time by different sensors are respectively transmitted to target terminal equipment of the target water conservancy system by adopting a TSMP algorithm, and the original water conservancy data corresponding to each target moment received by the target terminal equipment is respectively subjected to data cleaning by adopting a neighbor propagation clustering algorithm, so that a plurality of classes corresponding to each target moment are obtained, and each class comprises a plurality of data points; carrying out data compression on all data points corresponding to each target moment by adopting a Huffman compression algorithm, thereby obtaining water conservancy compression data corresponding to each target moment; and carrying out data fusion on the water conservancy compression data corresponding to each target moment by adopting a pre-established distributed data fusion algorithm, so as to obtain the water conservancy fusion data corresponding to each target moment. By adopting the technology, the efficiency and accuracy of water conservancy system information perception can be improved, so that the refinement, timeliness and intelligent level of water conservancy information management work are improved.

The data transmission module 604 may also be configured to: for the original water conservancy data corresponding to each target moment, respectively generating respective request messages through each sensor; the request message comprises a request synchronous message, a request data message, a time stamp for starting data transmission and a data transmission delay; the request synchronous message has a synchronous mark for representing whether the data to be transmitted are in time synchronization or not, and the request data message has mark data for representing whether the data to be transmitted are valid or not; for each generated request message, setting a synchronous mark of the request message according to whether a sensor generating the request message is communicated with the target terminal device before generating the request message and whether data to be transmitted of the sensor have time synchronization, setting mark data of the request message according to whether the target terminal device receives the data to be transmitted of the sensor, and sending the request message with the synchronous mark and the mark data being set to the target terminal device; and transmitting the original water conservancy data acquired by different sensors at each target moment in a preset period to target terminal equipment of the target water conservancy system according to each request message received by the target terminal equipment.

The data transmission module 604 may also be configured to: if the sensor generating the request message has communicated with the target terminal device before generating the request message and the data to be transmitted of the sensor is time-synchronized beyond a preset time period before the sensor generates the request message, or the sensor generating the request message has not communicated with the target terminal device before generating the request message, setting the synchronization flag of the request message to 1; and if the data to be transmitted of the sensor does not exceed the preset duration for time synchronization before the sensor generates the request message and the sensor does not communicate with the target terminal equipment after the time synchronization of the data to be transmitted, setting the synchronization flag of the request message to 0.

The data transmission module 604 may also be configured to: if the target terminal equipment does not receive the data to be transmitted of the sensor, setting the flag data of the request message to 0; and if the target terminal equipment receives the data to be transmitted of the sensor, setting the flag data of the request message to be 1.

The data transmission module 604 may also be configured to: for each request message received by the target terminal equipment, if the synchronization mark of the request message is 1 and the mark data of the request message is 0, performing time synchronization on the data to be transmitted corresponding to the request message, feeding back a response message of the request message to the sensor through the target terminal equipment, and not transmitting the data to be transmitted from the sensor for transmitting the request message to the target terminal equipment; for each request message received by the target terminal equipment, if the synchronization mark of the request message is 1 and the mark data of the request message is 1, performing time synchronization on the target data corresponding to the request message, feeding back a response message of the request message to the sensor through the target terminal equipment, and then transmitting the data to be transmitted from the sensor for transmitting the request message to the target terminal equipment; for each request message received by the target terminal device, if the synchronization mark of the request message is 0 and the mark data of the request message is 0, determining the data to be transmitted corresponding to the request message as invalid data, and then not executing any operation; for each request message received by the target terminal device, if the synchronization flag of the request message is 0 and the flag data of the request message is 1, feeding back a response message of the request message to the sensor through the target terminal device, and then transmitting the data to be transmitted from the sensor sending the request message to the target terminal device.

The data cleansing module 606 described above may also be used to: for the original water conservancy data corresponding to each target moment received by the target terminal equipment, taking each original water conservancy data corresponding to the target moment as a data point, respectively calculating the similarity between data points and the reference degree of each data point, and respectively calculating the attraction between data points, the attribution between data points and the self attraction of each data point based on the similarity between data points and the reference degree of each data point; for the original water conservancy data corresponding to each target moment, respectively carrying out multiple iterations on the attraction degree between each data point corresponding to the target moment and the attribution degree between each data point corresponding to the target moment based on a preset attenuation coefficient; the result of each iteration comprises a corresponding clustering center; stopping iteration until the clustering centers in the results of continuous multiple iterations are identical or reach the preset iteration times, taking each clustering center corresponding to the stopped iteration as a clustering center of one class, and aggregating all data points corresponding to the target moment into multiple classes.

The data compression module 608 described above may also be used to: counting the occurrence times of each data point corresponding to each target time for each target time, and constructing a variable length coding table corresponding to the target time according to the counting result corresponding to the target time; and respectively encoding all data points corresponding to each target moment based on a variable length encoding table corresponding to each target moment to obtain water conservancy compression data corresponding to each target moment.

Referring to fig. 6, the apparatus may further include:

the fusion algorithm building module 612 is configured to build a sensor network of the target water conservancy system based on the data transmission relationship between the sensors; the sensor network uses nodes to represent sensors, the sensor network uses directional edges among the nodes to represent data transmission relations, and the sensor network uses the direction of the directional edges to represent data transmission directions; setting the trust degree among the sensors and the self-trust degree of each sensor respectively, setting the adjacency degree among the sensors and the self-adjacency degree of each sensor respectively according to the distribution position of each sensor relative to a target area, and determining the signal sending sensor of each sensor and the signal receiving sensor of each sensor respectively according to the adjacency degree among the sensors; for each sensor, setting a data fusion initial value of the sensor according to whether the sensor acquires original water conservancy data at a first target moment and whether the sensor receives water conservancy compression data sent by a corresponding signal sending sensor at the first target moment, and setting the data fusion times of the sensor according to the number of target moments in a preset period; according to the trust degree among the sensors and the adjacency degree among the sensors, calculating to obtain the relative trust degree among the sensors; and respectively constructing corresponding fusion functions for each sensor according to the relative trust degree among the sensors, the data fusion initial value of each sensor and the data fusion times of each sensor, and forming the fusion functions corresponding to all the sensors into the distributed data fusion algorithm.

Based on this, the data fusion module 610 described above may also be used to: for each sensor, performing first data fusion on the data fusion initial value of the sensor by adopting a fusion function corresponding to the sensor, sending the result of the first data fusion of the sensor to a receiving sensor of the sensor, and then performing next data fusion on the result of the last data fusion of the sensor and the water conservancy data to be fused corresponding to the next data fusion of the sensor by adopting the fusion function until the result of the last data fusion of the sensor is obtained; the data fusion times of the sensor are equal to the number of target moments in a preset period; the water conservancy data to be fused corresponding to the next data fusion of the sensor comprises: and the sensor acquires the water conservancy compression data corresponding to the original water conservancy data at the next target moment and/or the last data fusion result of the corresponding signaling sensor received by the sensor.

The water conservancy panoramic information sensing device provided by the embodiment of the invention has the same implementation principle and the same generated technical effects as the water conservancy panoramic information sensing method embodiment, and for the sake of brief description, the corresponding contents in the method embodiment can be referred to where the device embodiment part is not mentioned.

The relative steps, numerical expressions and characters of the components and steps set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.

Finally, it should be noted that: the above examples are only specific embodiments of the present invention, and are not intended to limit the scope of the present invention, but it should be understood by those skilled in the art that the present invention is not limited thereto, and that the present invention is described in detail with reference to the foregoing examples: any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or perform equivalent substitution of some of the technical features, while remaining within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A water conservancy panorama information sensing method, the method comprising:

acquiring original water conservancy data of a target area in real time through a plurality of sensors of a target water conservancy system; each original water conservancy data has respective data acquisition time and respective data size; the raw water conservancy data comprises at least one of the following: water level data, rainfall data, flow rate data, water quality data;

The method comprises the steps of respectively transmitting original water conservancy data acquired at each target moment in a preset period of time by adopting a TSMP algorithm to target terminal equipment of a target water conservancy system;

adopting a neighbor propagation clustering algorithm to respectively carry out data cleaning on the original water conservancy data corresponding to each target moment received by the target terminal equipment, and obtaining a plurality of classes corresponding to each target moment; wherein each class comprises a plurality of data points, the clustering center of each class is a data point, and the clustering centers of different classes are different;

carrying out data compression on all data points corresponding to each target moment by adopting a Huffman compression algorithm to obtain water conservancy compression data corresponding to each target moment;

carrying out data fusion on the water conservancy compression data corresponding to each target moment by adopting a pre-established distributed data fusion algorithm to obtain water conservancy fusion data corresponding to each target moment; the distributed data fusion algorithm is established based on the data transmission relation among the sensors;

the method for transmitting the original water conservancy data acquired by different sensors at each target moment in a preset period to target terminal equipment of the target water conservancy system by adopting a TSMP algorithm comprises the following steps:

For the original water conservancy data corresponding to each target moment, respectively generating respective request messages through each sensor; the request message comprises a request synchronous message, a request data message, a time stamp for starting data transmission and a data transmission delay; the request synchronous message has a synchronous mark for representing whether the data to be transmitted are in time synchronization or not, and the request data message has mark data for representing whether the data to be transmitted are valid or not;

for each generated request message, setting a synchronous mark of the request message according to whether a sensor generating the request message is communicated with the target terminal device before generating the request message and whether data to be transmitted of the sensor have time synchronization, setting mark data of the request message according to whether the target terminal device receives the data to be transmitted of the sensor, and sending the request message with the synchronous mark and the mark data being set to the target terminal device;

transmitting the original water conservancy data acquired at each target moment in a preset period of time by different sensors to target terminal equipment of the target water conservancy system according to each request message received by the target terminal equipment;

And (3) respectively carrying out data cleaning on the original water conservancy data corresponding to each target moment received by the target terminal equipment by adopting a neighbor propagation clustering algorithm to obtain a plurality of classes corresponding to each target moment, wherein the steps comprise:

for the original water conservancy data corresponding to each target moment received by the target terminal equipment, taking each original water conservancy data corresponding to the target moment as a data point, respectively calculating the similarity between data points and the reference degree of each data point, and respectively calculating the attraction between data points, the attribution between data points and the self attraction of each data point based on the similarity between data points and the reference degree of each data point; the self-priming degree of each data point is the attraction degree between each data point and the self data point, and the similarity is a negative Euclidean distance;

for the original water conservancy data corresponding to each target moment, carrying out multiple iterations on the attraction degree among the data points corresponding to the target moment and the attribution degree among the data points corresponding to the target moment respectively according to the following formula based on a preset attenuation coefficient: r is (r) _{t + 1} (i,k)=(1 - ∂)r _{t + 1} (i,k) + ∂ r _t (i,k)，a _{t + 1} (i,k)=(1 - ∂)a _{t + 1} (i,k) + ∂ a _t (i, k); wherein t is the iteration number, ∂ is the attenuation coefficient, 0< ∂ < 1，r _{t + 1} (i, k) is the attraction between data point i and data point k after the t+1st iteration, a _{t + 1} (i, k) is the degree of ownership between data point i and data point k after the t+1st iteration, r _t (i, k) is the attraction between data point i and data point k after the t-th iteration, a _t (i, k) is the degree of attribution between the data point i and the data point k after the t-th iteration, and the result of each iteration comprises a corresponding clustering center; stopping iteration until the clustering centers in the results of continuous multiple iterations are identical or reach the preset iteration times, and stopping each corresponding iterationThe clustering centers are used as the clustering centers of one class, and all data points corresponding to the target moment are aggregated into a plurality of classes.

2. The method according to claim 1, wherein the step of setting the synchronization flag of the request message according to whether the sensor generating the request message has communicated with the target terminal device before generating the request message and whether time synchronization of data to be transmitted of the sensor occurs, comprises:

if the sensor generating the request message has communicated with the target terminal device before generating the request message and the data to be transmitted of the sensor is time-synchronized beyond a preset time period before the sensor generates the request message, or the sensor generating the request message has not communicated with the target terminal device before generating the request message, setting the synchronization flag of the request message to 1;

And if the data to be transmitted of the sensor does not exceed the preset duration for time synchronization before the sensor generates the request message and the sensor does not communicate with the target terminal equipment after the time synchronization of the data to be transmitted, setting the synchronization flag of the request message to 0.

3. The method according to claim 2, wherein the step of setting flag data of the request message according to whether the target terminal device receives data to be transmitted of the sensor, comprises:

if the target terminal equipment does not receive the data to be transmitted of the sensor, setting the flag data of the request message to 0;

and if the target terminal equipment receives the data to be transmitted of the sensor, setting the flag data of the request message to be 1.

4. A method according to claim 3, wherein the step of transmitting the raw water conservancy data collected by the different sensors at each target moment in a preset period to the target terminal device of the target water conservancy system according to each request message received by the target terminal device comprises the steps of:

for each request message received by the target terminal equipment, if the synchronization mark of the request message is 1 and the mark data of the request message is 0, performing time synchronization on the data to be transmitted corresponding to the request message, feeding back a response message of the request message to the sensor through the target terminal equipment, and not transmitting the data to be transmitted from the sensor for transmitting the request message to the target terminal equipment;

For each request message received by the target terminal equipment, if the synchronization mark of the request message is 1 and the mark data of the request message is 1, performing time synchronization on the target data corresponding to the request message, feeding back a response message of the request message to the sensor through the target terminal equipment, and then transmitting the data to be transmitted from the sensor for transmitting the request message to the target terminal equipment;

for each request message received by the target terminal device, if the synchronization mark of the request message is 0 and the mark data of the request message is 0, determining the data to be transmitted corresponding to the request message as invalid data, and then not executing any operation;

for each request message received by the target terminal device, if the synchronization flag of the request message is 0 and the flag data of the request message is 1, feeding back a response message of the request message to the sensor through the target terminal device, and then transmitting the data to be transmitted from the sensor sending the request message to the target terminal device.

5. The method of claim 1, wherein the step of establishing the distributed data fusion algorithm comprises:

Based on the data transmission relation among the sensors, a sensor network of the target water conservancy system is constructed; the sensor network uses nodes to represent sensors, the sensor network uses directional edges among the nodes to represent data transmission relations, and the sensor network uses the direction of the directional edges to represent data transmission directions;

setting the trust degree among the sensors and the self-trust degree of each sensor respectively, setting the adjacency degree among the sensors and the self-adjacency degree of each sensor respectively according to the distribution position of each sensor relative to a target area, and determining the signal sending sensor of each sensor and the signal receiving sensor of each sensor respectively according to the adjacency degree among the sensors;

for each sensor, setting a data fusion initial value of the sensor according to whether the sensor acquires original water conservancy data at a first target moment and whether the sensor receives water conservancy compression data sent by a corresponding signal sending sensor at the first target moment, and setting the data fusion times of the sensor according to the number of target moments in a preset period;

according to the trust degree among the sensors and the adjacency degree among the sensors, calculating to obtain the relative trust degree among the sensors;

And respectively constructing corresponding fusion functions for each sensor according to the relative trust degree among the sensors, the data fusion initial value of each sensor and the data fusion times of each sensor, and forming the fusion functions corresponding to all the sensors into the distributed data fusion algorithm.

6. The method of claim 5, wherein the step of performing data fusion on the hydraulic compression data corresponding to each target time by using a pre-established distributed data fusion algorithm to obtain hydraulic fusion data corresponding to each target time comprises the steps of:

for each sensor, performing first data fusion on the data fusion initial value of the sensor by adopting a fusion function corresponding to the sensor, sending the result of the first data fusion of the sensor to a receiving sensor of the sensor, and then performing next data fusion on the result of the last data fusion of the sensor and the water conservancy data to be fused corresponding to the next data fusion of the sensor by adopting the fusion function until the result of the last data fusion of the sensor is obtained; the data fusion times of the sensor are equal to the number of target moments in a preset period; the water conservancy data to be fused corresponding to the next data fusion of the sensor comprises: and the sensor acquires the water conservancy compression data corresponding to the original water conservancy data at the next target moment and/or the last data fusion result of the corresponding signaling sensor received by the sensor.

7. The method of claim 1, wherein the step of performing data compression on all data points corresponding to each target time by using a huffman compression algorithm to obtain hydraulic compression data corresponding to each target time comprises the steps of:

counting the occurrence times of each data point corresponding to each target time for each target time, and constructing a variable length coding table corresponding to the target time according to the counting result corresponding to the target time;

and respectively encoding all data points corresponding to each target moment based on a variable length encoding table corresponding to each target moment to obtain water conservancy compression data corresponding to each target moment.

8. A water conservancy panorama information sensing device, wherein said device comprises:

the data acquisition module is used for acquiring original water conservancy data of a target area in real time through a plurality of sensors of the target water conservancy system; each original water conservancy data has respective data acquisition time and respective data size; the raw water conservancy data comprises at least one of the following: water level data, rainfall data, flow rate data, water quality data;

The data transmission module is used for respectively transmitting the original water conservancy data acquired at each target moment in a preset period of time by using a TSMP algorithm to target terminal equipment of the target water conservancy system;

the data cleaning module is used for respectively carrying out data cleaning on the original water conservancy data corresponding to each target moment received by the target terminal equipment by adopting a neighbor propagation clustering algorithm to obtain a plurality of classes corresponding to each target moment; wherein each class comprises a plurality of data points, the clustering center of each class is a data point, and the clustering centers of different classes are different;

the data compression module is used for carrying out data compression on all data points corresponding to each target moment by adopting a Huffman compression algorithm to obtain water conservancy compression data corresponding to each target moment;

the data fusion module is used for carrying out data fusion on the water conservancy compression data corresponding to each target moment by adopting a pre-established distributed data fusion algorithm to obtain water conservancy fusion data corresponding to each target moment; the distributed data fusion algorithm is established based on the data transmission relation among the sensors;

the data transmission module is further used for:

The data cleaning module is also used for:

for the original water conservancy data corresponding to each target moment, carrying out multiple iterations on the attraction degree among the data points corresponding to the target moment and the attribution degree among the data points corresponding to the target moment respectively according to the following formula based on a preset attenuation coefficient: r is (r) _{t + 1} (i,k)=(1 - ∂)r _{t + 1} (i,k) + ∂ r _t (i,k)，a _{t + 1} (i,k)=(1 - ∂)a _{t + 1} (i,k) + ∂ a _t (i, k); wherein t is the iteration number, ∂ is the attenuation coefficient, 0< ∂ < 1，r _{t + 1} (i, k) is the attraction between data point i and data point k after the t+1st iteration, a _{t + 1} (i, k) is the degree of ownership between data point i and data point k after the t+1st iteration, r _t (i, k) is the attraction between data point i and data point k after the t-th iteration, a _t (i, k) is the degree of attribution between the data point i and the data point k after the t-th iteration, and the result of each iteration comprises a corresponding clustering center; stopping iteration until the clustering centers in the results of continuous multiple iterations are identical or reach the preset iteration times, taking each clustering center corresponding to the stopped iteration as a clustering center of one class, and aggregating all data points corresponding to the target moment into multiple classes.