Current transformer secondary output signal normalization system applied to coal mines

Abstract

The application provides a current transformer secondary output signal normalization system applied to a coal mine. The system comprises a host computer and a display unit, wherein the host computer and the display unit are in communication connection; the host computer comprises a plurality of signal connection ports and a digital processor; the signal connection ports are used for automatically adapting the secondary output signal of the current transformer and performing signal waveform simulation, the digital processor is internally provided with a normalization program, and the secondary output signal is normalized by the normalization program; the display unit comprises at least one fusion display interface and a plurality of signal display interfaces; the fusion display interface is used for displaying the normalized fusion signal, and the signal display interface is used for displaying the secondary output signal. The application can reduce manual intervention and adapt to the signal input of the multi-specification current transformer under the complex environment of the coal mine. Through classification and segmentation processing, the normalization effect of the abnormal signal is improved. The fusion interface is combined with the independent interface to meet different analysis requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of mine power supply, and in particular to a current transformer secondary output signal normalization system applied to coal mines. Background Art

[0002] In the mine power supply system, the primary side of the current transformer has a rated current input condition, and the secondary output is a variety of signal outputs such as 5A, 1A, 0.1A, 6V, 1V, 0.7V, and 0.1V. This leads to a wide variety of analog quantity acquisition circuit hardware for the upgraded equipment, which is not conducive to the unification of specifications for underground protection equipment in coal mines.

[0003] Therefore, it is urgent to develop a signal conversion device that can normalize the various signals output from the secondary side of the current transformer. Adding a signal conversion device will help upgrade the hardware uniformity of the equipment, facilitate coal mine equipment management, system maintenance, and safe operation. Summary of the Invention

[0004] The purpose of the present invention is to solve the problem that in a mine power supply system, when the primary side of the current transformer is under the condition of rated current input, the secondary output is 5A, 1A, 0.1A, 6V, 1V, 0.7V, 0.1V and other signal outputs, resulting in a wide variety of analog quantity acquisition circuit hardware for upgraded equipment, which is not conducive to the unification of specifications of underground protection equipment in coal mines.

[0005] In a first aspect, the present application proposes a current transformer secondary output signal normalization system for coal mines, comprising a host and a display unit, wherein: the host and the display unit are communicatively connected;

[0006] The host includes multiple signal connection ports and a digital processor; wherein the signal connection port is used to automatically adapt to the secondary output signal of the current transformer and perform signal waveform simulation; the digital processor has a built-in normalization program, and normalizes the secondary output signal through the normalization program;

[0007] The display unit includes at least one fusion display interface and multiple signal display interfaces; wherein the fusion display interface is used to display the fusion signal after normalization processing, and the signal display interface is used to display the secondary output signal.

[0008] In this embodiment, for a large number of devices that may exist on a mine, different current transformers exist for each device, which is difficult to achieve, and different current transformer detection devices are needed. The application automatically adapts the current transformer through the signal connection port, and then uniformly converts the signal of the current transformer into a digital signal output. The normalized program built in the digital processor performs normalization calculation on the digital signal output, converts the different signals of the current transformer into a standard dimension that can be displayed through the display unit, thereby eliminating the data differences between various devices with current transformers inside the coal mine, and the problem that data analysis cannot be performed. Multi-scene mixed use can be performed, and false judgments caused by format differences can be avoided. When displaying the test results of the current transformer, the integrated multi-channel normalized signal can be displayed through the display unit, the comprehensive detection of various devices with current transformers can be realized, a global perspective can be provided, and comprehensive analysis can be performed. For each device with a current transformer, the signal display interface can also provide detailed tracing of the individual current transformer, display any one original signal with a current transformer, thereby realizing fault positioning and normalized data comparison.

[0009] In a possible implementation manner of the first aspect, the normalization processing of the secondary output signal by the normalization program comprises:

[0010] Waveform features of the analog waveform obtained by simulating the signal waveform are extracted, and the waveform features are classified;

[0011] According to the waveform feature classification, the secondary output signals subjected to the normalization processing are classified, and the secondary output signals participating in the normalization calculation are determined after quality inspection and the corresponding normalization coefficients are determined;

[0012] The normalization feature parameters of the secondary output signals are determined one by one, and for abnormal secondary output signals in which the amplitude balance degree and the waveform distortion coefficient are obviously not in a linear relationship in the coordinate, the normalization feature parameters are calculated in segments;

[0013] According to the waveform feature classification, the normalization average values of the normalization feature parameters of different secondary output signals are determined;

[0014] According to the normalization average values, the normalization relative fusion curves of different secondary output signals are generated;

[0015] According to the normalization relative fusion curves, the fusion coefficients of the fused different secondary output signals in different operating states of the current transformer are determined;

[0016] And the different output signals are normalized by the fusion coefficients to determine the normalization values.

[0017] In this embodiment, after receiving the secondary output signal of the current transformer at the signal connection port, the signal waveform of the secondary output signal is simulated through time-frequency domain analysis, the signal characteristics are extracted, and then the abnormal signals and normal signals in different secondary output signals are distinguished through the signal characteristics; according to the quality inspection, the abnormal signals are removed, and the normalized coefficient after normalization calculation is output, which is used to represent the signal quality of the normalized output signal; then, the secondary output signal is analyzed one by one, and the normalized feature parameters of each waveform feature are generated, if there is amplitude balance abnormality or waveform coefficient abnormality in the secondary output signal, the segmented normalization is performed, the segmented normalization calculation is realized, and the error problem in complex working condition long signal detection is avoided. Finally, the normalized feature parameters of different signals are fused to realize the cooperative normalization of multi-source signals and enhance the analysis effect of multi-signal joint analysis.

[0018] Based on the first aspect, in possible implementation manners, the signal waveform simulation includes:

[0019] The time domain characteristics and frequency domain characteristics of the current transformer secondary output signal are analyzed, and the signal modal component that needs to be simulated is determined;

[0020] According to the signal modal component, a corresponding chaotic system is matched and a chaotic signal is generated through numerical integration;

[0021] According to the characteristic index of the current transformer secondary output signal, the amplitude modulation and phase modulation parameters of the chaotic signal are calculated;

[0022] According to the control parameters, the amplitude and phase of the chaotic signal are modulated;

[0023] The modulated signals are synthesized to obtain the simulated current transformer secondary output signal waveform, and the simulated waveform is output.

[0024] In this embodiment, waveform simulation is to more accurately extract the characteristics of the secondary output signal. The current signal of the current transformer in the coal mine is nonlinear, and nonlinear simulation is often required for feature extraction to produce a large amount of water to realize feature extraction. The present application performs waveform simulation on the basis of the waveform of the secondary signal, extracts key modal components, determines the simulated signal characteristics, and then uses the vertical moving integral calculation of the chaotic system to simulate the complex working condition random interference, determine the characteristics of the secondary output signal, and make the simulated chaotic signal as real as possible to meet the scene of the coal mine and improve the signal analysis effect in extreme scenes.

[0025] Based on the first aspect, in possible implementation manners, the signal connection port includes a signal detection unit, a signal analysis unit control unit;

[0026] The signal detection unit is configured to receive a secondary output signal of the current transformer.

[0027] The signal analysis unit determines the signal strength according to a preset threshold value, and determines a port specification to which the signal should be matched.

[0028] The control unit adjusts the signal strength of the signal connection port according to the port specification through a relay matrix or an electronic switch.

[0029] In this embodiment, the signal connection port is configured to receive the secondary output signal. However, in coal mines, signal adaptation is required for a variety of different devices with current transformers to obtain the most accurate results. Traditional technologies mainly rely on manual switching of ports or implementation of multiple types of current transformers for model adaptation. However, even if the adaptation is successful and the signal of the current transformer can be transmitted, signal interference may still occur due to environmental interference. Therefore, the present application physically isolates different signals through a relay matrix or an electronic switch, and adjusts the port specification, i.e., the gain multiple of the transmission line, to prevent crosstalk while ensuring the stability of the signal.

[0030] Based on the first aspect, in possible implementation manners, the normalization processing of the secondary output signal through the normalization program further includes:

[0031] Collecting raw data of the secondary output signal of the current transformer, including normal signals and abnormal signals;

[0032] Preprocessing the raw signals to determine training corpus; wherein the preprocessing includes sampling rate conversion, length fixing, and data cleaning;

[0033] Constructing a self-supervised signal normalization supervision model based on de-statistical instance normalization, training the model using the training corpus, and obtaining a trained target supervision model;

[0034] Inputting the normalized secondary output signal of the current transformer to be verified into the trained target supervision model to determine an enhanced verification signal;

[0035] By comparing the enhanced verification signal with the raw signal, the effectiveness of the normalization result is determined.

[0036] In this embodiment, during the normalization processing, in order to solve the linear scaling problem caused by the dependence of traditional normalization on fixed rules, as well as the distortion of the normalized signal and the loss of key features. The present application constructs a statistical instance normalization supervision model, which can prevent the distortion of the normalized signal in the case of enhanced signal and realize the verification of the enhanced signal. The effectiveness of the normalization effect is guaranteed.

[0037] Based on the first aspect, in a possible implementation, the self-supervisory signal normalization supervision model is constructed by the following steps:

[0038] Collect text description information and device description information of the current transformer secondary signal; wherein the text description information is first associated data of the influence of the signal characteristics on the normalization result; the device description information is second associated data of the influence of the device characteristics of the verified physical object on the normalization result;

[0039] Constructing first training sample data and second training sample data according to the first associated data and the second associated data;

[0040] Inputting the first training sample data into a first model substructure based on the semantic vector and outputting a first target vector;

[0041] Inputting the second training sample data into a second model substructure based on the device parameters and outputting a second target vector;

[0042] An attribute constraint model is constructed according to the first target vector and the second target vector, a target self-supervision model is trained, and a self-supervision signal normalization supervision model is generated.

[0043] In this embodiment, in the process of supervising the input secondary signal, the present application performs fusion association of text features and device parameters based on the secondary signal of the current transformer, and more accurately supervises different input parameters according to the attribute constraint model to improve time adaptability.

[0044] Based on the first aspect, in a possible implementation, the fusion display interface and the signal display interface of the display unit are connected;

[0045] Each signal display interface and fusion display interface is configured with a unique data transmission channel;

[0046] The fusion display interface is used to perform waveform processing on the normalized secondary output signal to generate a normalized waveform;

[0047] The fusion display interface and the signal display interface are connected to the preset visualization tool, and the corresponding display mode is generated according to the control instructions of the visualization tool.

[0048] In this embodiment, in order to solve the problems of traditional current transformers, such as either a single interface or multiple simultaneous displays, solid line overlap and data delay, a fusion signal interface and a signal display interface are set up to switch the display page according to user needs. At the same time, the only transmitted separate signal channel can be displayed in a targeted manner as an independent communication channel.

[0049] Based on the first aspect, in a possible implementation, the fusion display interface and the signal display interface are connected to a preset visualization tool, including:

[0050] Receive visualization instructions from the user and determine the signal display interface that needs to be fused;

[0051] Obtain raw data from different signal display interfaces and convert them into converted data that conforms to the preset visualization tool data format rules;

[0052] According to user needs, configure data visualization requirement data; wherein the visualization requirement data includes attribute data of data visualization elements and attribute data of data visualization scene data;

[0053] Use preset visualization tools to edit the converted data and generate data visualization elements that meet the requirements of attribute data;

[0054] Use the data visualization engine to edit data visualization elements and generate data visualization scene data that meets the attribute data requirements;

[0055] According to the data visualization scene data, different signal display interfaces are fused to generate data visualization resource files that conform to the preset visualization tool data format rules, and the visualization resource files are visualized through the fused display interface.

[0056] In this embodiment, in order to solve the problem of no visual interface, the interface fusion of the visualization engine can realize the visual display of multi-signal fusion and the visual display of the operating data of a single device.

[0057] Based on the first aspect, in a possible implementation, the display unit includes a preset visual interface configuration unit, the visual interface configuration unit being configured to schedule an interface card template according to a specification identifier of a secondary signal of the current transformer;

[0058] Among them, the specification identification is generated by the signal connection port on the host after the current transformer is connected;

[0059] The interface card template is a preset schedulable template for the fusion display interface and signal display interface.

[0060] In this embodiment, in order to solve the problem that traditional coal mines need to manually configure the visualization interface, this application can quickly generate a display interface for test results by configuring an interface card template.

[0061] Based on the first aspect, in a possible implementation, the display unit further includes an interface response mechanism; wherein the interface response mechanism is configured to:

[0062] Receive user operations on the fusion display interface and the signal display interface;

[0063] According to the operation behavior, a mapping relationship between the fusion display interface and the signal display interface and the operation behavior is established;

[0064] According to the mapping relationship, the corresponding target definition instructions are obtained on the preset rule engine;

[0065] Define instructions based on the target and respond to user operations.

[0066] In this embodiment, during the visual display process, in order to accurately understand the user's real-time intentions, this application sets up an operation linkage mechanism, and through the analysis of the operation linkage, determines the data mapping of the fusion display interface and the signal display interface to achieve collaborative analysis of multiple views. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] The accompanying drawings are used to further explain the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0068] Figure 1 This is a system operation composition diagram of the current transformer secondary output signal normalization system used in coal mines in the present invention;

[0069] Figure 2 is a flow chart of the normalization process of the normalization program in the present invention;

[0070] Figure 3 This is a flow chart of the waveform simulation process in the present invention;

[0071] Figure 4 This is an implementation diagram of the signal connection port in the present invention;

[0072] Figure 5 This is a flow chart of the process for determining the validity of the result of the present invention;

[0073] Figure 6 Schematic diagram of the construction process of the self-supervisory signal normalization supervision model in the present invention;

[0074] Figure 7 This is a flow chart showing the display mode implementation of the display unit in the present invention;

[0075] Figure 8 A diagram showing the docking process between the display interface and the visualization tool in the present invention;

[0076] Figure 9 A diagram showing the configuration process of the scheduling interface card of the visual interface configuration unit of the present invention;

[0077] Figure 10 This is a diagram showing the response process of the display unit interface in the present invention. DETAILED DESCRIPTION

[0078] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The embodiments are only used to explain the present invention and are not used to limit the scope of the present invention.

[0079] See Figure 1 , this application proposes a system block diagram of a current transformer secondary output signal normalization system for use in coal mines. The host used in this application is an electric power testing device or electric power monitoring device for the current transformer, which includes multiple signal connection ports. The signal connection port is a hardware interface composed of an adjustable resistor, an operational amplifier circuit, or an ADC module. When adapting the secondary signal of the current transformer, the adapted secondary output signal, including but not limited to the range, frequency, and waveform of the secondary signal, will be converted into a unified digital signal. The digital processor is electrically connected to the output end of the signal connection port, and a normalization algorithm is configured inside it. The normalization algorithm is a synchronous normalization algorithm that combines a signal amplification algorithm with a baseline calibration algorithm. The combination of the signal amplification algorithm and the baseline calibration algorithm can indirectly perform filtering during the normalization process. The filtering effect can eliminate device differences in data analysis. The data normalized by the digital processor will be transmitted to the display unit. The display unit is equipped with a data fusion algorithm based on weighted average and principal component analysis, which is used to perform comprehensive detection and display of multiple devices with current transformers that are monitored simultaneously in the coal mine, and display the overall load status of multiple devices with current transformers in the coal mine through a fusion display interface. For a single coal mine device with a current transformer, it can be split-screen displayed through the signal display interface, which is convenient for fault location and comparison of data before and after normalization to determine the cause of the fault. The signal normalization system of the present application is suitable for coal mine scenarios with mixed use of multiple models, which can reduce the complexity of operation. After normalization, overload warning analysis and harmonic detection analysis can be realized. Even if the signal formats of different devices are different, during analysis, because signal normalization is realized, it is not only easy to perform signal analysis, but also because after signal normalization, it can be determined whether there is a misjudgment by comparing similar signals. The fusion interface provides a global monitoring perspective, while the signal display interface of the split interface can realize detailed diagnosis of a single device.

[0080] Example 2:

[0081] See Figure 2 In actual implementation, when normalizing the collected current transformer signal, the waveform features are first extracted through time-frequency domain analysis. Time domain analysis includes peak value and mean value; frequency domain analysis includes extracting waveform features using FFT.

[0082] In the process of waveform feature classification, SVM, random forest, and K-means algorithms are used to perform signal classification. Signal classification is performed, and the classification results include normal, overload, short circuit, distortion, noise interference, etc.

[0083] During quality inspection and normalization calculations, the normalization coefficient can be calculated by applying dynamic scaling factors to signals with poor amplitude balance and introducing compensation parameters for waveform distortion signals. Preset thresholds or amplitude fluctuation range verification can also be used to exclude abnormal signals. Normalization characteristic parameters, such as phase difference or amplitude anomaly characteristics, can be determined signal by signal. Through waveform classification and quality inspection, abnormal signals can be excluded, preventing the occurrence of noise or faulty data, which can lead to errors in the normalization results.

[0084] During the segmented processing of abnormal signals, for nonlinear abnormal signals such as transient impulse waveforms, normalization parameters are calculated segmentally in the time domain, dividing the signal into multiple linear sub-segments. For example, calculations are performed every 100ms. A normalized relative fusion curve is then generated through curve fitting, reflecting the optimal fusion coefficient under different operating conditions. Segmented processing of normalized signals can enhance the transient impact of motor startup and short-circuit transients in complex coal mine environments without causing normalization distortion.

[0085] In the process of calculating the fusion coefficient and generating the fusion curve, weights can be assigned according to the signal confidence to collaboratively normalize the multi-source signals.

[0086] Example 3:

[0087] See Figure 3 : First, after the signal is connected to the port, the signal output from the secondary side of the current transformer will be analyzed in the time domain, including peak value, rise time, etc., and the frequency domain analysis includes FFT, wavelet transform, etc.; the secondary output signal of the current transformer, the fundamental wave, harmonics, transient impact components and other key modal components are extracted to determine the signal characteristics.

[0088] The secondary output signal has characteristic indicators such as nonlinearity and instability. Based on these characteristics, a chaotic signal is generated by performing numerical integration calculations using preset chaotic systems, such as the Lorenz system and Chua circuit, combined with amplitude modulation and phase modulation parameters. At this time, the generated chaotic signal simulates random interference components, such as electromagnetic noise and arc disturbance.

[0089] Chaotic signals will have amplitude-frequency characteristics such as rated current and harmonic content. The modulation parameters are calculated based on the amplitude-frequency characteristics to achieve dynamic matching between the chaotic signal and the modulation parameters, so that the simulated signal is close to the actual working condition. In this way, multiple groups of modulated chaotic signals are superimposed to cover the working scenarios of the current transformer, achieving a more realistic simulation of complex working conditions and obtaining a more accurate simulation waveform.

[0090] Example 4:

[0091] See Figure 4 In actual implementation, the signal connection port consists of three parts. The signal detection unit uses a high-precision ADC chip (such as the ADS1256) to collect signals and ensures electrical safety through isolation circuits (such as optocouplers). The signal analysis unit uses an FPGA to perform real-time threshold determination, dynamically matching signal strength to port specifications (such as 0-5V or 0-10V ranges). The control unit uses a relay matrix or solid-state electronic switches to switch the signal path and adjust the signal gain, achieving adaptive signal conditioning. If the signal strength exceeds the threshold, it triggers the interface to switch to the corresponding port specification for the range and protocol. The difficulty of traditional current transformers lies in the need for manual port switching and fixed specification design, requiring fixed specifications and adaptability. They are not suitable for mixed use scenarios with different transformation ratios or output ranges, making them unusable or insufficiently sensitive in arbitrary scenarios. When presetting thresholds, the thresholds can be dynamically applied based on the current transformer model. When using a relay matrix, a tree-like switching network is built based on the relay matrix, and the relay matrix is ​​controlled by trigger signals to respond to corresponding switching commands.

[0092] Example 5:

[0093] See Figure 5 :In actual implementation, data preprocessing will be performed on normal signals or abnormal signals in the original signal, including reducing the sampling frequency, eliminating outliers, filling in the actual points with differences, or converting normal signals and abnormal signals into fixed-length corpus. Then, through the sub-supervised learning framework, the statistical features of the waveform pattern and the text of the instance features are used as training corpus for normalization training by using the contrastive learning method. In this process, the model parameters are optimized by reconstructing the task or comparing the loss function to extract the normalized results of the signal. The normalized results are input into the trained model, and the robustness of the model is verified by adding controllable noise or time series perturbations. Finally, in the process of effectiveness evaluation, the key features of the normalized signal, such as waveform shape and time series, are judged based on the dynamic time warping distance. The normalization method of the present application can solve the problem that traditional technology relies on linear scaling and cannot adapt to complex current signals in coal mines, such as complex characteristics such as multimodal distribution or Gaussian noise. The signal under fault state is processed by the model to avoid normalization failure. The normalized results obtained are directly output for effective ones and feedback correction for invalid ones.

[0094] Example 6:

[0095] See Figure 6, the working condition description of the device corresponding to the current transformer in the traditional coal mine is often irrelevant to the device attribute, and the application eliminates a scheme for accurately identifying the device in a complex scene and accurately extracting the signal by using device information. First, or obtain the secondary signal of the current transformer, determine the text description information and device description information of the device corresponding to the current transformer in the cloud or local system according to the adaptation data of the current transformer. The text information is the specific information of the signal characteristics of such device, for example, the harmonic content of such device is high, there is transient impact, etc. In the case of high harmonic content, the calculation of the optimization result needs to increase the low-pass filtering weight, and then the first association data is constructed through these data.

[0096] For the second association data, the influence on the normalization result is determined according to the device type and device material information corresponding to the current transformer, for example, if the core of the device of the current transformer is silicon steel, it needs to be compensated dynamically because there is low magnetic saturation, therefore, the second association data also represents the device feature vector.

[0097] Then, the first encoder based on the device motion state and the second encoder based on the device material are generated by the first association data and the second association data, the two modal features are fused for training, and the device attribute constraint model is constructed to form a normalization supervision model. The attribute constraint model is a joint optimization constraint model constructed based on the device loss term. In actual implementation, the first model substructure is a pre-trained BERT model, and the second model substructure is a 3-layer fully connected network.

[0098] Embodiment 7:

[0099] Referring to Figure 7 In terms of display of the normalization result, the normalization result also represents the global state and local state of the current transformer device inside the coal mine, and the application realizes the cooperative display of the double-type interface. The fusion display interface and the signal display interface both have independent data transmission channels, and are mostly Ethernet, CAN bus or special serial port, realizing signal synchronization of data isolation. The fusion display interface displays the normalized comprehensive interface, and the signal display interface displays the single shunt display interface of a single current transformer. Each interface is allocated an independent communication link, but there are different communication channels to prevent delay.

[0100] The fusion interface uses a difference algorithm or wavelet transform data fusion technology to convert multi-channel signals into normalized signals. In this process, the fusion display interface will fuse multi-dimensional features, such as amplitude, phase and frequency. When displaying, the display mode is switched and rendered based on visualization tools. In this process, the display mode can be switched by switching the time domain waveform, the spectrum diagram and the vector diagram, and customized view layout and rendering can be achieved through the instructions of the user or administrator. When the present application is displayed, because there are independent channels and a fusion display interface, the data can be dynamically updated, while avoiding interface freezes or data misalignments, and different display modes can be switched as needed.

[0101] Example 8:

[0102] See Figure 8 In practice, traditional coal mine monitoring systems only visualize current transformer measurement signals. This prevents them from integrating and displaying the signals of individual current transformers to achieve dynamic scene configuration and response based on user needs. Therefore, upon receiving user commands, primarily visualization commands, the system sets the visualization display mode, selects the signal display interface, and triggers the corresponding data formatting logic. During the data conversion process, the raw signal data, including binary streams and analog voltages, is converted into a format compatible with visualization tools, allowing for viewing on web pages, mobile devices, or specific devices. The raw signal data is converted into a structured format compatible with visualization tools. User-defined visualization elements, such as waveform color and axis ranges, as well as scenes, are generated into corresponding configuration files, which display scene and visualization element properties. The data visualization engine can build visualization data implementation scenarios and generate corresponding industrial tables, such as line charts and heat maps. Furthermore, user-defined commands can be extended through scripts or plug-ins based on resource files. Finally, the visualized data is packaged into fused signal data to achieve expanded platform display and dynamic scene layout. The main function of the above two interfaces is to achieve seamless display of different terminals, allowing customers to freely switch product status data and determine whether there is a fault.

[0103] Example 9:

[0104] See Figure 9: In the interface output, the traditional coal mine needs to manually display interface parameters, and there is no automatic adaptation device, so after the current transformer and the interface are connected, the host computer determines the physical characteristics of the current transformer through the secondary output signal, generates a unique specification identifier, and according to the unique specification identifier, the corresponding interface card template can be matched in the preset end shift. The interface card template predefines the display elements required for different specifications of the transformer, such as waveform diagram, data table, alarm threshold line, and defines data binding logic, such as mapping the transformation ratio parameter to the coordinate axis scale. In the process of template matching, the specification identifier and the template are highly associated through the key-value pair mapping rule engine, which will ensure that the interface elements and the parameters of the current transformer are synchronized in real time.

[0105] Embodiment 10:

[0106] Referring to Figure 10 : In actual implementation, for the problem of complex operation of the system interface interaction of the current transformer in the coal mine, user operation and display result, the application realizes the recognition of user operation and automatic response through the recognition of user intent. In this process, through the visual instruction operation of the touch screen or physical button on the host computer and user gestures, and the corresponding interface elements in the interface, the mapping relationship between operation and response is constructed. Then, the preset rules are parsed through the inheritance rule engine to determine the user's convertible instructions and dynamically generate adjustment instructions. The target instructions output by the rule engine will call the rendering interface to display the abnormal and normal data in the abnormal waveform and the time axis in real time on the new canvas and update the data binding. The above-mentioned fusion interface and signal interface linkage operation can enhance the analysis function of view cooperation.

[0107] Finally, it should be noted that: the above-mentioned embodiments only express several embodiments of the present application, and are not used to limit the present application. For ordinary skilled in the art, any modification, equivalent replacement, improvement, etc. without departing from the concept of the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application patent should be subject to the appended claims.

Claims

1. A current transformer secondary output signal normalization system used in coal mines, characterized in that: It includes a host and a display unit, wherein the host and the display unit are communicatively connected; The host includes multiple signal connection ports and a digital processor; wherein the signal connection port is used to automatically adapt to the secondary output signal of the current transformer and perform signal waveform simulation; the digital processor has a built-in normalization program, and normalizes the secondary output signal through the normalization program; wherein the normalization process of the secondary output signal through the normalization program includes: Extract waveform features from the simulated waveform obtained by simulating the signal waveform and perform waveform feature classification; Classify the secondary output signals for normalization according to waveform characteristics, determine the secondary output signals involved in normalization calculation after quality inspection, and determine the corresponding normalization coefficients; Determine the normalized characteristic parameters of the secondary output signals one by one, and calculate the normalized characteristic parameters in sections for abnormal secondary output signals whose amplitude balance and waveform distortion coefficient are obviously not in a linear relationship under the coordinates; Classifying the waveform characteristics and determining the normalized average values ​​of the normalized characteristic parameters of different secondary output signals; Generate normalized relative fusion curves of different secondary output signals based on the normalized average value; According to the normalized relative fusion curve, the fusion coefficient of different secondary output signals under different operating conditions of the current transformer is determined; And through the fusion coefficient, the different output signals are normalized to determine the normalized value; The display unit includes at least one fusion display interface and N signal display interfaces; wherein the fusion display interface is used to display the fusion signal after normalization processing, and the signal display interface is used to display the secondary output signal.

2. The current transformer secondary output signal normalization system used in coal mines according to claim 1, characterized in that: The waveform simulation includes: Analyze the time domain and frequency domain characteristics of the current transformer secondary output signal to determine the signal modal components that need to be simulated; According to the signal modal components, the corresponding chaotic system is matched and the chaotic signal is generated through numerical integration; According to the characteristic index of the secondary output signal of the current transformer, the amplitude modulation and phase modulation parameters of the chaotic signal are calculated; According to the control parameters, the chaotic signal is amplitude and phase modulated; The modulated signals are synthesized to obtain a simulated current transformer secondary output signal waveform, and the simulated waveform is output.

3. The current transformer secondary output signal normalization system used in coal mines according to claim 1, characterized in that: The signal connection port includes a signal detection unit, a signal analysis unit and a control unit; Wherein, the signal detection unit is used to receive the secondary output signal of the current transformer; The signal analysis unit determines the signal strength based on a preset threshold and determines the port specifications that the signal should match; The control unit adjusts the adaptation signal strength of the signal connection port through a relay matrix or an electronic switch according to the port specifications.

4. The current transformer secondary output signal normalization system used in coal mines according to claim 1, characterized in that: The normalization process of the secondary output signal by a normalization program further includes: Collect the original data of the secondary output signal of the current transformer, including normal signals and abnormal signals; Preprocess the original signal to determine the training corpus; the preprocessing includes: sampling rate conversion, length fixing and data cleaning; Construct a self-supervisory signal normalization supervision model based on statistical instance normalization, train the model using training corpus, and obtain a trained target supervision model; Input the normalized secondary output signal of the current transformer to be verified into the trained target supervision model to determine the enhanced verification signal after enhancement; The effectiveness of the normalization result is judged by comparing the enhanced verification signal with the original signal.

5. The current transformer secondary output signal normalization system used in coal mines according to claim 4, characterized in that: The self-supervisory signal normalization supervision model includes the following construction steps: Collect text description information and device description information of the current transformer secondary signal; wherein the text description information is first associated data of the influence of the signal characteristics on the normalization result; the device description information is second associated data of the influence of the device characteristics of the verified physical object on the normalization result; Constructing first training sample data and second training sample data according to the first associated data and the second associated data; Inputting the first training sample data into a first model substructure based on the semantic vector and outputting a first target vector; Inputting the second training sample data into a second model substructure based on the device parameters and outputting a second target vector; An attribute constraint model is constructed according to the first target vector and the second target vector, a target self-supervision model is trained, and a self-supervision signal normalization supervision model is generated.

6. The current transformer secondary output signal normalization system used in coal mines according to claim 1, characterized in that: The fusion display interface and the signal display interface of the display unit are connected; Each signal display interface and fusion display interface is configured with a unique data transmission channel; The fusion display interface is used to perform waveform processing on the normalized secondary output signal to generate a normalized waveform; The fusion display interface and the signal display interface are connected to the preset visualization tool, and the corresponding display mode is generated according to the control instructions of the visualization tool.

7. The current transformer secondary output signal normalization system used in coal mines according to claim 1, characterized in that: The fusion display interface and the signal display interface are connected to a preset visualization tool, including: Receive visualization instructions from the user and determine the signal display interface that needs to be fused; Obtain raw data from different signal display interfaces and convert them into converted data that conforms to the preset visualization tool data format rules; According to user needs, configure data visualization requirement data; wherein the visualization requirement data includes attribute data of data visualization elements and attribute data of data visualization scene data; Use preset visualization tools to edit the converted data and generate data visualization elements that meet the requirements of attribute data; Use the data visualization engine to edit data visualization elements and generate data visualization scene data that meets the attribute data requirements; According to the data visualization scene data, different signal display interfaces are fused to generate data visualization resource files that conform to the preset visualization tool data format rules, and the visualization resource files are visualized through the fused display interface.

8. The current transformer secondary output signal normalization system used in coal mines according to claim 1, characterized in that: The display unit includes a preset visual interface configuration unit, which is used to dispatch an interface card template according to the specification identification of the secondary signal of the current transformer; Among them, the specification identification is generated by the signal connection port on the host after the current transformer is connected; The interface card template is a preset schedulable template for the fusion display interface and signal display interface.

9. The current transformer secondary output signal normalization system used in coal mines according to claim 1, characterized in that: The display unit further includes an interface response mechanism; wherein the interface response mechanism is used to: Receive user operations on the fusion display interface and the signal display interface; According to the operation behavior, a mapping relationship between the fusion display interface and the signal display interface and the operation behavior is established; According to the mapping relationship, the corresponding target definition instructions are obtained on the preset rule engine; Define instructions based on the target and respond to user operations.

Images