Power equipment state monitoring sensor and monitoring method thereof

Abstract

The invention discloses a power equipment state monitoring sensor and a monitoring method thereof, and solves the problem that the reliability and the accuracy of a partial discharge sensor are insufficient due to the fact that clutter interference cannot be effectively eliminated when the partial discharge sensor of an existing device collects signals, the power equipment state monitoring sensor comprises a data collection main body, a state database, a state evaluation module and a result pushing module, the data acquisition main body comprises a sensor shell, a signal detection part, a detection adaptation part and an anti-interference assembly, the signal detection part is sleeved with the anti-interference assembly, and the anti-interference assembly is used for protecting the signal detection part and assisting in shielding interference clutters; according to the invention, the data acquisition main body and the state evaluation module are arranged, the data acquisition main body can effectively eliminate clutter interference when acquiring the real-time state signal of the power equipment, so that the accuracy of the signal is ensured, and the state evaluation module can analyze and evaluate the real-time state data by pre-constructing a state evaluation model, so that the accuracy of the real-time state data is improved. And an accurate equipment state evaluation result is obtained.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power equipment monitoring, and in particular relates to an electric power equipment state monitoring sensor and a monitoring method thereof. Background Art

[0002] Power equipment is the most important core part of the power grid. The power transmission and transformation side of power equipment includes: high-voltage transmission lines; large power transformers; control equipment, etc. In order to enhance the safety and reliability of power equipment operation, timely diagnosis of faults, and extension of equipment life, at this stage, sensor technology is generally used to collect on-site information, combined with artificial intelligence, to complete the intelligent online monitoring of power equipment. The power equipment monitoring sensor is a detection device used to monitor and control various parameters in the operation of power equipment, so as to enable the power equipment to operate in a normal or optimal state.

[0003] Chinese patent CN112363100B discloses a fault detection device for a condition monitoring sensor, which includes a partial discharge sensor, which is arranged on a capacitive device to detect the state of the capacitive device; a fault detection module, which is connected to the partial discharge sensor through a secondary circuit to monitor the working state of the partial discharge sensor; a mounting component, which is arranged on the outside of the partial discharge sensor and has a fixed state for fixing the partial discharge sensor; however, the partial discharge sensor of the existing device cannot effectively eliminate clutter interference when collecting signals, so that the reliability and accuracy of the partial discharge sensor are insufficient. In view of the above problems, we propose a power equipment condition monitoring sensor and a monitoring method thereof. Summary of the invention

[0004] The purpose of the present invention is to provide an electric power equipment status monitoring sensor and a monitoring method thereof in view of the deficiencies in the prior art, thereby solving the problem that the local discharge sensor of the existing device cannot effectively eliminate clutter interference when collecting signals, resulting in insufficient reliability and accuracy of the local discharge sensor.

[0005] The existing partial discharge sensors cannot effectively eliminate clutter interference when collecting signals, which makes the reliability and accuracy of the partial discharge sensors insufficient. In view of the above problems, we propose a power equipment status monitoring sensor and a monitoring method thereof. In short, the power equipment status monitoring sensor includes a data acquisition subject, a status database, a status evaluation module, and a result push module. In the embodiment of the present invention, a data acquisition subject and a status evaluation module are provided. The data acquisition subject can effectively eliminate clutter interference when collecting real-time status signals of power equipment, thereby ensuring the accuracy of the signal. The status evaluation module can analyze and evaluate the real-time status data by pre-building a status evaluation model, and obtain accurate equipment status evaluation results, thereby overcoming the problem that the existing partial discharge sensors cannot effectively eliminate clutter interference when collecting signals, which makes the reliability and accuracy of the partial discharge sensors insufficient.

[0006] The present invention is implemented in this way: a power equipment state monitoring sensor, wherein the power equipment state monitoring sensor comprises:

[0007] A data acquisition subject, wherein the data acquisition subject is used to collect real-time status signals of power equipment in real time, amplify and convert the real-time status signals to obtain real-time status data, and upload the real-time status data to a status database;

[0008] A state database is used to normalize and pre-process the real-time state data and store the normalized real-time state data;

[0009] The state assessment module is connected to the state database and the data collection subject through communication. The state assessment module pre-builds a state assessment model, traverses the real-time state data, uses the state assessment model to analyze and evaluate the real-time state data, and outputs the equipment state assessment result;

[0010] The result push module is used to obtain the device status evaluation results and push the device status evaluation results to the user end in real time based on the local area network.

[0011] Preferably, the status database includes:

[0012] A data cleaning unit, used to obtain real-time status data and clean the real-time status data;

[0013] A data dimension reduction unit is used to obtain the real-time status data after data cleaning and perform dimension reduction and normalization processing on the real-time status data;

[0014] Among them, when the real-time status data is reduced in dimension and normalized, the real-time status data is processed based on the principal component analysis algorithm;

[0015] A data fusion unit, which obtains the normalized real-time status data, performs data fusion on the real-time status data, and uploads the processed real-time status data to the data storage unit;

[0016] The data storage unit is used to store the normalized real-time status data, and the data storage unit is electrically connected to the data cleaning unit, the data dimension reduction unit, and the data fusion unit respectively.

[0017] Preferably, the method for processing real-time status data based on the principal component analysis algorithm specifically includes:

[0018] Input multi-dimensional real-time status data, map the real-time status data to a low-dimensional space through the feature mapping function, and obtain low-dimensional status data, which is calculated by the following formula:

[0019]

[0020] Among them, x jis the real-time status data of the input j dimension, x i is the output low-dimensional state data, θ i is the inspiration function of the feature mapping function, i is the dimension of the output data, is the mean of the real-time status data, and α is the information weight corresponding to the real-time status data;

[0021]

[0022] Get low-dimensional state data, set the low-dimensional variance to 1, and obtain a low-dimensional variance set;

[0023] Among them, the variance of low-dimensional state data is calculated by the following formula;

[0024]

[0025] Load low-dimensional state data and low-dimensional variance set, calculate the covariance matrix, and perform eigendecomposition on the covariance matrix to obtain the eigendecomposition set;

[0026] Based on the principal component analysis algorithm, the l features with the greatest significance are calculated, and the l features and the corresponding feature vectors are retained to achieve dimensionality reduction and normalization processing of real-time status data.

[0027] Preferably, the status assessment module comprises:

[0028] The model building unit pre-builds the state assessment model by combining expert experience and the selective state space model, and iteratively trains the state assessment model;

[0029] The construction method of the state assessment model specifically includes:

[0030] Taking the selective state space model as the initial model, initializing the state transfer matrix and discretization parameters of the selective state space model;

[0031] Introducing a hybrid expert mechanism into the initial model, and replacing the input selection mechanism in the initial model with the hybrid expert mechanism;

[0032] Traverse the real-time state data in the state database, extract the training sample set from the state database, and assign weights to the training sample set based on the hybrid expert mechanism;

[0033] Among them, the weighting formula for the training sample set is:

[0034]

[0035] Among them, ω i For the input training sample x i The corresponding weight value;

[0036] Iteratively train the initial model with the weighted training samples until the initial model converges;

[0037] Determine whether the state evaluation model meets the preset accuracy threshold. If it meets the accuracy threshold, output the trained state evaluation model;

[0038] Among them, the judgment function of the state assessment model is:

[0039]

[0040] Among them, argmax(y i ) is the state evaluation model's evaluation result on the training sample state, N is the number of training samples, and g is the coupling coefficient of the state evaluation model;

[0041] The model evaluation unit traverses the real-time status data, uses the status evaluation model to analyze and evaluate the real-time status data, and outputs the equipment status evaluation result.

[0042] Preferably, the data collection subject includes:

[0043] A sensor housing, wherein a sensor protection cavity is provided in the sensor housing, a front-end controller is detachably mounted on the top of the sensor housing, and a sensor bearing plate is fixedly mounted in the sensor protection cavity;

[0044] A signal detection unit, the signal detection unit is arranged on the sensor bearing plate, and the signal detection unit is used to collect and detect the vibration signal generated when the power equipment is working;

[0045] A detection adaptation part, which is arranged in the sensor bearing plate and is used to fix signal detection parts of different specifications;

[0046] An anti-interference component is sleeved on the outside of the signal detection unit, and is used to protect the signal detection unit and assist in shielding interference noise.

[0047] Preferably, the signal detection unit comprises:

[0048] A piezoelectric ceramic sheet, the piezoelectric ceramic sheet is used to detect vibration signals of power equipment, and the piezoelectric ceramic sheet is electrically connected to the front-end controller through a signal amplification circuit;

[0049] The ceramic bearing seat is fixedly connected to the piezoelectric ceramic sheet and is used for bearing the piezoelectric ceramic sheet.

[0050] Preferably, the detection adaptation unit comprises:

[0051] An adaptive motor, wherein the adaptive motor is fixedly embedded on the sensor bearing plate;

[0052] An inner limit plate fixedly connected to the output end of the adaptive motor, wherein at least one set of inner limit grooves is formed on the inner limit plate;

[0053] An outer limit plate detachably mounted on the sensor bearing plate, wherein at least one set of outer limit grooves is formed in the outer limit plate;

[0054] At least one set of adaptive linkage blocks, wherein the adaptive linkage blocks are respectively slidably connected to the inner limit groove and the outer limit groove;

[0055] An adaptive clamping roller fixedly connected to the adaptive linkage block, wherein the adaptive clamping roller is used to clamp and fix ceramic bearing seats of different specifications.

[0056] Preferably, the anti-interference component comprises:

[0057] An anti-interference seat, which is fixedly mounted on the end of the sensor housing and is slidably connected to the piezoelectric ceramic sheet, and a clutter absorption cavity is provided in the anti-interference seat;

[0058] At least one set of clutter absorbing members, wherein the clutter absorbing members are detachably mounted in the anti-interference seat;

[0059] Wherein, the clutter absorbing member comprises:

[0060] An absorption capacitor, wherein the absorption capacitor is fixedly installed in the anti-interference seat;

[0061] At least one group of clutter attenuation rings, on which absorption coils are wound.

[0062] Preferably, a monitoring and protection component is provided in the sensor protection cavity, the monitoring and protection component is connected to the sensor bearing plate, and the monitoring and protection component is used to protect the piezoelectric ceramic sheet;

[0063] Wherein, the monitoring and protection component includes:

[0064] At least one set of guide sleeves, wherein the guide sleeves are fixedly connected to the sensor bearing plate;

[0065] A guide shock-absorbing column slidably connected to the guide sleeve, the guide shock-absorbing column is fixedly mounted on the side wall of the sensor housing, an outer wall of the guide shock-absorbing column is sleeved with a shock-absorbing spring, one end of the shock-absorbing spring is fixedly connected to the end wall of the guide sleeve;

[0066] A lifting linkage block, wherein the lifting linkage block is fixedly mounted on the side wall of the guide sleeve, a reset shock absorbing seat is fixedly mounted on the lifting linkage block, and a reset shock absorbing groove is provided in the reset shock absorbing seat;

[0067] A reset linkage block slidably connected to the reset damping groove, the reset linkage block is slidably installed in a linkage guide seat, the linkage guide seat is fixedly connected to the side wall of the sensor housing, and the side wall of the reset linkage block is elastically connected with a driving ratchet;

[0068] A linkage gear meshes with the driving ratchet for transmission, one side of the linkage gear is connected with a reset torsion spring, and the reset torsion spring is fixedly installed on the inner wall of the sensor housing.

[0069] On the other hand, the present invention also provides a power equipment state monitoring sensor monitoring method, the power equipment state monitoring sensor monitoring method specifically comprising:

[0070] Collect the real-time status signal of the power equipment in real time, amplify and convert the real-time status signal to obtain the real-time status data, and upload the real-time status data to the status database;

[0071] Normalize and pre-process the real-time status data, and store the normalized real-time status data;

[0072] Pre-build a status assessment model, traverse the real-time status data, use the status assessment model to analyze and evaluate the real-time status data, and output the equipment status assessment results;

[0073] Obtain the device status assessment results and push them to the user end in real time based on the local area network.

[0074] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0075] In the embodiment of the present invention, a data acquisition body and a status evaluation module are provided. The data acquisition body can effectively eliminate clutter interference when collecting real-time status signals of power equipment, thereby ensuring the accuracy of the signal, and the status evaluation module can analyze and evaluate the real-time status data by pre-building a status evaluation model to obtain accurate equipment status evaluation results.

[0076] In the embodiment of the present invention, the real-time status data is processed by using the principal component analysis algorithm, so that the eigenvectors with high significance in the real-time status data can be extracted, which facilitates the construction and training of the status assessment model and ensures the assessment accuracy of the status assessment model.

[0077] In the embodiment of the present invention, an anti-interference component is provided. The anti-interference component can protect the signal detection unit while effectively shielding and eliminating interfering clutter, thereby ensuring that the signal detection unit obtains an accurate vibration signal. Through effective clutter suppression technology, the signal-to-noise ratio can be significantly improved, ensuring that the extracted vibration signal is purer and more reliable.

[0078] In an embodiment of the present invention, a detection adaptation part is provided, which is composed of an adaptive motor, an inner limit plate, an outer limit plate, an adaptive linkage block, and an adaptive clamping roller. The inner limit plate, the outer limit plate, the adaptive linkage block, and the adaptive clamping roller cooperate with each other to achieve clamping and fixing of piezoelectric ceramic sheets of different specifications and models, thereby ensuring the stable operation of the piezoelectric ceramic sheets.

[0079] In an embodiment of the present invention, a monitoring and protection component is provided, which is composed of a guide shock-absorbing column, a lifting linkage block, a reset linkage block, a driving ratchet, and a linkage gear. The guide shock-absorbing column, the lifting linkage block, the reset linkage block, the driving ratchet, and the linkage gear cooperate to form a dynamic shock-absorbing protection system, thereby avoiding damage to the piezoelectric ceramic piece when the external vibration frequency is too large, thereby further ensuring the normal and stable operation of the piezoelectric ceramic piece. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 It is a structural schematic diagram of the power equipment status monitoring sensor provided by the present invention;

[0081] Figure 2 It is a schematic diagram of the implementation flow of real-time status data processing based on the principal component analysis algorithm provided by the present invention;

[0082] Figure 3 It is a schematic diagram of the implementation flow of the state assessment model construction method provided by the present invention;

[0083] Figure 4 It is a structural schematic diagram of the data acquisition subject provided by the present invention;

[0084] Figure 5 It is a three-dimensional structural schematic diagram of the data acquisition subject provided by the present invention;

[0085] Figure 6 It is a top view of the data acquisition body provided by the present invention;

[0086] Figure 7 yes Figure 6 AA section view;

[0087] Figure 8 is a schematic diagram of the structure of the detection adaptation unit provided by the present invention;

[0088] Fig. 9 is a front view of the detection adaptation part provided by the present invention;

[0089] Fig.10 is a bottom view of the detection adaptation portion provided by the present invention;

[0090] Fig.11 is a schematic diagram of the structure of the anti-interference component provided by the present invention;

[0091] Fig.12 is a schematic diagram of the three-dimensional structure of the anti-interference component provided by the present invention;

[0092] Fig.13 It is a schematic diagram of the internal structure of the sensor protection cavity provided by the present invention;

[0093] Fig.14 It is a structural schematic diagram of the monitoring and protection component provided by the present invention;

[0094] Fig.15 is an axonometric diagram of the monitoring and protection assembly provided by the present invention;

[0095] Fig.16 It is a schematic diagram of the implementation flow of the power equipment status monitoring sensor monitoring method provided by the present invention.

[0096] In the figure: 10-data acquisition body, 11-sensor housing, 111-sensor protection chamber, 12-front-end controller, 13-sensor bearing plate, 14-monitoring protection component, 141-guide sleeve, 142-guide shock-absorbing column, 143-shock-absorbing spring, 144-lifting linkage block, 145-reset shock-absorbing seat, 146-linkage guide seat, 147-reset linkage block, 148-drive ratchet, 149-linkage gear, 15-detection adaptation part, 151-adaptive motor, 152-inner limit plate, 153-outer limit plate, 154-inner limit slot, 155-outer limit slot, 156-adaptive linkage Moving block, 157-adaptive clamping roller, 16-reset torsion spring, 2-anti-interference component, 21-anti-interference seat, 22-clutter absorber, 221-absorption capacitor, 222-clutter weakening ring, 223-absorption coil, 23-clutter absorption cavity, 3-signal detection unit, 31-piezoelectric ceramic sheet, 32-ceramic bearing seat, 200-state database, 210-data cleaning unit, 220-data dimension reduction unit, 230-data fusion unit, 240-data storage unit, 300-state evaluation module, 310-model construction unit, 320-model evaluation unit, 400-result push module. DETAILED DESCRIPTION

[0097] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of this application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0098] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0099] The existing partial discharge sensors cannot effectively eliminate clutter interference when collecting signals, which makes the reliability and accuracy of the partial discharge sensors insufficient. In view of the above problems, we propose a power equipment status monitoring sensor and a monitoring method thereof. In short, the power equipment status monitoring sensor includes a data acquisition body 10, a status database 200, a status evaluation module 300, and a result push module 400. In the embodiment of the present invention, a data acquisition body 10 and a status evaluation module 300 are provided. The data acquisition body 10 can effectively eliminate clutter interference when collecting real-time status signals of power equipment, thereby ensuring the accuracy of the signal, and the status evaluation module 300 can analyze and evaluate the real-time status data by pre-building a status evaluation model, and obtain accurate equipment status evaluation results, thereby overcoming the problem that the existing partial discharge sensors cannot effectively eliminate clutter interference when collecting signals, which makes the reliability and accuracy of the partial discharge sensors insufficient.

[0100] The embodiment of the present invention provides a sensor for monitoring the state of electric power equipment. Figure 1 The schematic diagram of the structure of the power equipment status monitoring sensor is shown, and the power equipment status monitoring sensor specifically includes:

[0101] The data acquisition body 10 is used to collect the real-time status signal of the power equipment in real time, amplify and convert the real-time status signal to obtain real-time status data, and upload the real-time status data to the status database 200;

[0102] The state database 200 is used to normalize and pre-process the real-time state data and store the normalized real-time state data;

[0103] The state evaluation module 300 is connected to the state database 200 and the data acquisition body 10 in communication. The state evaluation module 300 pre-builds a state evaluation model, traverses the real-time state data, uses the state evaluation model to analyze and evaluate the real-time state data, and outputs the device state evaluation result;

[0104] The result push module 400 is used to obtain the device status evaluation result and push the device status evaluation result to the user terminal in real time based on the local area network.

[0105] In the embodiment of the present invention, a data acquisition body 10 and a status evaluation module 300 are provided. The data acquisition body 10 can effectively eliminate clutter interference when collecting real-time status signals of power equipment, thereby ensuring the accuracy of the signal, and the status evaluation module 300 can analyze and evaluate the real-time status data by pre-building a status evaluation model to obtain accurate equipment status evaluation results.

[0106] In this embodiment, the state database 200 includes:

[0107] A data cleaning unit 210 is used to obtain real-time status data and clean the real-time status data;

[0108] It should be noted that when cleaning the real-time status data, deletion, replacement, and interpolation methods can be used to handle missing values, and duplicate items can be deleted and merged to remove duplicate values.

[0109] The data dimension reduction unit 220 is used to obtain the real-time status data after data cleaning, and perform dimension reduction and normalization processing on the real-time status data;

[0110] Among them, when the real-time status data is reduced in dimension and normalized, the real-time status data is processed based on the principal component analysis algorithm;

[0111] The data fusion unit 230 acquires the normalized real-time status data, performs data fusion on the real-time status data, and uploads the processed real-time status data to the data storage unit 240;

[0112] The data storage unit 240 is used to store the normalized real-time status data. The data storage unit 240 is electrically connected to the data cleaning unit 210 , the data dimension reduction unit 220 , and the data fusion unit 230 .

[0113] The embodiment of the present invention provides a method for processing real-time status data based on a principal component analysis algorithm. Figure 2 The figure shows a schematic diagram of the process flow of the method for processing real-time status data based on the principal component analysis algorithm. The method for processing real-time status data based on the principal component analysis algorithm specifically includes:

[0114] Step S101, input multi-dimensional real-time status data, map the real-time status data to a low-dimensional space through a feature mapping function, obtain low-dimensional status data, and calculate it through the following formula:

[0115]

[0116] Among them, x j is the real-time status data of the input j dimension, x i is the output low-dimensional state data, θ iis the heuristic function of the feature mapping function, i is the dimension of the output data, and in this embodiment, the dimension of the output data can be 2-3, is the mean of the real-time status data, and α is the information weight corresponding to the real-time status data;

[0117] It should be noted that real-time status data includes but is not limited to vibration signals, power equipment frequency, real-time temperature, and equipment voltage / current data, and power equipment includes but is not limited to thermal power generating sets, hydropower generating sets, wind power generating sets, solar photovoltaic generating sets, transmission lines, transformers, power generation switchgear, lightning arresters, and insulators.

[0118]

[0119] Step S102, obtaining low-dimensional state data, setting the low-dimensional variance to 1, and obtaining a low-dimensional variance set;

[0120] Among them, the variance of low-dimensional state data is calculated by the following formula;

[0121]

[0122] Step S103, loading low-dimensional state data and low-dimensional variance sets, calculating covariance matrices, and performing eigendecomposition on the covariance matrix to obtain eigendecomposition sets;

[0123] Step S104, based on the principal component analysis algorithm, calculate the l features with the greatest significance, retain the l features and the corresponding feature vectors, and realize the dimensionality reduction and normalization processing of the real-time status data.

[0124] In the embodiment of the present invention, the real-time status data is processed by using the principal component analysis algorithm, so that the eigenvectors with high significance in the real-time status data can be extracted, which facilitates the construction and training of the status assessment model and ensures the assessment accuracy of the status assessment model.

[0125] The embodiment of the present invention provides a state assessment module 300, Figure 1 The schematic diagram of the structure of the state evaluation module 300 is shown. The state evaluation module 300 specifically includes:

[0126] The model building unit 310 pre-builds a state assessment model based on expert experience and a selective state space model, and iteratively trains the state assessment model.

[0127] The model evaluation unit 320 traverses the real-time status data, analyzes and evaluates the real-time status data using the status evaluation model, and outputs the device status evaluation result.

[0128] In this embodiment, data interaction is achieved between the model building unit 310 and the model evaluation unit 320 by means of a DTU or 5G communication connection.

[0129] The embodiment of the present invention provides a method for constructing a state assessment model. Figure 3 The schematic diagram of the implementation flow of the construction method of the state assessment model is shown. The construction method of the state assessment model specifically includes:

[0130] Step S201, taking the selective state space model as the initial model, initializing the state transfer matrix and discretization parameters of the selective state space model;

[0131] It should be noted that the selective state space model is an advanced time series modeling tool that optimizes model performance and interpretability by introducing a variable selection mechanism. Selective state space models (SSMs) are developed based on traditional state space models. By introducing a variable selection mechanism, the model allows the model to automatically identify and retain variables that have a significant impact on the prediction results during the modeling process, while eliminating irrelevant or redundant information.

[0132] Step S202, introducing a hybrid expert mechanism into the initial model, and replacing the input selection mechanism in the initial model with the hybrid expert mechanism;

[0133] Step S203, traversing the real-time state data in the state database 200, extracting a training sample set from the state database 200, and assigning weights to the training sample set based on a hybrid expert mechanism;

[0134] In this embodiment, the selective state space model is combined with the hybrid expert mechanism to assign weights to the training sample set, thereby providing an effective tool for solving the prediction problem of complex dynamic systems.

[0135] Among them, the weighting formula for the training sample set is:

[0136]

[0137] Among them, ω i For the input training sample x i The corresponding weight value;

[0138] Step S204, iteratively training the initial model with the weighted training samples until the initial model converges;

[0139] Step S205, determining whether the state assessment model meets a preset accuracy threshold. In this embodiment, the preset accuracy threshold is set to 0.96-0.98;

[0140] Step S206: if the accuracy threshold is met, output the trained state evaluation model;

[0141] If it does not meet the accuracy threshold, return to step S203.

[0142] Among them, the judgment function of the state assessment model is:

[0143]

[0144] Among them, argmax(y i ) is the state evaluation model's evaluation result on the state of the training sample, N is the number of training samples, g is the coupling coefficient of the state evaluation model, and in this embodiment, the model coupling coefficient is set to 0.1-0.3.

[0145] In a further preferred embodiment of the present invention, Figure 4-Figure 7 As shown, the data collection body 10 includes:

[0146] A sensor housing 11, wherein a sensor protection cavity 111 is disposed in the sensor housing 11, a front-end controller 12 is detachably mounted on the top of the sensor housing 11, and a sensor bearing plate 13 is fixedly mounted in the sensor protection cavity 111;

[0147] In this embodiment, the sensor housing 11 is made of PVC or stainless steel, and the sensor housing 11 is a round seat or cylindrical structure with a hollow interior. The front-end controller 12 can be a PLC controller, and the front-end controller 12 is fixedly installed by fastening bolts or clamps. The sensor carrier plate 13 is fixedly embedded in the inner wall of the sensor protection cavity 111 by means of threads or snaps.

[0148] A signal detection unit 3, wherein the signal detection unit 3 is disposed on the sensor bearing plate 13, and the signal detection unit 3 is used to collect and detect vibration signals generated when the power equipment is working;

[0149] A detection adaptation part 15, wherein the detection adaptation part 15 is disposed in the sensor carrier plate 13, and the detection adaptation part 15 is used to fix signal detection parts 3 of different specifications;

[0150] The anti-interference component 2 is sleeved on the outside of the signal detection unit 3, and is used to protect the signal detection unit 3 and assist in shielding interference clutter.

[0151] In the embodiment of the present invention, an anti-interference component 2 is provided. The anti-interference component 2 can protect the signal detection unit 3 while effectively shielding and eliminating interference clutter, thereby ensuring that the signal detection unit 3 obtains an accurate vibration signal. Through effective clutter suppression technology, the signal-to-noise ratio can be significantly improved, ensuring that the extracted vibration signal is purer and more reliable.

[0152] In this embodiment, the signal detection unit 3 includes:

[0153] A piezoelectric ceramic sheet 31, the piezoelectric ceramic sheet 31 is used to detect a vibration signal of an electric power device, and the piezoelectric ceramic sheet 31 is electrically connected to the front-end controller 12 via a signal amplification circuit;

[0154] The ceramic support seat 32 is fixedly connected to the piezoelectric ceramic sheet 31 and is used for supporting the piezoelectric ceramic sheet 31 .

[0155] In this embodiment, the piezoelectric ceramic sheet 31 is made of modified lead zirconate titanate (PZT) and barium titanate materials, and the piezoelectric ceramic sheet 31 can be a circular, annular or disc-shaped structure. The piezoelectric ceramic sheet 31 and the ceramic support seat 32 are fixedly connected by welding or an integrated connection.

[0156] In a further preferred embodiment of the present invention, Figure 8-Figure 10 As shown, the detection adaptation unit 15 includes:

[0157] An adaptive motor 151, wherein the adaptive motor 151 is fixedly embedded on the sensor carrier board 13, and the adaptive motor 151 is electrically connected to the front-end controller 12, and the adaptive motor 151 can be a single-phase asynchronous motor;

[0158] An inner limit plate 152 fixedly connected to the output end of the adaptive motor 151, wherein at least one set of inner limit grooves 154 is formed on the inner limit plate 152;

[0159] An outer limiting plate 153 detachably mounted on the sensor bearing plate 13, wherein at least one set of outer limiting grooves 155 is formed in the outer limiting plate 153;

[0160] It should be noted that the inner limit plate 152 and the outer limit plate 153 are rotatably connected, and both the inner limit plate 152 and the outer limit plate 153 can be circular plates or round seat structures. The outer limit plate 153 is detachably connected to the sensor carrier plate 13 through at least one group of pillars, and one side of the inner limit plate 152 is fixedly connected to the output end of the adaptive motor 151 by a snap or a fastening bolt. The number of the inner limit grooves 154 and the outer limit grooves 155 remains consistent and can be 3-10 groups. The shape of the inner limit groove 154 can be a rectangular groove, an S-shaped groove or an arc groove.

[0161] At least one set of adaptive linkage blocks 156, wherein the adaptive linkage blocks 156 are slidably connected to the inner limit groove 154 and the outer limit groove 155 respectively;

[0162] The adaptive clamping roller 157 is fixedly connected to the adaptive linkage block 156, and the adaptive clamping roller 157 is used to clamp and fix ceramic bearing seats 32 of different specifications.

[0163] In this embodiment, the adaptive linkage block 156 may be a round block or a rectangular block structure, and the adaptive linkage block 156 is fixedly connected to the adaptive clamping roller 157 by mortise and tenon or riveting.

[0164] During operation, when it is necessary to clamp piezoelectric ceramic pieces 31 of different specifications and models, the adaptive motor 151 is turned on. The adaptive motor 151 is started to drive the inner limit plate 152 to rotate. The rotation of the inner limit plate 152 can drive the inner limit groove 154 to rotate. The rotation of the inner limit groove 154 drives the adaptive linkage block 156 to slide along the outer limit groove 155, and then the adaptive linkage block 156 drives the adaptive clamping roller 157 to move, thereby clamping and fixing piezoelectric ceramic pieces 31 of different specifications.

[0165] In the embodiment of the present invention, a detection adaptation part 15 is provided, and the detection adaptation part 15 is composed of an adaptive motor 151, an inner limit plate 152, an outer limit plate 153, an adaptive linkage block 156, and an adaptive clamping roller 157. The inner limit plate 152, the outer limit plate 153, the adaptive linkage block 156, and the adaptive clamping roller 157 cooperate with each other to achieve clamping and fixing of piezoelectric ceramic pieces 31 of different specifications and models, thereby ensuring the stable operation of the piezoelectric ceramic pieces 31.

[0166] In a further preferred embodiment of the present invention, Figure 11-Figure 12 As shown, the anti-interference component 2 includes:

[0167] An anti-interference seat 21, wherein the anti-interference seat 21 is fixedly mounted at the end of the sensor housing 11, and the anti-interference seat 21 is slidably connected to the piezoelectric ceramic sheet 31, and a clutter absorption cavity 23 is provided in the anti-interference seat 21;

[0168] At least one set of clutter absorbing members 22, wherein the clutter absorbing members 22 are detachably mounted in the anti-interference seat 21;

[0169] It should be noted that the anti-interference seat 21 is fixedly installed on the end of the sensor housing 11 by fastening bolts or welding, and the noise absorption cavity 23 is also filled with insulating shielding material, which can be a nylon mixed material or aluminum foil Mylar material. The noise absorption parts 22 are circumferentially arranged on the upper surface of the anti-interference seat 21.

[0170] Wherein, the clutter absorbing member 22 comprises:

[0171] An absorption capacitor 221, wherein the absorption capacitor 221 is fixedly installed in the anti-interference seat 21;

[0172] At least one group of clutter attenuation rings 222 , on which an absorption coil 223 is wound.

[0173] In this embodiment, the absorption capacitor 221 is fixedly installed in the anti-interference seat 21 by means of fastening bolts or snaps, the noise attenuation ring 222 is longitudinally arranged on the outside of the absorption capacitor 221, the absorption coil 223 is electrically connected to the absorption capacitor 221, and the noise attenuation ring 222 and the absorption coil 223 work together to absorb and weaken the noise.

[0174] In a further preferred embodiment of the present invention, Figure 13-Figure 15 As shown, a monitoring protection component 14 is disposed in the sensor protection cavity 111, and the monitoring protection component 14 is connected to the sensor bearing plate 13, and the monitoring protection component 14 is used to protect the piezoelectric ceramic sheet 31;

[0175] Wherein, the monitoring and protection component 14 includes:

[0176] At least one set of guide sleeves 141, the guide sleeves 141 are fixedly connected to the sensor bearing plate 13, the guide sleeves 141 are fixedly installed on the sensor bearing plate 13 in a circumferential direction, and the guide sleeves 141 are fixedly installed by plugging or riveting;

[0177] A guide damping column 142 slidably connected to the guide sleeve 141, the guide damping column 142 is fixedly mounted on the side wall of the sensor housing 11, and a damping spring 143 is sleeved on the outer wall of the guide damping column 142, and one end of the damping spring 143 is fixedly connected to the end wall of the guide sleeve 141;

[0178] It should be noted that the guide shock-absorbing column 142 is made of an elastic material, and the guide shock-absorbing column 142 can be a cylindrical or rectangular column structure.

[0179] A lifting linkage block 144, wherein the lifting linkage block 144 is fixedly mounted on the side wall of the guide sleeve 141, a reset damping seat 145 is fixedly mounted on the lifting linkage block 144, and a reset damping groove is provided in the reset damping seat 145;

[0180] A reset linkage block 147 slidably connected to the reset damping groove, the reset linkage block 147 is slidably installed in a linkage guide seat 146, the linkage guide seat 146 is fixedly connected to the side wall of the sensor housing 11, and the side wall of the reset linkage block 147 is elastically connected with a driving ratchet 148;

[0181] A linkage gear 149 meshes with the driving ratchet 148 for transmission, and a reset torsion spring 16 is connected to one side of the linkage gear 149 . The reset torsion spring 16 is fixedly mounted on the inner wall of the sensor housing 11 .

[0182] In this embodiment, one end of the linkage gear 149 away from the reset torsion spring 16 is also rotatably connected to the inner wall of the transmission housing through a bearing, and the reset torsion spring 16 is fixedly mounted on the inner wall of the sensor housing 11 by mortise and tenon or riveting.

[0183] In the embodiment of the present invention, a monitoring and protection component 14 is provided, and the monitoring and protection component 14 is composed of a guide shock-absorbing column 142, a lifting linkage block 144, a reset linkage block 147, a driving ratchet 148, and a linkage gear 149. The guide shock-absorbing column 142, the lifting linkage block 144, the reset linkage block 147, the driving ratchet 148, and the linkage gear 149 cooperate with each other to form a dynamic shock-absorbing protection system, thereby avoiding damage to the piezoelectric ceramic piece 31 when the external vibration frequency is too large, thereby further ensuring the normal and stable operation of the piezoelectric ceramic piece 31.

[0184] In a further preferred embodiment of the present invention, a method for monitoring a power equipment status monitoring sensor is also provided. Fig.16 The schematic diagram of the implementation process of the power equipment status monitoring sensor monitoring method is shown, and the power equipment status monitoring sensor monitoring method specifically includes:

[0185] Step S10, real-time acquisition of real-time status signals of power equipment, amplification and conversion of the real-time status signals to obtain real-time status data, and uploading the real-time status data to the status database 200;

[0186] Step S20, normalizing and preprocessing the real-time status data, and storing the normalized real-time status data;

[0187] Step S30, pre-constructing a status assessment model, traversing the real-time status data, using the status assessment model to analyze and assess the real-time status data, and outputting the device status assessment result;

[0188] Step S40, obtaining the device status evaluation result, and pushing the device status evaluation result to the user terminal in real time based on the local area network.

[0189] On the other hand, an embodiment of the present invention further provides a computer device, the computer device comprising a memory and a processor, the memory storing a computer program, and the computer program implementing any one of the above-mentioned methods when executed by the processor.

[0190] As a non-volatile computer-readable storage medium, the memory can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as the program instructions / modules corresponding to the power equipment status monitoring sensor monitoring method in the embodiment of the present application. The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the use of the power equipment status monitoring sensor monitoring method, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the local module via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0191] In summary, the present invention provides an electric power equipment state monitoring sensor and a monitoring method thereof. In an embodiment of the present invention, a data acquisition body 10 and a state evaluation module 300 are provided. The data acquisition body 10 can effectively eliminate clutter interference when collecting real-time state signals of electric power equipment, thereby ensuring the accuracy of the signal, and the state evaluation module 300 can analyze and evaluate the real-time state data by pre-building a state evaluation model to obtain accurate equipment state evaluation results.

[0192] It should be noted that, for the above-mentioned embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described order of actions, because according to the present invention, some steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0193] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be in the form of telecommunication or other forms.

[0194] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the scope of protection of the invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on these embodiments, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in this field can still combine, add, delete or make other adjustments to the features in the various embodiments of the present invention according to the circumstances without conflict, without making creative work, so as to obtain different other technical solutions that do not deviate from the concept of the present invention in essence, and these technical solutions also belong to the scope of protection of the present invention.

Claims

1. A power equipment status monitoring sensor, characterized in that: The power equipment status monitoring sensor comprises: A data acquisition subject, wherein the data acquisition subject is used to collect real-time status signals of power equipment in real time, amplify and convert the real-time status signals to obtain real-time status data, and upload the real-time status data to a status database; A state database is used to normalize and pre-process the real-time state data and store the normalized real-time state data; The state assessment module is connected to the state database and the data collection subject through communication. The state assessment module pre-builds a state assessment model, traverses the real-time state data, uses the state assessment model to analyze and evaluate the real-time state data, and outputs the equipment state assessment result; The result push module is used to obtain the device status evaluation results and push the device status evaluation results to the user end in real time based on the local area network.

2. The power equipment status monitoring sensor according to claim 1, characterized in that: The state database includes: A data cleaning unit, used to obtain real-time status data and clean the real-time status data; A data dimension reduction unit is used to obtain the real-time status data after data cleaning and perform dimension reduction and normalization processing on the real-time status data; Among them, when the real-time status data is reduced in dimension and normalized, the real-time status data is processed based on the principal component analysis algorithm; A data fusion unit, which obtains the normalized real-time status data, performs data fusion on the real-time status data, and uploads the processed real-time status data to the data storage unit; The data storage unit is used to store the normalized real-time status data, and the data storage unit is electrically connected to the data cleaning unit, the data dimension reduction unit, and the data fusion unit respectively.

3. The power equipment status monitoring sensor according to claim 2, characterized in that: The method for processing real-time status data based on the principal component analysis algorithm specifically includes: Input multi-dimensional real-time status data, map the real-time status data to a low-dimensional space through the feature mapping function, and obtain low-dimensional status data, which is calculated by the following formula: Among them, x j is the real-time status data of the input j dimension, x i is the output low-dimensional state data, θ i is the inspiration function of the feature mapping function, i is the dimension of the output data, is the mean of the real-time status data, and α is the information weight corresponding to the real-time status data; Get low-dimensional state data, set the low-dimensional variance to 1, and obtain a low-dimensional variance set; Among them, the variance of low-dimensional state data is calculated by the following formula; Load low-dimensional state data and low-dimensional variance set, calculate the covariance matrix, and perform eigendecomposition on the covariance matrix to obtain the eigendecomposition set; Based on the principal component analysis algorithm, the l features with the greatest significance are calculated, and the l features and the corresponding feature vectors are retained to achieve dimensionality reduction and normalization processing of real-time status data.

4. The power equipment status monitoring sensor according to claim 1, characterized in that: The status assessment module includes: The model building unit pre-builds the state assessment model by combining expert experience and the selective state space model, and iteratively trains the state assessment model; The construction method of the state assessment model specifically includes: Taking the selective state space model as the initial model, initializing the state transfer matrix and discretization parameters of the selective state space model; Introducing a hybrid expert mechanism into the initial model, and replacing the input selection mechanism in the initial model with the hybrid expert mechanism; Traverse the real-time state data in the state database, extract the training sample set from the state database, and assign weights to the training sample set based on the hybrid expert mechanism; Among them, the weighting formula for the training sample set is: Among them, ω i For the input training sample x i The corresponding weight value; Iteratively train the initial model with the weighted training samples until the initial model converges; Determine whether the state evaluation model meets the preset accuracy threshold. If it meets the accuracy threshold, output the trained state evaluation model; Among them, the judgment function of the state assessment model is: Among them, argmax(y i ) is the state evaluation model's evaluation result on the state of the training sample, N is the number of training samples, and g is the coupling coefficient of the state evaluation model; The model evaluation unit traverses the real-time status data, uses the status evaluation model to analyze and evaluate the real-time status data, and outputs the equipment status evaluation result.

5. The power equipment status monitoring sensor according to any one of claims 2 to 4, characterized in that: The data collection subjects include: A sensor housing, wherein a sensor protection cavity is provided in the sensor housing, a front-end controller is detachably mounted on the top of the sensor housing, and a sensor bearing plate is fixedly mounted in the sensor protection cavity; A signal detection unit, the signal detection unit is arranged on the sensor bearing plate, and the signal detection unit is used to collect and detect the vibration signal generated when the power equipment is working; A detection adaptation part, which is arranged in the sensor bearing plate and is used to fix signal detection parts of different specifications; An anti-interference component is sleeved on the outside of the signal detection unit, and is used to protect the signal detection unit and assist in shielding interference noise.

6. The power equipment status monitoring sensor according to claim 5, characterized in that: The signal detection unit comprises: A piezoelectric ceramic sheet, the piezoelectric ceramic sheet is used to detect vibration signals of power equipment, and the piezoelectric ceramic sheet is electrically connected to the front-end controller through a signal amplification circuit; The ceramic bearing seat is fixedly connected to the piezoelectric ceramic sheet and is used for bearing the piezoelectric ceramic sheet.

7. The power equipment status monitoring sensor according to claim 6, characterized in that: The detection adaptation unit comprises: An adaptive motor, wherein the adaptive motor is fixedly embedded on the sensor bearing plate; An inner limit plate fixedly connected to the output end of the adaptive motor, wherein at least one set of inner limit grooves is formed on the inner limit plate; An outer limit plate detachably mounted on the sensor bearing plate, wherein at least one set of outer limit grooves is formed in the outer limit plate; At least one set of adaptive linkage blocks, wherein the adaptive linkage blocks are respectively slidably connected to the inner limit groove and the outer limit groove; An adaptive clamping roller fixedly connected to the adaptive linkage block, wherein the adaptive clamping roller is used to clamp and fix ceramic bearing seats of different specifications.

8. The power equipment status monitoring sensor according to claim 7, characterized in that: The anti-interference component comprises: An anti-interference seat, which is fixedly mounted on the end of the sensor housing and is slidably connected to the piezoelectric ceramic sheet, and a clutter absorption cavity is provided in the anti-interference seat; At least one set of clutter absorbing members, wherein the clutter absorbing members are detachably mounted in the anti-interference seat; Wherein, the clutter absorbing member comprises: An absorption capacitor, wherein the absorption capacitor is fixedly installed in the anti-interference seat; At least one group of clutter attenuation rings, on which absorption coils are wound.

9. The power equipment status monitoring sensor according to claim 7, characterized in that: A monitoring and protection component is arranged in the sensor protection cavity, the monitoring and protection component is connected to the sensor bearing plate, and the monitoring and protection component is used to protect the piezoelectric ceramic sheet; Wherein, the monitoring and protection component includes: At least one set of guide sleeves, wherein the guide sleeves are fixedly connected to the sensor bearing plate; A guide shock-absorbing column slidably connected to the guide sleeve, the guide shock-absorbing column is fixedly mounted on the side wall of the sensor housing, an outer wall of the guide shock-absorbing column is sleeved with a shock-absorbing spring, one end of the shock-absorbing spring is fixedly connected to the end wall of the guide sleeve; A lifting linkage block, wherein the lifting linkage block is fixedly mounted on the side wall of the guide sleeve, a reset shock absorbing seat is fixedly mounted on the lifting linkage block, and a reset shock absorbing groove is provided in the reset shock absorbing seat; A reset linkage block slidably connected to the reset damping groove, the reset linkage block is slidably installed in a linkage guide seat, the linkage guide seat is fixedly connected to the side wall of the sensor housing, and the side wall of the reset linkage block is elastically connected with a driving ratchet; A linkage gear meshes with the driving ratchet for transmission, one side of the linkage gear is connected with a reset torsion spring, and the reset torsion spring is fixedly installed on the inner wall of the sensor housing.

10. A monitoring method for a power equipment status monitoring sensor, implemented by using the power equipment status monitoring sensor according to any one of claims 1 to 9, characterized in that: The electric power equipment status monitoring sensor monitoring method specifically includes: Collect the real-time status signal of the power equipment in real time, amplify and convert the real-time status signal to obtain the real-time status data, and upload the real-time status data to the status database; Normalize and pre-process the real-time status data, and store the normalized real-time status data; Pre-build a status assessment model, traverse the real-time status data, use the status assessment model to analyze and evaluate the real-time status data, and output the equipment status assessment results; Obtain the device status assessment results and push them to the user end in real time based on the local area network.

Images