Booster station intelligent inspection system and method based on Internet of Things

Abstract

The invention provides an intelligent inspection system and method for a booster station based on the Internet of Things, and relates to the technical field of intelligent inspection, and the system employs an artificial intelligence technology based on deep learning to carry out the real-time temperature monitoring and data analysis of a temperature sensitive point in the booster station, so as to capture the temperature time sequence correlation characteristics of different time scales. And the abnormal temperature fluctuation condition of the temperature sensitive point is intelligently identified based on the self-adaptive fusion feature of the multi-time scale temperature information. The method has the advantages that real-time monitoring and accurate early warning of temperature sensitive points in the booster station can be achieved, the situations of false alarm and missing alarm are effectively reduced, and the intelligent operation and maintenance level of the booster station is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent inspection, and in particular relates to an intelligent inspection system and method for a substation based on the Internet of Things. Background Art

[0002] With the rapid development of power systems and the growing demand for intelligence, traditional booster station operation and maintenance methods have been unable to meet the needs of efficient, safe and economical operation of modern power systems. Especially for offshore booster stations operating in harsh environments such as high temperature, high humidity, and high salt fog, real-time monitoring of temperature-sensitive points and abnormal warnings are particularly important.

[0003] Traditional inspection methods rely on regular manual inspections and cannot achieve real-time monitoring of temperature-sensitive points, resulting in the inability to detect and respond to temperature anomalies in a timely manner. This is not only inefficient, but also easily affected by human factors. In recent years, with the rapid development of Internet of Things technology, new solutions have been provided for intelligent inspection of booster stations. Through Internet of Things technology, real-time monitoring of booster station equipment can be achieved, providing the possibility for equipment status monitoring and fault warning.

[0004] At present, the technical means for temperature monitoring of booster stations mainly include temperature sensor collection, data analysis and abnormal warning. However, most intelligent inspection systems use simple threshold comparison methods in data analysis, lacking in-depth understanding of temperature change trends and patterns, which can easily lead to false alarms and missed alarms. Therefore, an optimized intelligent inspection system and method for booster stations based on the Internet of Things is expected. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art and provide a new technical solution of a booster station intelligent inspection system and method based on the Internet of Things.

[0006] According to a first aspect of the present invention, there is provided a booster station intelligent inspection system based on the Internet of Things, which comprises:

[0007] The temperature monitoring module is used to collect the real-time temperature time series of temperature-sensitive points through the temperature sensor of the Internet of Things;

[0008] A data transmission module, used to transmit the time series of the real-time temperature to a background booster station intelligent inspection server through the wireless communication network of the Internet of Things;

[0009] A multi-scale temperature time series feature extraction module is used to extract multi-scale temperature time series features on the time series of the real-time temperature in the background booster station intelligent inspection server to obtain a first time scale temperature time series associated feature vector and a second time scale temperature time series associated feature vector;

[0010] A multi-scale feature balanced fusion module is used to perform adaptive balanced fusion on the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector in the background booster station intelligent inspection server to obtain a multi-time scale temperature time series associated feature vector;

[0011] The temperature monitoring result generation module is used to determine the monitoring result based on the multi-time scale temperature time series correlation feature vector in the background booster station intelligent inspection server.

[0012] In the above-mentioned IoT-based substation intelligent inspection system, the multi-scale temperature time series feature extraction module is used to: input the real-time temperature time series into a temperature time series feature extraction dual-stream network comprising a 1DCNN layer and a Bi-LSTM layer to obtain the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector.

[0013] In the above-mentioned IoT-based substation intelligent inspection system, the multi-scale feature balanced fusion module is used to: input the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector into the balanced threshold feature vector adaptive fusion module to obtain the multi-time scale temperature time series associated feature vector.

[0014] In the above-mentioned IoT-based substation intelligent inspection system, the multi-scale feature balanced fusion module includes: a multi-dimensional fusion unit, which is used to pass the first time scale temperature series association feature vector and the second time scale temperature series association feature vector through a multi-dimensional fusion module to obtain a first multi-time scale temperature series fusion feature vector, a second multi-time scale temperature series fusion feature vector and a third multi-time scale temperature series fusion feature vector; a balanced threshold value calculation unit, which is used to respectively calculate the balanced threshold values ​​of the first multi-time scale temperature series fusion feature vector, the second multi-time scale temperature series fusion feature vector and the third multi-time scale temperature series fusion feature vector to obtain a first balanced threshold value, a second balanced threshold value and a third balanced threshold value; a balanced fusion unit, which is used to perform weighted fusion of the first time scale temperature series association feature vector and the second time scale temperature series association feature vector based on the first balanced threshold value, the second balanced threshold value and the third balanced threshold value to obtain the multi-time scale temperature series association feature vector.

[0015] In the above-mentioned IoT-based substation intelligent inspection system, the multi-dimensional fusion unit includes: a cascade fusion subunit, used to cascade the first time scale temperature series association feature vector and the second time scale temperature series association feature vector to obtain the first multi-time scale temperature series fusion feature vector; an element addition fusion subunit, used to add the first time scale temperature series association feature vector and the second time scale temperature series association feature vector by position to obtain the second multi-time scale temperature series fusion feature vector; an element point multiplication fusion subunit, used to point multiply the first time scale temperature series association feature vector and the second time scale temperature series association feature vector by position to obtain the third multi-time scale temperature series fusion feature vector.

[0016] In the above-mentioned IoT-based substation intelligent inspection system, the balance threshold value calculation unit is used to: multiply the first multi-time-scale temperature series fusion feature vector by a first predetermined transformation vector to obtain a first threshold scoring coefficient; add the first threshold scoring coefficient and the first bias parameter and then pass the sigmoid activation function to obtain the first balance threshold value.

[0017] In the above-mentioned IoT-based substation intelligent inspection system, the balance fusion unit includes: a weight parameter calculation subunit, which is used to determine the first weight parameter and the second weight parameter based on the first balance threshold value, the second balance threshold value and the third balance threshold value, wherein the first weight parameter is the average of the first balance threshold value, the second balance threshold value and the third balance threshold value, and the second weight parameter is the difference between one and the first weight parameter; a position-by-position weighting subunit, which is used to weight the first time scale temperature time series association feature vector position-by-position with the first weight parameter to obtain a weighted first time scale temperature time series association feature vector, and to weight the second time scale temperature time series association feature vector position-by-position with the second weight parameter to obtain a weighted second time scale temperature time series association feature vector; a fusion subunit, which is used to add the corresponding elements of the weighted first time scale temperature time series association feature vector and the weighted second time scale temperature time series association feature vector to obtain the multi-time scale temperature time series association feature vector.

[0018] In the above-mentioned IoT-based substation intelligent inspection system, the temperature monitoring result generation module is used to: input the multi-time-scale temperature series correlation feature vector into the classifier-based temperature monitor to obtain the monitoring result, and the monitoring result is used to indicate whether there is abnormal temperature fluctuation at the temperature sensitive point.

[0019] According to a second aspect of the present invention, there is provided a method for intelligent inspection of a booster station based on the Internet of Things, which comprises:

[0020] The temperature sensor of the Internet of Things collects the time series of the real-time temperature of the temperature-sensitive point;

[0021] The time series of the real-time temperature is transmitted to the background booster station intelligent inspection server through the wireless communication network of the Internet of Things;

[0022] In the background booster station intelligent inspection server, multi-scale temperature time series feature extraction is performed on the time series of the real-time temperature to obtain a first time scale temperature time series associated feature vector and a second time scale temperature time series associated feature vector;

[0023] In the background booster station intelligent inspection server, the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector are adaptively balanced and fused to obtain a multi-time scale temperature time series associated feature vector;

[0024] In the background booster station intelligent inspection server, a monitoring result is determined based on the multi-time-scale temperature time series correlation feature vector.

[0025] In the above-mentioned booster station intelligent inspection method based on the Internet of Things, in the background booster station intelligent inspection server, the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector are adaptively balanced and fused to obtain a multi-time scale temperature time series associated feature vector, including:

[0026] The first time scale temperature time series correlation feature vector and the second time scale temperature time series correlation feature vector are input into the equalization threshold feature vector adaptive fusion module to obtain the multi-time scale temperature time series correlation feature vector.

[0027] A technical effect of the present invention is:

[0028] The IoT-based booster station intelligent inspection system and method provided in this application uses artificial intelligence technology based on deep learning to perform real-time temperature monitoring and data analysis on temperature-sensitive points inside the booster station, so as to capture the temperature time series correlation characteristics of different time scales, and intelligently identify abnormal temperature fluctuations of temperature-sensitive points based on the adaptive fusion characteristics of multi-time scale temperature information. In this way, real-time monitoring and accurate early warning of temperature-sensitive points inside the booster station can be achieved, effectively reducing false alarms and missed alarms, and improving the intelligent operation and maintenance level of the booster station. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a block diagram of an IoT-based intelligent inspection system for substations according to an embodiment of the present application.

[0030] Figure 2 Schematic diagram of the architecture of an IoT-based intelligent inspection system for substations according to an embodiment of the present application.

[0031] Figure 3 It is a block diagram of a multi-scale feature balancing fusion module in a booster station intelligent inspection system based on the Internet of Things according to an embodiment of the present application.

[0032] Figure 4 It is a block diagram of a multi-dimensional fusion unit in the IoT-based substation intelligent inspection system according to an embodiment of the present application.

[0033] Figure 5 It is a block diagram of the balancing fusion unit in the IoT-based substation intelligent inspection system according to an embodiment of the present application.

[0034] Figure 6 The present invention is a flowchart of an intelligent inspection method for booster stations based on the Internet of Things according to an embodiment of the present application. DETAILED DESCRIPTION

[0035] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present application.

[0036] The embodiments of the present application will be described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and should not be construed as limitations on the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0037] The term "first" or "second" in the specification and claims of this application may include one or more of the features explicitly or implicitly. In the description of this application, unless otherwise specified, "plurality" means two or more. In addition, "and / or" in the specification and claims means at least one of the connected objects, and the character " / " generally means that the objects connected before and after are in an "or" relationship.

[0038] As mentioned in the above background technology, in the prior art, most of the technical means for temperature monitoring of booster stations still remain at the traditional simple threshold comparison method. Although this method realizes real-time monitoring of temperature to a certain extent, its application effect is not ideal due to the lack of in-depth analysis and understanding of temperature change trends and patterns. In practical applications, this simple method often leads to frequent false alarms and missed reports, affecting the normal operation and maintenance of the booster station. In response to the above technical problems, the technical concept of the present application is to use artificial intelligence technology based on deep learning to perform real-time temperature monitoring and data analysis on temperature-sensitive points inside the booster station, so as to capture the temperature time series correlation characteristics of different time scales, and intelligently identify abnormal temperature fluctuations of temperature-sensitive points based on the adaptive fusion characteristics of multi-time scale temperature information. In this way, real-time monitoring and accurate early warning of temperature-sensitive points inside the booster station can be realized, effectively reducing false alarms and missed reports, and improving the intelligent operation and maintenance level of the booster station.

[0039] Figure 1 It is a block diagram of an IoT-based intelligent inspection system for substations according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of the architecture of the booster station intelligent inspection system based on the Internet of Things according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to an embodiment of the present application, a booster station intelligent inspection system 100 based on the Internet of Things includes:

[0040] The temperature monitoring module 110 is used to collect the time series of the real-time temperature of the temperature sensitive point through the temperature sensor of the Internet of Things;

[0041] The data transmission module 120 is used to transmit the time series of the real-time temperature to the background booster station intelligent inspection server through the wireless communication network of the Internet of Things;

[0042] A multi-scale temperature time series feature extraction module 130 is used to extract multi-scale temperature time series features on the time series of the real-time temperature in the background booster station intelligent inspection server to obtain a first time scale temperature time series associated feature vector and a second time scale temperature time series associated feature vector;

[0043] A multi-scale feature balanced fusion module 140 is used to perform adaptive balanced fusion on the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector in the background booster station intelligent inspection server to obtain a multi-time scale temperature time series associated feature vector;

[0044] The temperature monitoring result generating module 150 is used to determine the monitoring result based on the multi-time-scale temperature time series correlation feature vector in the background booster station intelligent inspection server.

[0045] In the above-mentioned booster station intelligent inspection system 100 based on the Internet of Things, the temperature monitoring module 110 is used to collect the time series of the real-time temperature of the temperature sensitive point through the temperature sensor of the Internet of Things. It should be understood that the damage of electrical equipment is usually a gradual process, and there may be multiple intermediate states between the normal state and the fault state. Simple threshold comparison methods can often only determine whether the equipment is currently in a normal or faulty state, but cannot predict the potential abnormal conditions of the equipment. In addition, the normal operating temperature range of electrical equipment may be different under different operating environments (such as temperature, humidity, pressure, etc.). A single threshold cannot meet the monitoring requirements under all environmental conditions, which may cause the normal operating state of the equipment to be misjudged as a fault state under certain environmental conditions, or fail to provide timely warning when a fault occurs. To this end, in the technical solution of the present application, the time series of the real-time temperature of the temperature sensitive points inside the booster station is obtained to obtain the real-time temperature change information during the operation of the booster station equipment, and the abnormal temperature fluctuation is identified based on the time series dynamic change characteristics of the temperature.

[0046] In the above-mentioned booster station intelligent inspection system 100 based on the Internet of Things, the data transmission module 120 is used to transmit the time series of the real-time temperature to the background booster station intelligent inspection server through the wireless communication network of the Internet of Things. It should be understood that the background server has powerful data processing and analysis capabilities. After the real-time temperature data is transmitted to the server, the computing resources of the background server can be used to perform real-time analysis of the data to intelligently identify abnormal temperature fluctuations at temperature-sensitive points. In addition, by transmitting the real-time temperature data to the background server, remote monitoring and centralized management of the booster station can be achieved, so that operators and managers can access and monitor the status of the booster station through the network at any time and any place, without having to go to the site for inspection in person.

[0047] In the above-mentioned IoT-based substation intelligent inspection system 100, the multi-scale temperature time series feature extraction module 130 is used to perform multi-scale temperature time series feature extraction on the real-time temperature time series in the background substation intelligent inspection server to obtain a first time scale temperature time series associated feature vector and a second time scale temperature time series associated feature vector.

[0048] In a specific example of the present application, the processing method for multi-scale temperature time series feature extraction of the real-time temperature time series is to input the real-time temperature time series into a temperature time series feature extraction dual-stream network including a 1DCNN layer and a Bi-LSTM layer to obtain the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector. It should be understood that, considering that temperature data is time series data, it has the characteristic of changing over time, and because there are many internal devices in the booster station and the scene is complex, the temperature changes of its temperature sensitive points are usually complex and variable, and may show different characteristics and laws at different time scales. Therefore, in the technical solution of the present application, a temperature time series feature extraction dual-stream network including a 1DCNN layer and a Bi-LSTM layer is used to perform a comprehensive and in-depth data analysis of the real-time temperature time series.

[0049] Specifically, the 1DCNN (one-dimensional convolutional neural network) layer is a network structure that performs well in processing time series data. Through a series of convolution operations, it can effectively extract local features from the data, which helps to understand short-term patterns and instantaneous changes in the data. The Bi-LSTM (bidirectional long short-term memory network) layer can process the front-end dependencies of the data and identify complex patterns and long-term trends across multiple time points. By combining the two network structures, the temperature time series correlation characteristics of temperature-sensitive points at different time scales can be fully extracted, providing rich feature information for subsequent anomaly identification.

[0050] In the above-mentioned booster station intelligent inspection system 100 based on the Internet of Things, the multi-scale feature balanced fusion module 140 is used to perform adaptive balanced fusion of the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector in the background booster station intelligent inspection server to obtain a multi-time scale temperature time series associated feature vector. In a specific example of the present application, the processing method of adaptively balancing and fusion of the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector is to input the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector into the balanced threshold feature vector adaptive fusion module to obtain the multi-time scale temperature time series associated feature vector. It should be understood that, considering that the temperature time series associated feature vectors of different time scales have different emphases and importance when describing the temperature changes of temperature sensitive points, direct splicing or addition may ignore some key information. Therefore, in the technical solution of the present application, the balanced threshold feature vector adaptive fusion module is used to fuse the temperature time series associated feature vectors of two time scales. Specifically, the equalization threshold feature vector adaptive fusion module performs multi-dimensional correlation feature learning on the first time scale temperature time series correlation feature vector and the second time scale temperature time series correlation feature vector to explore the intrinsic connection and complementarity between the two, thereby adaptively adjusting the feature fusion weights of the first time scale temperature time series correlation feature vector and the second time scale temperature time series correlation feature vector to fully retain and utilize the temperature time series correlation features at different time scales, reduce dependence on a single time scale feature, more comprehensively reflect the temperature change characteristics of temperature sensitive points, and improve the accuracy and reliability of temperature anomaly identification.

[0051] Figure 3 FIG. 1 is a block diagram of a multi-scale feature balance fusion module in a booster station intelligent inspection system based on the Internet of Things according to an embodiment of the present application. Figure 3 As shown, the multi-scale feature balanced fusion module 140 includes:

[0052] A multi-dimensional fusion unit 141 is used to obtain a first multi-time scale temperature time series fusion feature vector, a second multi-time scale temperature time series fusion feature vector and a third multi-time scale temperature time series fusion feature vector by passing the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector through a multi-dimensional fusion module;

[0053] A balanced threshold value calculation unit 142 is used to calculate the balanced threshold values ​​of the first multi-time scale temperature time series fusion feature vector, the second multi-time scale temperature time series fusion feature vector and the third multi-time scale temperature time series fusion feature vector to obtain a first balanced threshold value, a second balanced threshold value and a third balanced threshold value respectively;

[0054] The equalization fusion unit 143 is used to perform weighted fusion on the first time scale temperature time series association feature vector and the second time scale temperature time series association feature vector based on the first equalization threshold value, the second equalization threshold value and the third equalization threshold value to obtain the multi-time scale temperature time series association feature vector.

[0055] Figure 4 FIG. 1 is a block diagram of a multi-dimensional fusion unit in a booster station intelligent inspection system based on the Internet of Things according to an embodiment of the present application. Figure 4 As shown, the multi-dimensional fusion unit 141 includes:

[0056] A cascade fusion subunit 1411, configured to cascade the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector to obtain the first multi-time scale temperature time series fusion feature vector;

[0057] An element addition fusion subunit 1412 is used to add the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector by position to obtain the second multi-time scale temperature time series fusion feature vector;

[0058] The element point multiplication fusion subunit 1413 is used to perform position point multiplication on the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector to obtain the third multi-time scale temperature time series fusion feature vector.

[0059] Specifically, the equalization threshold value calculation unit 142 is used to: multiply the first multi-time-scale temperature series fusion feature vector by a first predetermined transformation vector to obtain a first threshold scoring coefficient; add the first threshold scoring coefficient and the first bias parameter and then pass the sigmoid activation function to obtain the first equalization threshold value.

[0060] Figure 5 FIG. 1 is a block diagram of a balanced fusion unit in a booster station intelligent inspection system based on the Internet of Things according to an embodiment of the present application. Figure 5 As shown, the equalization fusion unit 143 includes:

[0061] a weight parameter calculation subunit 1431, configured to determine a first weight parameter and a second weight parameter based on the first equalization threshold value, the second equalization threshold value, and the third equalization threshold value, wherein the first weight parameter is an average of the first equalization threshold value, the second equalization threshold value, and the third equalization threshold value, and the second weight parameter is a difference between one and the first weight parameter;

[0062] The position-by-position weighting subunit 1432 is used to weight the first time-scale temperature time series association feature vector position-by-position with the first weight parameter to obtain a weighted first time-scale temperature time series association feature vector, and to weight the second time-scale temperature time series association feature vector position-by-position with the second weight parameter to obtain a weighted second time-scale temperature time series association feature vector;

[0063] The fusion subunit 1433 is used to add the corresponding elements of the weighted first time scale temperature time series association feature vector and the weighted second time scale temperature time series association feature vector to obtain the multi-time scale temperature time series association feature vector.

[0064] That is, the multi-scale feature equalization fusion module 140 is used to process the first time scale temperature time series correlation feature vector and the second time scale temperature time series correlation feature vector using the following adaptive fusion formula to obtain the multi-time scale temperature time series correlation feature vector, wherein the adaptive fusion formula is:

[0065]

[0066] Wherein, v1 is the temperature time series associated feature vector of the first time scale, v2 is the temperature time series associated feature vector of the second time scale, W1, W2 and W3 are respectively the first predetermined transformation vector, the second predetermined transformation vector and the third predetermined transformation vector, b1, b2 and b3 are respectively the first bias parameter, the second bias parameter and the third bias parameter; sigmoid represents the activation function, concat(·,·) represents the cascade processing, represents positional addition, ⊙ represents positional point multiplication, t1, t2 and t3 are the first, second and third equalization thresholds, respectively, and t1, t2, t3∈[0, 1], v c is the multi-time-scale temperature series correlation feature vector.

[0067] In the above-mentioned booster station intelligent inspection system 100 based on the Internet of Things, the temperature monitoring result generation module 150 is used to determine the monitoring result based on the multi-time scale temperature time series associated feature vector in the background booster station intelligent inspection server. In a specific example of the present application, the processing method for determining the monitoring result based on the multi-time scale temperature time series associated feature vector is to input the multi-time scale temperature time series associated feature vector into a temperature monitor based on a classifier to obtain the monitoring result, and the monitoring result is used to indicate whether there is an abnormal fluctuation in the temperature of the temperature sensitive point. That is, the multi-time scale temperature time series associated feature vector is finally identified as abnormal by using a temperature monitor based on a classifier. Specifically, the classifier-based temperature monitor can effectively distinguish the normal temperature fluctuation mode and the abnormal temperature fluctuation mode of the temperature sensitive point by using a large amount of training data for training optimization, and identify the temperature fluctuation information in the multi-time scale temperature time series associated feature vector, thereby outputting the corresponding monitoring result. In this way, when the monitoring result shows that the temperature of the temperature sensitive point has abnormal fluctuations, an early warning signal can be automatically issued to remind operators and managers to deal with it in time, thereby avoiding equipment failure and loss.

[0068] In the technical solution of the present application, the first time scale temperature time series association feature vector and the second time scale temperature time series association feature vector respectively represent the local temperature time series association features with different time series association scales determined based on one-dimensional convolutional coding and bidirectional long short-term memory coding of the time series of the real-time temperature. Taking into account the difference in time series association scales between the two, when the first time scale temperature time series association feature vector and the second time scale temperature time series association feature vector are input into the equalization threshold feature vector adaptive fusion module, the multi-time scale temperature time series association feature vector obtained will have the problem of insufficient aggregation of temperature different time series association scale adaptive fusion distribution, thereby affecting the classification convergence efficiency through the classifier, that is, affecting the efficiency of classification training and the accuracy of classification results.

[0069] Based on this, in a preferred example of the present application, the multi-time-scale temperature time series associated feature vector is input into a classifier-based temperature monitor to obtain a monitoring result, including: multiplying the multi-time-scale temperature time series associated feature vector by the length of the multi-time-scale temperature time series associated feature vector and the square root of the length of the multi-time-scale temperature time series associated feature vector to obtain a multi-time-scale temperature time series associated full-width representation vector and a multi-time-scale temperature time series associated half-width representation vector; performing point subtraction on the sum of the norms of the multi-time-scale temperature time series associated full-width representation vector and the multi-time-scale temperature time series associated feature vector, and calculating the square root of the absolute value of each position of the point subtraction result vector to obtain a multi-time-scale temperature time series associated full-width semantic change vector; performing point subtraction on the multi-time-scale temperature time series associated half-width representation vector and the multi-time-scale temperature time series associated feature vector. The binary norm of the vector is subtracted, and the square root of the absolute value of each position of the point subtraction result vector is calculated to obtain a multi-time-scale temperature time series associated half-width semantic change vector; the logarithm with base 2 of each eigenvalue of the multi-time-scale temperature time series associated full-width semantic change vector and the multi-time-scale temperature time series associated half-width semantic change vector are calculated respectively to obtain a multi-time-scale temperature time series associated full-width semantic change information vector and a multi-time-scale temperature time series associated half-width semantic change information vector; the weighted sum of the multi-time-scale temperature time series associated full-width semantic change information vector and the multi-time-scale temperature time series associated half-width semantic change information vector is calculated with a balanced hyperparameter as a weight to obtain an optimized multi-time-scale temperature time series associated feature vector; the optimized multi-time-scale temperature time series associated feature vector is input into the classifier-based temperature monitor to obtain the monitoring result.

[0070] Here, the multi-time-scale temperature time series associated feature vector is taken as a feature set, wherein the change semantics in units of the eigenvalues ​​of the multi-time-scale temperature time series associated feature vector are expressed, in order to dynamically aggregate the semantic set as a whole composed of different change semantics of the multi-time-scale temperature time series associated feature vector without ignoring the individual semantic change information, the individual features of the multi-time-scale temperature time series associated feature vector and the collective expression of the aggregation scale of the multi-time-scale temperature time series associated feature vector are taken as the feature full amplitude and half amplitude, and the low-rank negative correlation of different dimensions of the overall semantics of the feature set of the multi-time-scale temperature time series associated feature vector is taken as the phase and scaled, so as to dynamically adjust the change relationship of the semantic content of the multi-time-scale temperature time series associated feature vector, thereby improving the aggregation of the semantic information expression of the overall feature set of the multi-time-scale temperature time series associated feature vector, and improving the classification convergence efficiency of the multi-time-scale temperature time series associated feature vector when classified by a classifier-based temperature monitor.

[0071] In summary, the IoT-based intelligent inspection system for booster stations according to the embodiment of the present application is explained, which uses artificial intelligence technology based on deep learning to perform real-time temperature monitoring and data analysis on temperature-sensitive points inside the booster station, so as to capture the temperature time series correlation characteristics of different time scales, and intelligently identify abnormal temperature fluctuations of temperature-sensitive points based on the adaptive fusion characteristics of multi-time scale temperature information. In this way, real-time monitoring and accurate early warning of temperature-sensitive points inside the booster station can be achieved, effectively reducing false alarms and missed alarms, and improving the intelligent operation and maintenance level of the booster station.

[0072] Figure 6 FIG. 1 is a flow chart of a method for intelligent inspection of a booster station based on the Internet of Things according to an embodiment of the present application. Figure 6 As shown, according to the embodiment of the present application, the intelligent inspection method of the booster station based on the Internet of Things includes the following steps:

[0073] Step S1, collecting the time series of the real-time temperature of the temperature sensitive point through the temperature sensor of the Internet of Things;

[0074] Step S2, transmitting the time series of the real-time temperature to the background booster station intelligent inspection server through the wireless communication network of the Internet of Things;

[0075] Step S3, in the background booster station intelligent inspection server, performing multi-scale temperature time series feature extraction on the time series of the real-time temperature to obtain a first time scale temperature time series associated feature vector and a second time scale temperature time series associated feature vector;

[0076] Step S4, in the background booster station intelligent inspection server, adaptively balancing and fusing the first time scale temperature time series correlation feature vector and the second time scale temperature time series correlation feature vector to obtain a multi-time scale temperature time series correlation feature vector;

[0077] Step S5: In the background booster station intelligent inspection server, a monitoring result is determined based on the multi-time-scale temperature time series correlation feature vector.

[0078] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned booster station intelligent inspection method based on the Internet of Things have been referred to above. Figures 1 to 5 The description of the IoT-based substation intelligent inspection system has been introduced in detail, and therefore, its repeated description will be omitted.

[0079] The basic principle of the present invention is described above in conjunction with specific embodiments. However, it should be pointed out that the advantages, strengths, effects, etc. mentioned in the present invention are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present invention. In addition, the specific details of the above embodiments are only for the purpose of illustration and facilitation of understanding, rather than limitation, and the above details do not limit the present invention to being implemented by adopting the above specific details.

[0080] In the above embodiments, the description of each embodiment has its own emphasis. For the parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is only schematic. For example, the unit division is only a logical function division, and there may be other division methods in actual implementation. The unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical sub-unit, that is, it may be located in one place, or it may be distributed on multiple network sub-units. Some or all of the units can be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0081] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference to a figure in a claim should not be considered as limiting the claim to which it relates.

[0082] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. An intelligent inspection system for booster stations based on the Internet of Things, characterized in that: include: The temperature monitoring module is used to collect the real-time temperature time series of temperature-sensitive points through the temperature sensor of the Internet of Things; A data transmission module, used to transmit the time series of the real-time temperature to a background booster station intelligent inspection server through the wireless communication network of the Internet of Things; A multi-scale temperature time series feature extraction module is used to extract multi-scale temperature time series features on the time series of the real-time temperature in the background booster station intelligent inspection server to obtain a first time scale temperature time series associated feature vector and a second time scale temperature time series associated feature vector; A multi-scale feature balanced fusion module is used to perform adaptive balanced fusion on the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector in the background booster station intelligent inspection server to obtain a multi-time scale temperature time series associated feature vector; The temperature monitoring result generation module is used to determine the monitoring result based on the multi-time scale temperature time series correlation feature vector in the background booster station intelligent inspection server.

2. According to the Internet of Things-based booster station intelligent inspection system of claim 1, it is characterized in that: The multi-scale temperature time series feature extraction module is used to: The time series of the real-time temperature is input into a temperature time series feature extraction dual-stream network including a 1DCNN layer and a Bi-LSTM layer to obtain the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector.

3. The IoT-based booster station intelligent inspection system according to claim 2 is characterized in that: The multi-scale feature balanced fusion module is used to: The first time scale temperature time series correlation feature vector and the second time scale temperature time series correlation feature vector are input into the equalization threshold feature vector adaptive fusion module to obtain the multi-time scale temperature time series correlation feature vector.

4. The IoT-based booster station intelligent inspection system according to claim 3 is characterized in that: The multi-scale feature balanced fusion module includes: A multi-dimensional fusion unit, used for fusing the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector through a multi-dimensional fusion module to obtain a first multi-time scale temperature time series fusion feature vector, a second multi-time scale temperature time series fusion feature vector and a third multi-time scale temperature time series fusion feature vector; a balance threshold value calculation unit, used to calculate the balance threshold values ​​of the first multi-time scale temperature time series fusion feature vector, the second multi-time scale temperature time series fusion feature vector and the third multi-time scale temperature time series fusion feature vector respectively to obtain a first balance threshold value, a second balance threshold value and a third balance threshold value; The equalization fusion unit is used to perform weighted fusion on the first time scale temperature time series association feature vector and the second time scale temperature time series association feature vector based on the first equalization threshold value, the second equalization threshold value and the third equalization threshold value to obtain the multi-time scale temperature time series association feature vector.

5. The IoT-based booster station intelligent inspection system according to claim 4 is characterized in that: The multi-dimensional fusion unit comprises: A cascade fusion subunit, configured to cascade the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector to obtain the first multi-time scale temperature time series fusion feature vector; An element addition fusion subunit, used for adding the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector by position to obtain the second multi-time scale temperature time series fusion feature vector; The element point multiplication fusion subunit is used to perform position point multiplication on the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector to obtain the third multi-time scale temperature time series fusion feature vector.

6. The IoT-based intelligent inspection system for booster stations according to claim 5 is characterized in that: The equalization threshold value calculation unit is used to: Multiplying the first multi-time-scale temperature time series fusion feature vector by a first predetermined transformation vector to obtain a first threshold scoring coefficient; The first threshold scoring coefficient and the first bias parameter are added together and then passed through a sigmoid activation function to obtain the first equalization threshold value.

7. The IoT-based intelligent inspection system for booster stations according to claim 6 is characterized in that: The equalization fusion unit comprises: a weight parameter calculation subunit, configured to determine a first weight parameter and a second weight parameter based on the first equalization threshold value, the second equalization threshold value, and the third equalization threshold value, wherein the first weight parameter is an average of the first equalization threshold value, the second equalization threshold value, and the third equalization threshold value, and the second weight parameter is a difference between one and the first weight parameter; a position-by-position weighting subunit, configured to weight the first time-scale temperature time series associated feature vector position-by-position with the first weight parameter to obtain a weighted first time-scale temperature time series associated feature vector, and to weight the second time-scale temperature time series associated feature vector position-by-position with the second weight parameter to obtain a weighted second time-scale temperature time series associated feature vector; The fusion subunit is used to add the weighted first time scale temperature time series association feature vector and the weighted second time scale temperature time series association feature vector by corresponding elements to obtain the multi-time scale temperature time series association feature vector.

8. The IoT-based booster station intelligent inspection system according to claim 7 is characterized in that: The temperature monitoring result generating module is used for: The multi-time-scale temperature time series correlation feature vector is input into a classifier-based temperature monitor to obtain the monitoring result, and the monitoring result is used to indicate whether there is abnormal temperature fluctuation at the temperature sensitive point.

9. A method for intelligent inspection of booster stations based on the Internet of Things, characterized in that: include: The temperature sensor of the Internet of Things collects the time series of the real-time temperature of the temperature-sensitive point; The time series of the real-time temperature is transmitted to the background booster station intelligent inspection server through the wireless communication network of the Internet of Things; In the background booster station intelligent inspection server, multi-scale temperature time series feature extraction is performed on the time series of the real-time temperature to obtain a first time scale temperature time series associated feature vector and a second time scale temperature time series associated feature vector; In the background booster station intelligent inspection server, the first time scale temperature time series associated feature vector and the second time scale temperature time series associated feature vector are adaptively balanced and fused to obtain a multi-time scale temperature time series associated feature vector; In the background booster station intelligent inspection server, a monitoring result is determined based on the multi-time-scale temperature time series correlation feature vector.

10. The method for intelligent inspection of booster stations based on the Internet of Things according to claim 9 is characterized in that: In the background booster station intelligent inspection server, adaptively balancing and fusing the first time scale temperature time series correlation feature vector and the second time scale temperature time series correlation feature vector to obtain a multi-time scale temperature time series correlation feature vector, including: The first time scale temperature time series correlation feature vector and the second time scale temperature time series correlation feature vector are input into the equalization threshold feature vector adaptive fusion module to obtain the multi-time scale temperature time series correlation feature vector.

Images