Switch cabinet line temperature measurement early warning system based on electric field induction electricity taking

Abstract

The invention relates to the technical field of power equipment, in particular to a switch cabinet line temperature measurement and early warning system based on electric field induction electricity taking, and the system comprises an energy collection module which is used for capturing a magnetic field with corresponding position changes according to a deployed coil assembly, and converting the magnetic field into electric energy; the temperature sensor module is used for acquiring first temperature information of a specified monitoring point of a switch cabinet line in real time; the edge calculation module is used for screening and correcting abnormal information in all the first temperature information according to a preset abnormal detection algorithm every first preset time to obtain second temperature information corresponding to each piece of first temperature information; the information processing module judges whether all the second temperature information is abnormal or not according to the first temperature threshold value; and when any second temperature information is abnormal, an alarm signal for warning related personnel is generated. The beneficial effects are that temperature monitoring can be continuously carried out on key parts, potential risks can be found in time, electric field induction is utilized to take electricity, and the maintenance cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment, and in particular to a temperature measurement and early warning system for switchgear lines based on electric field induction power generation. Background Art

[0002] A switchgear is a complete set of switch equipment, and its internal structure includes various protection devices such as circuit breakers, operating mechanisms, and sensors. The main function of the switchgear is to switch on and off, control, and protect electrical equipment during the stable operation of the power system. Since the high-voltage electrical appliances inside the switchgear will generate resistance loss, hysteresis eddy current loss, and dielectric loss, and the opening and closing structures are oxidized and worn, the triggering and aging of the circuit breaker and the loosening of the busbar joints will cause the temperature of the internal lines of the switchgear to rise. The switchgear has strong airtightness and poor ventilation. If the heat cannot be removed in time, the excessive temperature rise will cause the contacts and cable compression nodes to overheat and burn out, and even lead to serious accidents such as the switchgear catching fire, power outage, and explosion. The abnormal operation or failure of the electrical equipment inside the switchgear is usually manifested as a temperature rise. Therefore, the on-line temperature detection and early warning of the switchgear are the most direct and effective means for monitoring electrical equipment.

[0003] The traditional temperature measurement technology uses an infrared temperature gun for manual detection. However, this method can only detect at a straight point, and is often blocked and unable to measure. It requires manual regular inspections, with a large workload, untimely monitoring, and close-range temperature measurement, resulting in a low safety factor. Summary of the Invention

[0004] (I) Technical Problems to be Solved

[0005] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a temperature measurement and early warning system for switchgear lines based on electric field induction power generation, which solves the technical problems of manual detection with an infrared temperature gun, being often blocked and unable to measure, requiring manual regular inspections, with a large workload, untimely monitoring, close-range temperature measurement, and low safety factor.

[0006] (II) Technical Solutions

[0007] To achieve the above object, the main technical solutions adopted by the present invention include:

[0008] In a first aspect, an embodiment of the present invention provides a temperature measurement and early warning system for switchgear lines based on electric field induction power generation, including: an energy collection module, a temperature sensor module, an edge computing module, and an information processing module; the energy collection module and the temperature sensor are arranged inside the switchgear, the edge computing module is arranged within a preset local range of the switchgear, and the information processing module is arranged at a remote end of the switchgear;

[0009] The energy collection module is electrically connected to the temperature sensor module, the temperature sensor module is communicatively connected to the edge computing module, and the edge computing module is communicatively connected to the information processing module;

[0010] An energy harvesting module, configured to capture a magnetic field with a corresponding position change according to a pre-deployed coil group and convert it into electric energy to power the temperature sensor module;

[0011] A temperature sensor module, configured to obtain first temperature information of a specified monitoring point on the switch cabinet line in real time;

[0012] An edge computing module, configured to obtain the first temperature information detected by the temperature sensor module in real time;

[0013] Moreover, every first preset time, preprocess all the first temperature information detected within the first preset time, screen and correct abnormal information in all the first temperature information according to a pre-set anomaly detection algorithm, and obtain second temperature information corresponding to each first temperature information;

[0014] An information processing module, configured to determine whether all the second temperature information is abnormal according to a pre-set first temperature threshold;

[0015] When any of the second temperature information is abnormal, generate an alarm signal to alert relevant personnel.

[0016] Optionally, the edge computing module preprocesses all the first temperature information detected within the first preset time, screens and corrects abnormal information in all the first temperature information according to a pre-set anomaly detection algorithm, and obtains second temperature information corresponding to each first temperature information, including:

[0017] Preprocess all the first temperature information detected within the first preset time; the preprocessing includes data cleaning, data sorting, and missing value filling;

[0018] According to all the preprocessed first temperature information and a pre-set formula 1, obtain a probability density function corresponding to each first temperature information; the formula 1 is:

[0019]

[0020] where, F K is the probability density function corresponding to the first temperature information with index K, N is the number of first temperature information detected within this time period, x i is the i-th first temperature information;

[0021] According to the probability density function corresponding to each preprocessed first temperature information and a pre-set second probability density threshold, determine whether this preprocessed first temperature information is abnormal information;

[0022] When it is determined that any pre - processed first temperature information is abnormal information, the abnormal information is corrected based on a pre - set correction algorithm to obtain second temperature information corresponding to the first temperature information;

[0023] When it is determined that any pre - processed first temperature information is normal information, the second temperature information is the first temperature information.

[0024] Optionally, the edge computing module corrects the abnormal information based on a pre - set correction algorithm to obtain second temperature information corresponding to the first temperature information, including:

[0025] According to the normal information among all the first temperature information and a pre - set formula two, the Euclidean distance between the abnormal information and all the normal information is obtained; the formula two is:

[0026]

[0027] where d z is the Euclidean distance between the abnormal information and the z - th normal information, r is the number of normal information, m is the abnormal information, and n z is the z - th normal information;

[0028] Based on all the Euclidean distances corresponding to the abnormal information, the number of the nearest neighbors corresponding to the abnormal information is obtained; the nearest neighbor is the normal information with the smallest Euclidean distance from the abnormal information;

[0029] According to the number of the nearest neighbors corresponding to the abnormal information, the neighbor weights corresponding to each nearest neighbor pre - set, and a pre - set formula three, the corrected information of the abnormal information, that is, the second temperature information, is obtained; the formula three is:

[0030]

[0031] where y is the second temperature information corresponding to the abnormal information, n is the number of the nearest neighbors, w k is the neighbor weight of the k - th nearest neighbor, and q k is the k - th nearest neighbor.

[0032] Optionally, the edge computing module is further configured to:

[0033] According to a pre - set correction effect evaluation algorithm, determine whether the corrected information corresponding to each abnormal information is reasonable;

[0034] When any corrected information is unreasonable, the corrected information is eliminated, that is, the corresponding second temperature information is eliminated;

[0035] The correction effect evaluation algorithm includes: time series algorithm, MSE algorithm, RMSE algorithm, MAE algorithm or coefficient of determination algorithm.

[0036] Optionally, the edge computing module is further configured to:

[0037] Whenever it is determined that any preprocessed first temperature information is abnormal information, save the abnormal information and the detection time corresponding to the abnormal information to a pre-set abnormal information database;

[0038] And, every second preset time, count the number of abnormal information within the second preset time, and determine whether to send an alarm signal to the information processing module according to a pre-set third abnormal information quantity threshold and the number of abnormal information within the second preset time;

[0039] And, every third preset time, back up and delete all abnormal information and the detection time corresponding to each abnormal information in the abnormal information database;

[0040] And, in real time, determine whether to send an alarm signal to the information processing module according to the number of abnormal information in the abnormal information database and a pre-set fourth abnormal information quantity threshold.

[0041] Optionally, the information processing module is further configured to:

[0042] Obtain in real time through the edge computing module the detection time corresponding to each second temperature information, where the detection time corresponding to the second temperature information is the time when the temperature sensor module obtains the first temperature information corresponding to the second temperature information;

[0043] Based on all the second temperature information obtained within the first preset time and the detection time corresponding to each second temperature information, establish a corresponding temperature time series;

[0044] Extract a first temperature distribution feature matrix corresponding to the temperature time series according to a pre-set feature extraction algorithm; the feature extraction algorithm includes: wavelet transform algorithm, autoregressive algorithm or Fourier transform;

[0045] Input the first temperature distribution feature matrix into a pre-set temperature prediction model to obtain a future temperature prediction result;

[0046] The future temperature prediction result includes third temperature information corresponding to each unit time within a future fourth preset time;

[0047] Judge whether all the third temperature information is abnormal according to a pre-set fifth temperature threshold;

[0048] When any of the third temperature information is abnormal, generate an alarm signal to alert relevant personnel.

[0049] Optionally, the information processing module extracts a first temperature distribution feature matrix corresponding to the temperature time series according to a preset feature extraction algorithm, including:

[0050] Construct a second temperature distribution matrix according to the temperature time series;

[0051] Perform singular value decomposition on the second temperature distribution matrix based on a preset singular value decomposition formula to obtain a singular value set corresponding to the second temperature distribution matrix; the singular value decomposition formula is:

[0052] A = U∑V T ;

[0053] where A is the second temperature distribution matrix, U and V are orthogonal matrices, Σ is a diagonal matrix, and T is the transpose;

[0054] The singular value set includes a preset number of singular values screened based on the diagonal matrix;

[0055] Construct a third temperature distribution feature matrix according to the left singular vector and right singular vector corresponding to each singular value in the singular value set;

[0056] Perform feature extraction on the temperature time series according to a preset wavelet transform algorithm to obtain a corresponding fourth temperature distribution feature matrix;

[0057] Obtain a first temperature distribution feature matrix corresponding to the temperature time series according to the third temperature distribution feature matrix and the fourth temperature distribution feature matrix, and a preset formula four; the formula four is:

[0058] F = F 1 + w 1 F 2 ;

[0059] where F is the first temperature distribution feature matrix, F 1 is the third temperature distribution feature matrix, F 2 is the fourth temperature distribution feature matrix, and w 1 is a preset weight function.

[0060] Optionally, the information processing module performs feature extraction on the temperature time series according to a preset wavelet transform algorithm to obtain a corresponding fourth temperature distribution feature matrix, including:

[0061] Obtain a corresponding change parameter according to the temperature time series; the change parameter is a parameter for the temperature in the temperature time series to change with time;

[0062] According to the change parameter, the temperature time series, and the preset Formula Five, a corresponding fourth temperature distribution feature matrix is obtained; Formula Five is as follows:

[0063]

[0064] Wherein, is the average temperature corresponding to the temperature time series, O is the change parameter, Ψ * is the complex conjugate of the wavelet basis function, a is the scaling factor, and b is the translation factor.

[0065] Optionally, the temperature prediction model includes a prediction model obtained by training an LSTM model through a preset first training set;

[0066] The first training set includes at least one first temperature distribution feature matrix with known temperature prediction results.

[0067] Optionally, the energy harvesting module includes: an energy harvesting unit and an energy storage unit;

[0068] The energy harvesting unit is electrically connected to the energy storage unit, and the energy storage unit is electrically connected to the temperature sensor module;

[0069] The energy harvesting unit is configured to capture a magnetic field with a corresponding position change according to a pre-deployed coil group and convert it into electric energy;

[0070] The energy storage unit is configured to store the electric energy converted by the energy harvesting unit and supply power to the temperature sensor module.

[0071] (III) Beneficial Effects

[0072] The beneficial effects of the present invention are as follows: A switchgear line temperature measurement and early warning system based on electric field induction power generation of the present invention uses electric field induction power generation to supply power to the temperature sensor module, and at the same time, the edge computing module provided locally in the switchgear preliminarily processes the information detected by the temperature sensor module. Compared with the prior art, it can continuously monitor the temperature of key parts, timely discover potential risks, and use electric field induction power generation to reduce the maintenance cost. Description of the Drawings

[0073] Figure 1 is a schematic structural diagram of a switchgear line temperature measurement and early warning system based on electric field induction power generation provided by an embodiment of the present invention;

[0074] Figure 2 is a structural block diagram of a switchgear line temperature measurement and early warning system based on electric field induction power generation provided by an embodiment of the present invention. Detailed Embodiments

[0075] For better explaining the present invention for easy understanding, the present invention will be described in detail below in conjunction with the accompanying drawings through specific embodiments.

[0076] A switchgear line temperature measurement and early warning system based on electric field induction power generation proposed in an embodiment of the present invention uses electric field induction power generation to supply power to a temperature sensor module, and at the same time, preliminarily processes the information detected by the temperature sensor module through an edge computing module set locally in the switchgear. Compared with the prior art, it can continuously monitor the temperature of key parts, timely discover potential risks, and reduce maintenance costs by using electric field induction power generation.

[0077] To better understand the above technical solution, the exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more clear and thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0078] Embodiment 1

[0079] This embodiment provides a switchgear line temperature measurement and early warning system based on electric field induction power generation, as Figure 1 shown, including: an energy collection module, a temperature sensor module, an edge computing module, and an information processing module; the energy collection module and the temperature sensor are arranged inside the switchgear, the edge computing module is arranged within a preset range locally in the switchgear, and the information processing module is arranged at a remote end of the switchgear;

[0080] The energy collection module is electrically connected to the temperature sensor module, the temperature sensor module is communicatively connected to the edge computing module, and the edge computing module is communicatively connected to the information processing module;

[0081] The energy collection module is configured to capture a magnetic field with a corresponding position change according to a pre-deployed coil group and convert it into electric energy to supply power to the temperature sensor module;

[0082] The temperature sensor module is configured to obtain first temperature information of a specified monitoring point of the switchgear line in real time;

[0083] The edge computing module is configured to obtain the first temperature information detected by the temperature sensor module in real time;

[0084] And, every first preset time, preprocess all the first temperature information detected within the first preset time, screen and correct abnormal information in all the first temperature information according to a pre-set abnormal detection algorithm, and obtain second temperature information corresponding to each first temperature information;

[0085] An information processing module determines whether all the second temperature information is abnormal according to a preset first temperature threshold;

[0086] When any of the second temperature information is abnormal, an alarm signal is generated to alert relevant personnel.

[0087] A switchgear line temperature measurement and early warning system based on electric field induction power generation provided in this embodiment, while using electric field induction power generation to supply power to the temperature sensor module, preliminarily processes the information detected by the temperature sensor module through an edge computing module set locally in the switchgear. Compared with the prior art, it can continuously monitor the temperature of key parts, timely discover potential risks, and reduce maintenance costs by using electric field induction power generation.

[0088] Embodiment 2

[0089] A switchgear line temperature measurement and early warning system based on electric field induction power generation provided in this embodiment, as Figure 2 shown, includes: an information processing module provided at the remote end of the switchgear, and a switchgear network monitored and warned by the information processing module; the switchgear network includes at least one switchgear;

[0090] An energy collection module and a temperature sensor module are correspondingly arranged inside each switchgear; an edge computing module is arranged near each switchgear (i.e., within a locally preset range).

[0091] Each energy collection module is electrically connected to its corresponding temperature sensor module, each temperature sensor module is communicatively connected to its corresponding edge computing module, and all the edge computing modules are communicatively connected to the information processing module.

[0092] Among them, any energy collection module is used to capture the magnetic field changing at the corresponding position according to a pre-deployed coil group and convert it into electric energy to supply power to the temperature sensor module; the coil group is usually arranged near the conductor close to the current source. When current passes through these conductors, an alternating magnetic field will be generated around, and the coil can capture the change of these magnetic fields and generate a corresponding electromotive force according to its change rate. In order to maximize the energy collection efficiency, it is necessary to carefully select the coil material and number of turns and ensure that it is correctly arranged in a suitable position to maximize the capture of magnetic field changes. The coil should be as close as possible to the current source so that it can effectively sense the magnetic field change.

[0093] Any temperature sensor module is used to obtain the first temperature information of a specified monitoring point on the switchgear line in real time; generally, the temperature sensor module is a temperature sensor group, including multiple temperature sensors (temperature measurement sensors) arranged at different detection points. In order to ensure comprehensive and accurate monitoring of the temperature changes inside the switchgear, it is necessary to plan the installation positions of the temperature sensors, that is, arrange corresponding temperature sensors at the cable joints with large currents, circuit breaker contacts and other parts prone to heat generation, as well as at different levels and corners of the switchgear; at this time, the first temperature information includes the temperature information detected by all temperature sensors (that is, the first temperature information includes multiple dimensions, and each dimension has the position coding of the corresponding temperature sensor). Each temperature sensor uses wireless technologies such as Zigbee, Lora or Bluetooth to send the temperature information detected by itself, the detection time of the temperature information and the position coding corresponding to the temperature sensor to the corresponding edge computing module in real time.

[0094] Any edge computing module is used to obtain the first temperature information detected by the temperature sensor module in real time; and, every first preset time, preprocess all the first temperature information detected within the first preset time, screen and correct the abnormal information in all the first temperature information according to the pre-set abnormal detection algorithm, and obtain the second temperature information corresponding to each first temperature information.

[0095] The information processing module determines whether all the second temperature information is abnormal according to the pre-set first temperature threshold;

[0096] When any of the second temperature information is abnormal, an alarm signal is generated to alert relevant personnel.

[0097] Since the information processing module monitors and warns all switchgears in the switchgear network, and the amount of information is huge, corresponding edge computing modules are set locally for each switchgear to preliminarily process the corresponding switchgear information, which can reduce the processing burden of the information processing module and relieve the pressure in the data transmission process, while having a higher reaction speed and better real-time performance. On this basis, the processing frequency of the edge computing module for the first temperature information can be set according to the actual situation of the corresponding switchgear (such as the importance or danger level).

[0098] Furthermore, considering that the changes in current and magnetic field may be intermittent, energy storage elements are needed, that is, the energy harvesting module includes an energy harvesting unit and an energy storage unit; the energy harvesting unit is electrically connected to the energy storage unit, and the energy storage unit is electrically connected to the temperature sensor module; the energy harvesting unit is used to capture the magnetic field changing at the corresponding position according to the pre-deployed coil group and convert it into electric energy; the energy storage unit is used to store the electric energy converted by the energy harvesting unit and supply power to the temperature sensor module.

[0099] Furthermore, the alternating current generated from the coil needs to be converted into direct current by a rectifier circuit (such as a bridge rectifier circuit), and then the output voltage is adjusted by a voltage regulator circuit to stabilize it at a level suitable for use by the temperature sensor module (such as 3.3V or 5V).

[0100] The present embodiment provides a switch cabinet line temperature measurement and early warning system based on electric field induction power supply, which delegates part of the computing tasks to the edge computing module corresponding to each switch cabinet, effectively reducing the processing burden of the information processing module, alleviating the pressure of the data transmission process, reducing the processing and transmission delays, and improving the real-time and effectiveness of monitoring and early warning.

[0101] Example 3

[0102] This embodiment provides a switch cabinet line temperature measurement and early warning system based on electric field induction power extraction, including: an energy collection module, a temperature sensor module, an edge computing module and an information processing module; the energy collection module and the temperature sensor are arranged inside the switch cabinet, the edge computing module is arranged within a local preset range of the switch cabinet, and the information processing module is arranged at the remote end of the switch cabinet;

[0103] The energy collection module is electrically connected to the temperature sensor module, the temperature sensor module is communicatively connected to the edge computing module, and the edge computing module is communicatively connected to the information processing module;

[0104] The energy collection module is used to capture the magnetic field changing at the corresponding position according to the pre-deployed coil group and convert it into electrical energy to power the temperature sensor module;

[0105] A temperature sensor module is used to obtain the first temperature information of a specified monitoring point of the switch cabinet line in real time;

[0106] An edge computing module, used for acquiring first temperature information detected by the temperature sensor module in real time;

[0107] And every first preset time, all the first temperature information detected within the first preset time is preprocessed; the preprocessing includes data cleaning, data sorting and missing value filling. Data cleaning aims to remove or correct incorrect, incomplete or irrelevant data to ensure that a temperature sensor has only one temperature reading at each time point, and to ensure that the temperature reading of the temperature sensor is reasonable. Data sorting is mainly to make the data format appropriate, including timestamp standardization and sorting based on timestamps. Missing value filling is the process of processing missing data in a data set. Commonly used methods include forward filling, backward filling, interpolation, mean filling and median filling.

[0108] For the first temperature information after preprocessing, the abnormal information in all the first temperature information is screened and corrected through a pre-set abnormal detection algorithm to ensure the correctness and stability of the data. Commonly used abnormal detection algorithms include the threshold method, the standard deviation method, the K-nearest neighbor method, etc. In this embodiment, according to all the preprocessed first temperature information and a pre-set formula 1, the probability density function corresponding to each first temperature information is obtained; the formula 1 is:

[0109]

[0110] where F K is the probability density function corresponding to the first temperature information with index K, N is the number of the first temperature information detected during this time period, x i is the i-th first temperature information;

[0111] According to the probability density function corresponding to each preprocessed first temperature information and a pre-set second probability density threshold, it is determined whether the preprocessed first temperature information is abnormal information;

[0112] When it is determined that any preprocessed first temperature information is abnormal information, according to the normal information in all the first temperature information and a pre-set formula 2, the Euclidean distance between this abnormal information and all the normal information is obtained; the formula 2 is:

[0113]

[0114] where d z is the Euclidean distance between this abnormal information and the z-th normal information, r is the number of normal information, m is the abnormal information, n z is the z-th normal information;

[0115] Based on all the Euclidean distances corresponding to this abnormal information, the number of the nearest neighbors corresponding to this abnormal information is obtained; the nearest neighbor is the normal information with the smallest Euclidean distance from the abnormal information;

[0116] According to the number of the nearest neighbors corresponding to this abnormal information, the neighbor weight corresponding to each nearest neighbor pre-set, and a pre-set formula 3, the corrected information of this abnormal information, that is, the second temperature information, is obtained; the formula 3 is:

[0117]

[0118] where y is the second temperature information corresponding to this abnormal information, n is the number of the nearest neighbors, w k is the neighbor weight of the k-th nearest neighbor, q k is the k-th nearest neighbor.

[0119] When it is determined that any of the pre - processed first temperature information is normal information, the second temperature information is the first temperature information.

[0120] Further, after data correction, it can be evaluated through a correction effect evaluation algorithm to ensure that the corrected data is as close as possible to the true value and does not introduce new errors or distort the trend of the original data. That is, according to the pre - set correction effect evaluation algorithm, it is judged whether the correction information corresponding to each abnormal information is reasonable;

[0121] When any correction information is unreasonable, the correction information is excluded, that is, the corresponding second temperature information is excluded;

[0122] The correction effect evaluation algorithm includes: time - series algorithm, MSE algorithm, RMSE algorithm, MAE algorithm or coefficient of determination algorithm.

[0123] Further, the edge computing module is also used for: whenever it is determined that any of the pre - processed first temperature information is abnormal information, the abnormal information and the detection time corresponding to the abnormal information are saved to a pre - set abnormal information database;

[0124] And, every second preset time, count the number of abnormal information within the second preset time, and judge whether to send an alarm signal to the information processing module according to the pre - set third abnormal information quantity threshold and the number of abnormal information within the second preset time;

[0125] And, every third preset time, back up and delete all abnormal information and the detection time corresponding to each abnormal information in the abnormal information database;

[0126] And, in real - time, judge whether to send an alarm signal to the information processing module according to the number of abnormal information in the abnormal information database and the pre - set fourth abnormal information quantity threshold.

[0127] A switch cabinet line temperature measurement and early warning system based on electric - field induction power taking provided by this embodiment uses the edge computing module as the basis for judging whether an abnormal event occurs for the preliminarily processed information, enhancing the security of the system while reducing the burden on the information processing module.

[0128] Further, the information processing module is also used for:

[0129] Obtain in real - time through the edge computing module the detection time corresponding to each second temperature information, and the detection time corresponding to the second temperature information is the time when the temperature sensor module obtains the first temperature information corresponding to the second temperature information;

[0130] Based on all the second temperature information obtained within the first preset time and the detection time corresponding to each second temperature information, establish a corresponding temperature-time series;

[0131] Extract the first temperature distribution feature matrix corresponding to the temperature-time series according to a preset feature extraction algorithm; the feature extraction algorithm includes: wavelet transform algorithm, autoregressive algorithm or Fourier transform;

[0132] Input the first temperature distribution feature matrix into a preset temperature prediction model to obtain a future temperature prediction result;

[0133] The future temperature prediction result includes the third temperature information corresponding to each unit time within the future fourth preset time;

[0134] Judge whether all the third temperature information is abnormal according to a preset fifth temperature threshold;

[0135] When any of the third temperature information is abnormal, generate an alarm signal to alert relevant personnel.

[0136] Further, the information processing module extracts the first temperature distribution feature matrix corresponding to the temperature-time series according to a preset feature extraction algorithm, including:

[0137] Construct a second temperature distribution matrix according to the temperature-time series;

[0138] Perform singular value decomposition on the second temperature distribution matrix based on a preset singular value decomposition formula to obtain a singular value set corresponding to the second temperature distribution matrix; the singular value decomposition formula is:

[0139] A = U∑V T ;

[0140] where A is the second temperature distribution matrix, U and V are orthogonal matrices, Σ is a diagonal matrix, and T is the transpose;

[0141] The singular value set includes a preset number of singular values screened based on the diagonal matrix;

[0142] Construct a third temperature distribution feature matrix according to the left singular vector and right singular vector corresponding to each singular value in the singular value set;

[0143] Perform feature extraction on the temperature-time series according to a preset wavelet transform algorithm to obtain a corresponding fourth temperature distribution feature matrix;

[0144] Obtain the first temperature distribution feature matrix corresponding to the temperature-time series according to the third temperature distribution feature matrix and the fourth temperature distribution feature matrix, and a preset formula four; the formula four is:

[0145] F = F 1 + w 1 F 2 ;

[0146] Wherein, F is the first temperature distribution feature matrix, F 1 is the third temperature distribution feature matrix, F 2 is the fourth temperature distribution feature matrix, w 1 is a preset weight function.

[0147] Furthermore, the information processing module extracts features from the temperature time series according to a preset wavelet transform algorithm to obtain the corresponding fourth temperature distribution feature matrix, including:

[0148] Obtain the corresponding change parameter according to the temperature time series; the change parameter is the parameter of the temperature changing with time in the temperature time series;

[0149] Obtain the corresponding fourth temperature distribution feature matrix according to the change parameter, the temperature time series, and a preset formula five; the formula five is:

[0150]

[0151] Wherein, is the average temperature corresponding to the temperature time series, O is the change parameter, Ψ * is the complex conjugate of the wavelet basis function, a is the scaling factor, and b is the translation factor.

[0152] Furthermore, the pre-trained models that the temperature prediction model can select include the random forest model, XGBoost model, LGBM model, convolutional neural network model, recurrent neural network model, long short-term memory network model, gated recurrent model, self-compiler model, or vector machine model. And in this embodiment, the LSTM model is used for temperature prediction. That is, the temperature prediction model includes a prediction model obtained by training the LSTM model with a preset first training set;

[0153] The first training set includes at least one first temperature distribution feature matrix with known temperature prediction results.

[0154] When training the temperature prediction model, it is necessary to ensure that the first temperature distribution feature matrix used is processed, including but not limited to steps such as missing value filling, standardization, or normalization.

[0155] The LSTM model requires the input data to have a time dimension, so it is necessary to convert the first temperature distribution feature matrix into a three-dimensional tensor form (samples, time_steps, and features) suitable for the LSTM model.

[0156] Samples: Each independent time series sample.

[0157] Time_steps: The time step, that is, the number of historical data points included in each sample.

[0158] Features: The number of features at each time point.

[0159] After preparing the input data, it needs to be divided into a training set and a test set to evaluate the model performance.

[0160] Use the prepared training data to train the model. After training is completed, the test set can be used to evaluate the model performance and make predictions.

[0161] Through the above steps, the first temperature distribution feature matrix processed by wavelet transform can be used as the input to perform temperature prediction.

[0162] A switchgear line temperature measurement and early warning system based on electric field induction power generation provided by this embodiment, while using electric field induction power generation to supply power to the temperature sensor module, preliminarily processes the information detected by the temperature sensor module through an edge computing module set locally in the switchgear. Compared with the prior art, it can continuously monitor the temperature of key parts, discover potential risks in a timely manner, and reduce maintenance costs by using electric field induction power generation. On this basis, the system provided by this embodiment makes predictions about the future based on real-time data, ensuring the safety and stability of the switchgear during use and providing protection for the life and property safety of users.

[0163] In the description of the present invention, it should be understood that the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "a plurality" means two or more unless otherwise specifically defined.

[0164] In the present invention, unless otherwise clearly specified and limited, the terms "installation", "connection", "connection", "fixation" and other terms should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium; it can be the internal communication of two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0165] In the present invention, unless otherwise clearly defined and limited, a first feature being "on" or "under" a second feature may mean that the first and second features are in direct contact, or that the first and second features are in indirect contact via an intermediate medium. Further, a first feature being "above", "over" and "on top of" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the horizontal height of the first feature is higher than that of the second feature. A first feature being "under", "below" and "beneath" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the horizontal height of the first feature is lower than that of the second feature.

[0166] In the description of this specification, the description of terms such as "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples", etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0167] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A switch cabinet line temperature measurement and early warning system based on electric field induction power extraction, characterized in that: include: Energy collection module, temperature sensor module, edge computing module and information processing module; The energy collection module and the temperature sensor are arranged inside the switch cabinet, the edge computing module is arranged within the local preset range of the switch cabinet, and the information processing module is arranged at the remote end of the switch cabinet; The energy collection module is electrically connected to the temperature sensor module, the temperature sensor module is communicatively connected to the edge computing module, and the edge computing module is communicatively connected to the information processing module; An energy collection module, used to capture the magnetic field changing at the corresponding position according to the pre-deployed coil group and convert it into electrical energy to power the temperature sensor module; A temperature sensor module is used to obtain the first temperature information of a specified monitoring point of the switch cabinet line in real time; An edge computing module, used for acquiring first temperature information detected by the temperature sensor module in real time; And, at every first preset time, pre-processing all first temperature information detected within the first preset time, screening and correcting abnormal information in all first temperature information according to a preset abnormality detection algorithm, and obtaining second temperature information corresponding to each first temperature information; An information processing module, for determining whether all the second temperature information is abnormal according to a preset first temperature threshold; When any of the second temperature information is abnormal, an alarm signal is generated to alert relevant personnel.

2. The switch cabinet line temperature measurement and early warning system based on electric field induction power extraction according to claim 1 is characterized in that: The edge computing module pre-processes all first temperature information detected within a first preset time, screens and corrects abnormal information in all first temperature information according to a preset abnormality detection algorithm, and obtains second temperature information corresponding to each first temperature information, including: Preprocessing all first temperature information detected within a first preset time; the preprocessing includes data cleaning, data sorting and missing value filling; According to all the preprocessed first temperature information and the preset formula 1, the probability density function corresponding to each first temperature information is obtained; the formula 1 is: Among them, F K is the probability density function corresponding to the first temperature information with index K, N is the number of first temperature information detected in this time period, x i is the i-th first temperature information; According to the probability density function corresponding to each pre-processed first temperature information and a preset second probability density threshold, determining whether the pre-processed first temperature information is abnormal information; When it is determined that any pre-processed first temperature information is abnormal information, the abnormal information is corrected based on a preset correction algorithm to obtain second temperature information corresponding to the first temperature information; When any pre-processed first temperature information is determined to be normal information, the second temperature information is the first temperature information.

3. The switch cabinet line temperature measurement and early warning system based on electric field induction power extraction according to claim 2 is characterized in that: The edge computing module corrects the abnormal information based on a preset correction algorithm to obtain second temperature information corresponding to the first temperature information, including: According to the normal information in all the first temperature information and the preset formula 2, the Euclidean distance between the abnormal information and all the normal information is obtained; the formula 2 is: Among them, d z is the Euclidean distance between the abnormal information and the zth normal information, r is the number of normal information, m is the abnormal information, n z is the zth normal information; Based on all Euclidean distances corresponding to the abnormal information, the number of nearest neighbors corresponding to the abnormal information is obtained; the nearest neighbor is the normal information with the smallest Euclidean distance to the abnormal information; According to the number of nearest neighbors corresponding to the abnormal information and the preset neighbor weight corresponding to each nearest neighbor, and the preset formula three, the corrected information of the abnormal information is obtained, that is, the second temperature information; the formula three is: Among them, y is the second temperature information corresponding to the abnormal information, n is the number of nearest neighbors, and w k is the neighbor weight of the kth nearest neighbor, q k is the kth nearest neighbor.

4. The switch cabinet line temperature measurement and early warning system based on electric field induction power extraction according to claim 3 is characterized in that: The edge computing module is also used for: According to the preset correction effect evaluation algorithm, determine whether the correction information corresponding to each abnormal information is reasonable; When any correction information is unreasonable, the correction information is discarded, that is, the corresponding second temperature information is discarded; The correction effect evaluation algorithm includes: time series algorithm, MSE algorithm, RMSE algorithm, MAE algorithm or determination coefficient algorithm.

5. The switch cabinet line temperature measurement and early warning system based on electric field induction power extraction according to claim 2 is characterized in that: The edge computing module is also used for: Whenever any pre-processed first temperature information is determined to be abnormal information, the abnormal information and the detection time corresponding to the abnormal information are saved in a preset abnormal information database; and, at every second preset time, counting the number of abnormal information within the second preset time, and determining whether to send an alarm signal to the information processing module according to a preset third abnormal information number threshold and the number of abnormal information within the second preset time; And, every third preset time, back up and delete all abnormal information and the detection time corresponding to each abnormal information in the abnormal information database; And, judging whether to send an alarm signal to the information processing module in real time according to the amount of abnormal information in the abnormal information database and a fourth preset abnormal information amount threshold.

6. The switch cabinet line temperature measurement and early warning system based on electric field induction power extraction according to claim 1 is characterized in that: The information processing module is also used for: The detection time corresponding to each second temperature information is obtained in real time by the edge computing module, where the detection time corresponding to the second temperature information is the time when the temperature sensor module obtains the first temperature information corresponding to the second temperature information; Establishing a corresponding temperature time series based on all second temperature information acquired within a first preset time and a detection time corresponding to each second temperature information; Extracting a first temperature distribution feature matrix corresponding to the temperature time series according to a preset feature extraction algorithm; The feature extraction algorithm includes: wavelet transform algorithm, autoregressive algorithm or Fourier transform; Inputting the first temperature distribution characteristic matrix into a preset temperature prediction model to obtain a future temperature prediction result; The future temperature prediction result includes third temperature information corresponding to each unit time within a fourth preset time in the future; According to a preset fifth temperature threshold, determining whether all the third temperature information is abnormal; When any of the third temperature information is abnormal, an alarm signal is generated to alert relevant personnel.

7. The switch cabinet line temperature measurement and early warning system based on electric field induction power extraction according to claim 6 is characterized in that: The information processing module extracts a first temperature distribution feature matrix corresponding to the temperature time series according to a preset feature extraction algorithm, including: Constructing a second temperature distribution matrix according to the temperature time series; The second temperature distribution matrix is ​​subjected to singular value decomposition based on a preset singular value decomposition formula to obtain a singular value set corresponding to the second temperature distribution matrix; the singular value decomposition formula is: A=U∑V T ; Where A is the second temperature distribution matrix, U and V are orthogonal matrices, Σ is a diagonal matrix, and T is the transpose; The singular value set includes a preset number of singular values ​​screened based on a diagonal matrix; Constructing a third temperature distribution characteristic matrix according to the left singular vector and the right singular vector corresponding to each singular value in the singular value set; Perform feature extraction on the temperature time series according to a preset wavelet transform algorithm to obtain a corresponding fourth temperature distribution feature matrix; According to the third temperature distribution characteristic matrix and the fourth temperature distribution characteristic matrix, and the preset formula 4, the first temperature distribution characteristic matrix corresponding to the temperature time series is obtained; the formula 4 is: F=F1+w1F2; Among them, F is the first temperature distribution characteristic matrix, F1 is the third temperature distribution characteristic matrix, F2 is the fourth temperature distribution characteristic matrix, and w1 is a preset weight function.

8. The switch cabinet line temperature measurement and early warning system based on electric field induction power extraction according to claim 7 is characterized in that: The information processing module extracts features from the temperature time series according to a preset wavelet transform algorithm to obtain a corresponding fourth temperature distribution feature matrix, including: According to the temperature time series, a corresponding variation parameter is obtained; the variation parameter is a parameter of the temperature variation over time in the temperature time series; According to the variation parameter and the temperature time series, and the preset formula 5, the corresponding fourth temperature distribution characteristic matrix is ​​obtained; the formula 5 is: in, is the average temperature corresponding to the temperature time series, O is the variation parameter, Ψ * is the complex conjugate of the wavelet basis function, a is the scaling factor, and b is the translation factor.

9. The switch cabinet line temperature measurement and early warning system based on electric field induction power extraction according to claim 6 is characterized in that: The temperature prediction model includes a prediction model obtained by training an LSTM model with a preset first training set; The first training set includes at least one first temperature distribution feature matrix of known temperature prediction results.

10. The switch cabinet line temperature measurement and early warning system based on electric field induction power extraction according to claim 1 is characterized in that: The energy collection module includes: an energy collection unit and an energy storage unit; The energy collection unit is electrically connected to the energy storage unit, and the energy storage unit is electrically connected to the temperature sensor module; The energy collection unit is used to capture the magnetic field changing at the corresponding position according to the pre-deployed coil group and convert it into electrical energy; The energy storage unit is used to store the electric energy converted by the energy collection unit to supply power to the temperature sensor module.

Images