Remote monitoring and alarming method and system for temperature and humidity of power transformation cabinet

Abstract

The invention discloses a power transformation cabinet temperature and humidity remote monitoring and alarming method and system, and relates to the technical field of power transformation cabinet temperature and humidity monitoring and alarming, and the method comprises the following steps: a temperature and humidity monitoring system carries out the continuous monitoring of the temperature and humidity in a power transformation cabinet at a preset sensitivity, and monitors the temperature and humidity fluctuation; and any abnormal condition exceeding a safety range can be found and alarmed in time. By introducing the deep learning model to predict the load change trend, the response capability of the temperature and humidity monitoring system is remarkably improved. A traditional system depends on a fixed alarm threshold value and cannot deal with load fluctuation and rapid change in time, and a deep learning model intelligently predicts a load aggravation trend and generates an abnormal fluctuation index based on parameters such as real-time current waveform distortion and load response time. When the load is abnormal, the system automatically adjusts the alarm sensitivity, quickly captures the temperature and humidity change and triggers the alarm, thereby avoiding the temperature and humidity abnormity problem caused by the load fluctuation, and ensuring the safe operation of the equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of temperature and humidity monitoring and alarming for substation cabinets, and particularly relates to a method and system for remote monitoring and alarming of the temperature and humidity of substation cabinets. Background Art

[0002] Remote monitoring and alarming of the temperature and humidity of substation cabinets means connecting temperature and humidity sensors to a remote monitoring system to achieve real-time monitoring of the temperature and humidity inside the substation cabinets. Substation cabinets are usually used for centralized management and control of power equipment. The internal environment requires the temperature and humidity to be maintained within a certain range to ensure the normal operation of the equipment and avoid failures. Through remote monitoring, operators can view the temperature and humidity data inside the substation cabinets in real time, and when abnormal temperature and humidity are detected, the system automatically sends an alarm signal. This can early warn of potential equipment failures or environmental problems. For example, too high a temperature may cause the equipment to overheat and be damaged, and too high a humidity may trigger electrical failures. Through remote monitoring and alarming of temperature and humidity, the operation safety of the equipment can be improved, the probability of failures can be reduced, and the stable operation of the power system can be ensured.

[0003] The remote monitoring and alarming system for the temperature and humidity of substation cabinets usually uses a temperature and humidity monitoring system. The temperature and humidity monitoring system plays a crucial role in this process. It is responsible for collecting the temperature and humidity data inside the substation cabinets in real time and transmitting these data to the remote monitoring platform through sensors. The system can continuously monitor environmental changes within the set threshold range and automatically alarm for abnormal fluctuations (such as too high a temperature or too high a humidity). When the temperature and humidity inside the substation cabinets exceed the preset range, the monitoring system triggers an alarm mechanism to notify the operator or management platform to take intervention measures in a timely manner. This can not only prevent potential safety hazards such as overheating, corrosion, and electrical failures caused by abnormal temperature and humidity of the equipment, but also achieve 24-hour uninterrupted monitoring through the remote monitoring system, improving the operation safety and reliability of the substation equipment. Through remote monitoring and real-time alarming, the temperature and humidity monitoring system effectively reduces human negligence, improves the equipment's fault prevention ability, and ensures the safe operation of the substation equipment.

[0004] The prior art has the following deficiencies: In many temperature and humidity monitoring systems, the alarm threshold is usually preset (i.e., fixed sensitivity), that is, a fixed temperature and humidity range is set during the design. When the environmental data exceeds this range, the system will trigger an alarm. Although this static threshold setting is effective in a stable environment, there may be certain hidden dangers when facing dynamic and instantaneous changes. Especially in the case of large load fluctuations, the temperature and humidity inside the substation cabinet may change rapidly. For example, when the load suddenly increases, the heat generated by electrical equipment will cause the temperature to rise rapidly, and the humidity may also fluctuate sharply. However, due to the lag in the response of sensors and monitoring systems, and the alarm threshold is usually static, the system fails to adjust the threshold in time to cope with these rapid changes. This may lead to the sharp fluctuations of temperature and humidity exceeding the normal range. However, because the system fails to recognize this change in a short time or fails to dynamically adjust the alarm threshold, the alarm fails to be triggered in time, thus missing the best response opportunity. As a result, temperature and humidity problems may not be detected in time at the initial stage, which may cause equipment overheating, corrosion or failure, and may even seriously affect the safe operation of substation equipment.

[0005] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] The object of the present invention is to provide a method and system for remote monitoring and alarming of the temperature and humidity of a substation cabinet. By introducing a deep learning model for intelligent prediction of the load change trend, the dynamic response ability of the temperature and humidity monitoring system is significantly improved. Traditional systems rely on fixed alarm thresholds, which are difficult to cope with load fluctuations and rapid changes, and are prone to missing the alarm opportunity. The deep learning model predicts the load intensification trend based on parameters such as real-time current waveform distortion and load response time, and generates a load abnormal fluctuation index. When an abnormal load is detected, the system will intelligently adjust the alarm sensitivity, timely capture the temperature and humidity changes and trigger an alarm. This dynamic adjustment mechanism effectively avoids the temperature and humidity abnormality problems caused by load fluctuations, and ensures the safe operation of equipment, so as to solve the problems in the above background art.

[0007] To achieve the above object, the present invention provides the following technical solutions: A method for remote monitoring and alarming of the temperature and humidity of a substation cabinet, comprising the following steps: The temperature and humidity monitoring system continuously monitors the temperature and humidity inside the substation cabinet with a preset sensitivity, monitors the temperature and humidity fluctuations, and timely discovers and alarms any abnormal conditions exceeding the safe range; While monitoring the temperature and humidity, the operation parameter information of the substation cabinet is collected in real time, and the working state and load changes of the substation cabinet are reflected through the collected data, providing a data basis for subsequent load trend analysis; Preprocess the operation parameter data collected in real time, extract the key features reflecting the load change trend of the substation cabinet from the preprocessed operation parameter data, and analyze the extracted features under the detection window to quantify and understand the specific trend and pattern of the load change of the substation cabinet; Input the quantified load change trend features into a pre-trained deep learning model, and use the intelligent ability of the model to intelligently predict the future load change trend of the substation; According to the prediction results of the deep learning model, divide the load change trend of the substation cabinet into balanced power load and intensified power load; For the case of balanced power load, the temperature and humidity monitoring system continues to monitor with a preset sensitivity to ensure that the temperature and humidity are maintained within a safe range and prevent equipment failures caused by excessive temperature and humidity; For the case of intensified power load, based on the prediction results of the deep learning model, dynamically increase the sensitivity of temperature and humidity monitoring, keenly capture the changes in temperature and humidity, and trigger an alarm in a timely manner.

[0008] Preferably, extract the key features reflecting the load change trend of the substation cabinet from the preprocessed operation parameter data. The extracted parameters include the response speed of the equipment in the substation cabinet to load changes and the degree of distortion of the current signal. Under the detection window, analyze the response speed of the equipment in the substation cabinet to load changes and the degree of distortion of the current signal, and generate the instantaneous load response time reference value and the current waveform distortion reference value respectively. Quantify and understand the specific trend and pattern of the load change of the substation cabinet through the instantaneous load response time reference value and the current waveform distortion reference value.

[0009] Preferably, the specific steps for analyzing the response speed of the equipment in the substation cabinet to load changes under the detection window to generate the instantaneous load response time reference value are as follows: First, collect the operation parameters of various equipment in the substation cabinet in real time, and preprocess the obtained data. Through the preprocessed data, evaluate the rate and amplitude of load changes, so as to analyze the impact of load changes on equipment performance. The calculation expression of the load change rate is as follows: , where, is the change in load power, is the measurement duration of the load power change, is the load change rate; After obtaining the load change rate , establish an instantaneous load response model for the equipment. The speed of equipment response determines the instantaneous load response time. By analyzing the relationship between the load change rate and the equipment state, establish a dynamic response function, and calculate the instantaneous load response time of the equipment based on the characteristics of load changes. The calculation expression is as follows: , In the formula, is the instantaneous load response time of the device, is the load response proportionality coefficient, is the non-linear exponent of the load change, is the power factor, is the current, is the power factor adjustment coefficient, is the non-linear exponent of the power factor and the current, used to control the influence degree of the non-linear relationship on the response time; Based on the load change rate and the instantaneous load response time of the device, a reference value of the instantaneous load response time is generated, and the generation formula is as follows: , In the formula, is the natural base, is the instantaneous power output of the device, is the current load attenuation coefficient, is the weight coefficient of the ratio of power to current, is the exponential adjustment parameter, is the reference value of the instantaneous load response time.

[0010] Preferably, the specific steps for analyzing the distortion degree of the current signal under the detection window to generate a reference value of the current waveform distortion are as follows: First, the current signal is decomposed into high-frequency and low-frequency components, and the distortion characteristics in the signal are deeply analyzed. Through the fast Fourier transform, the current signal is converted into a frequency-domain signal, and the formula after the frequency-domain representation of the current signal is generated as: , In the formula, the frequency-domain representation of the signal, representing the amplitude and phase information of the current signal in the frequency domain, covering all frequency components, is the frequency, is the original current signal, describing the intensity and waveform of the current at any time t, is the natural base, is the imaginary unit, is the frequency of the signal, is the time variable; To extract the high-frequency and low-frequency components from the frequency-domain signal, the frequency-domain signal is segmented by the frequency threshold : If the frequency range is within the range, representing the stable and periodic part of the signal, it is labeled as the low-frequency component of the frequency-domain signal; If the frequency range within which corresponds to the rapid changes, noise, and nonlinear effects in the signal, it is calibrated as the high-frequency component of the frequency-domain signal ; After splitting the high-frequency and low-frequency components of the frequency-domain signal, energy analysis is used to measure the distortion degree of the current waveform. To accurately measure the amplitude of the distortion, the distortion energy index of the current signal is calculated, and the calculation expression is as follows: , In the formula, is the distortion energy index of the current waveform, is the high-frequency frequency range of the frequency-domain signal, is the lower limit of the high-frequency range, is the upper limit of the high-frequency range, is the frequency weighting factor; Based on the energy of the high-frequency component and the low-frequency stable part, through weighted synthesis, a reference value for the distortion of the current waveform is generated to quantify the overall distortion degree of the current waveform. The generation formula is as follows: , In the formula, is the low-frequency frequency range of the frequency-domain signal, is the lower limit of the low-frequency range, is the upper limit of the low-frequency range, is the weight coefficient, is the reference value for the distortion of the current waveform.

[0011] Preferably, the generated instantaneous load response time reference value and the reference value for the distortion of the current waveform after analysis are input into a pre-trained deep learning model, and based on the deep learning model, a load abnormal fluctuation index is generated to intelligently predict the future load change trend of the substation.

[0012] Preferably, when intelligently predicting the future load change trend of the substation through the intelligent ability of a pre-trained deep learning model, the generated load abnormal fluctuation index is compared and analyzed with a pre-set reference threshold for the load abnormal fluctuation index, and the load change trend of the switch cabinet is classified. The classification steps are as follows: If the load abnormal fluctuation index is greater than the pre-set reference threshold for the load abnormal fluctuation index, the load change trend of the switch cabinet is classified as an increase in power load; If the load abnormal fluctuation index is less than or equal to the pre-set reference threshold for the load abnormal fluctuation index, the load change trend of the switch cabinet is classified as a balanced power load.

[0013] Preferably, for the situation of intensified power load, based on the prediction results of the deep learning model, the specific steps to dynamically increase the sensitivity of temperature and humidity monitoring, keenly capture the changes in temperature and humidity, and trigger an alarm in a timely manner are as follows: After the load change trend of the switch cabinet is classified as an intensified power load, at this time, adjust the sensitivity of the temperature and humidity monitoring system. Determine how to increase the sensitivity by calculating the dynamic sensitivity adjustment coefficient, so as to ensure that the sensitivity of the temperature and humidity monitoring system is effectively adjusted according to the intensified load situation. The calculation expression is as follows: , In the formula, is the load abnormal fluctuation index, is the reference threshold of the load abnormal fluctuation index, is the maximum value of the load abnormal fluctuation index, and are both non-linear adjustment parameters, used to control the load abnormal fluctuation index For the reference threshold of the non-linear amplification degree, is used to control the load abnormal fluctuation index For the maximum value of the load abnormal fluctuation index of the non-linear influence, is the dynamic sensitivity adjustment coefficient; According to the dynamic sensitivity adjustment coefficient Adjust the sensitivity of the temperature and humidity monitoring system to capture the change trend of the temperature and humidity in the switch cabinet with higher sensitivity. When the rapid rise of temperature and humidity is captured, the temperature and humidity monitoring system immediately issues a warning. The calculation expression is as follows: , In the formula, is the adjusted temperature and humidity monitoring sensitivity, is the preset basic sensitivity, is the adjustment coefficient, is the load abnormal fluctuation index change rate, is the adjustment coefficient.

[0014] A remote monitoring and alarm system for the temperature and humidity of a switch cabinet includes a temperature and humidity monitoring module, an operating parameter acquisition module, a data preprocessing and feature extraction module, a deep learning prediction module, a load change trend classification module, a load balance monitoring module, and a dynamic sensitivity adjustment and alarm module: The temperature and humidity monitoring module. The temperature and humidity monitoring system continuously monitors the temperature and humidity inside the switch cabinet with a preset sensitivity, monitors the temperature and humidity fluctuations, and promptly discovers and alarms any abnormal conditions beyond the safe range; The operating parameter acquisition module collects the operating parameter information of the switch cabinet in real time while monitoring the temperature and humidity. The data collected reflects the working state and load changes of the switch cabinet, providing a data basis for subsequent load trend analysis; The data preprocessing and feature extraction module preprocesses the real-time collected operating parameter data, extracts key features reflecting the load change trend of the switch cabinet from the preprocessed operating parameter data, and analyzes the extracted features under the detection window to quantify and understand the specific trends and patterns of the switch cabinet load changes; The deep learning prediction module inputs the quantified load change trend features into a pre-trained deep learning model, and uses the intelligent capabilities of the model to make an intelligent prediction of the future load change trend of the substation; The load change trend classification module divides the load change trend of the switch cabinet into balanced power load and increased power load according to the prediction results of the deep learning model; The load balance monitoring module, for the case of balanced power load, the temperature and humidity monitoring system continues to monitor with a preset sensitivity to ensure that the temperature and humidity are maintained within a safe range and prevent equipment failures caused by excessive temperature and humidity; The dynamic sensitivity adjustment and alarm module, for the case of increased power load, based on the prediction results of the deep learning model, dynamically increases the sensitivity of temperature and humidity monitoring, keenly captures the changes in temperature and humidity, and triggers an alarm in a timely manner.

[0015] In the above technical solution, the technical effects and advantages provided by the present invention are as follows: Through the accurate prediction of the load change trend of the switch cabinet by the deep learning model, the present invention optimizes the load management and the monitoring of the equipment operating state. By collecting operating parameters in real time and extracting key features, the deep learning model can intelligently analyze the load change trend and predict future load changes based on historical data and real-time parameters. This intelligent prediction can not only identify whether the load is increasing, but also quantify the change pattern of the switch cabinet load, so as to reasonably plan and manage the equipment load. When the load changes greatly, the system will dynamically adjust the alarm threshold or monitoring strategy to ensure that the equipment is still in a safe working state during load fluctuations. This optimization measure can greatly improve the operating stability of the equipment, reduce the risk of failures caused by excessive or uneven load, and at the same time improve the energy efficiency and long-term stable operation ability of the switch cabinet. Description of the Drawings

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0017] Figure 1 This is the method flow chart of a method for remote monitoring and alarming of temperature and humidity in a substation cabinet according to the present invention.

[0018] Figure 2 This is the module schematic diagram of a system for remote monitoring and alarming of temperature and humidity in a substation cabinet according to the present invention. Specific embodiments

[0019] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that this disclosure will be more complete and thorough, and will fully convey the concept of the example embodiments to those skilled in the art.

[0020] The present invention provides a method for remote monitoring and alarming of temperature and humidity in a substation cabinet as shown in Figure 1 the following steps: The temperature and humidity monitoring system continuously monitors the temperature and humidity inside the substation cabinet with a preset sensitivity, monitors the temperature and humidity fluctuations, and promptly discovers and alarms any abnormal conditions beyond the safe range. The setting of the preset sensitivity is based on an in-depth analysis of the conventional operating environment of the substation cabinet, taking into account the heat dissipation capacity, ventilation conditions of the equipment, and the temperature and humidity change range in historical data. The system collects the data inside the substation cabinet in real time through high-precision temperature and humidity sensors and compares it with the preset thresholds. When the temperature or humidity exceeds the preset range, the system immediately triggers an alarm to notify the operation and maintenance personnel to take corresponding measures. This initial monitoring step ensures that the system can operate stably when the load is balanced and guarantees the normal working environment of the equipment.

[0021] While monitoring the temperature and humidity, the operation parameter information of the substation cabinet is collected in real time, and the working state and load changes of the substation cabinet are reflected through the collected data, providing a data basis for subsequent load trend analysis. The collection of operation parameters is completed by a variety of sensors and measuring devices installed inside the substation cabinet, and these devices can monitor the working state of electrical equipment in real time. For example, current sensors monitor the current changes of the equipment, voltage sensors record voltage fluctuations, and power sensors measure the total power output, etc. These data are transmitted to the data collection module of the monitoring system through wired or wireless communication methods to ensure the real-time and accuracy of the data. Obtaining these operation parameter information in real time helps to comprehensively understand the working state of the substation cabinet and provides key support for subsequent prediction of load change trends.

[0022] Preprocess the operation parameter data collected in real time, extract the key features reflecting the load change trend of the switchgear from the preprocessed operation parameter data, and analyze the extracted features under the detection window to quantify and understand the specific trends and patterns of the switchgear load change; The preprocessing steps include operations such as data cleaning, denoising, missing value filling, and standardization. First, the system detects and removes outliers and noise data to reduce the impact of data errors on subsequent analysis. Then, for missing data points, interpolation or other filling methods are used to complete them to ensure the integrity of the data. Finally, the data is standardized to make data with different dimensions have the same scale, facilitating subsequent feature extraction and model training. The preprocessed data is cleaner and more consistent, providing a high-quality data foundation for feature extraction and deep learning models, improving the accuracy and reliability of predictions.

[0023] Extract the key features reflecting the load change trend of the switchgear from the preprocessed operation parameter data. The extracted parameters include the response speed of the equipment in the switchgear to load changes and the degree of distortion of the current signal. Under the detection window, analyze the response speed of the equipment in the switchgear to load changes and the degree of distortion of the current signal, and generate the instantaneous load response time reference value and the current waveform distortion reference value respectively. Quantify and understand the specific trends and patterns of the switchgear load change through the instantaneous load response time reference value and the current waveform distortion reference value.

[0024] When the response speed of the equipment in the switchgear to load changes slows down, it usually indicates that the power load of the distribution cabinet shows an increasing trend, resulting in rapid changes in temperature and humidity. The reason is that when the load of the distribution cabinet increases, the electrical equipment needs to process larger currents and powers, which will cause changes in the electrical characteristics of the equipment. For example, when the load increases, the working pressure of the transformer, switchgear, and other electrical components in the switchgear increases, and the generated heat will rise significantly. The heat conduction and heat dissipation capabilities of the equipment itself will also be affected by the increased load, resulting in a rapid increase in temperature. In addition, the increased load may cause current instability, resulting in large power fluctuations, further increasing the heat accumulation and humidity changes inside the equipment. In this case, the lag in equipment response (i.e., the slowdown in the response speed of the equipment to load changes) is often due to the failure of the load and heat dissipation capabilities of the electrical equipment to be adjusted synchronously. Due to the lag in heat release of the equipment, the temperature and humidity changes will not be immediately reflected, but over time, the temperature and humidity inside the switchgear will rise sharply, resulting in abnormal fluctuations in the monitored temperature and humidity values. At this time, if the temperature and humidity monitoring system fails to identify this change in time, the alarm opportunity may be missed, affecting the normal operation of the switchgear and even potentially causing equipment failure or damage.

[0025] The specific steps for analyzing the response speed of equipment in the substation cabinet to load changes and generating the instantaneous load response time reference value under the detection window are as follows: First, collect various equipment operation parameters in the substation cabinet in real time, such as load, current, power factor, etc., and preprocess the obtained data. Data preprocessing includes removing noise, filling missing values, smoothing processing, etc., to ensure the accuracy of subsequent analysis. Through the preprocessed data, evaluate the rate and amplitude of load changes, so as to analyze the impact of load changes on equipment performance. The calculation expression of the load change rate is as follows: , In the formula, is the change in load power, is the measurement duration of the load power change, is the load change rate; In this step, the load change rate is used to preliminarily evaluate the speed of load fluctuations. Since the response of the equipment is not instantaneous, the change rate will reflect the initial impact of the sudden increase in load on the equipment, and further affect the dynamic characteristics of temperature and humidity changes.

[0026] Obtain the load change rate After that, establish an equipment instantaneous load response model. This model describes the response ability of the equipment to load changes. The speed of the equipment response determines the instantaneous load response time. By analyzing the relationship between the load change rate and the equipment state (such as temperature, humidity, power factor, etc.), establish a dynamic response function. Based on the characteristics of load changes, calculate the equipment instantaneous load response time. The calculation expression is as follows: , In the formula, is the equipment instantaneous load response time, that is, the reaction delay of the equipment to load changes, is the load response proportionality coefficient, which reflects the sensitivity of the equipment to load changes, is the non-linear exponent of load change, which is used to adjust the influence weight of the load change rate on the response time, is the power factor, which is the ratio of the reactive power of the equipment load to the apparent power, and the value range is 0-1, reflecting the efficiency of the load and the stability of equipment operation, is the current, indicating the current magnitude of the equipment, which directly reflects the current demand of the load on the equipment, is the power factor adjustment coefficient, which controls the influence degree of the power factor on the response time, is the non-linear exponent of the power factor and current, which is used to control the influence degree of the non-linear relationship of This step accurately simulates the response delay of the device to load changes by combining parameters such as the load change rate, power factor, and current. The greater the load change rate, the longer the response time, indicating that the load intensifies and may cause a lag in temperature and humidity changes.

[0027] Based on the load change rate and the instantaneous load response time of the device of the modeling results, a reference value for the instantaneous load response time is generated. This reference value reflects the response ability of the device and can predict the change trend of temperature and humidity when the load intensifies. When the load intensifies, the response ability of the device weakens, resulting in a lag in temperature and humidity changes. On the contrary, when the load is balanced, the response speed is faster and the temperature and humidity fluctuations are smaller. The generation formula is as follows: , In the formula, is the natural base, is the instantaneous power output of the device, is the current load attenuation coefficient, which reflects the current load ratio 's influence degree on the instantaneous load response time, is the weight coefficient of the power-to-current ratio, which reflects the power-to-current ratio 's influence weight on the instantaneous load response time, is the exponential adjustment parameter, which affects the non-linear effect of the power-to-current ratio . is the reference value of the instantaneous load response time.

[0028] This step comprehensively considers the influence of the changes in load and current on load response and adjusts the dynamic characteristics of the device response through an exponential function. It can accurately generate a reference value for the instantaneous load response time to help the system dynamically evaluate the change trend of temperature and humidity when the load changes. If the reference value is large, it means that the load intensifies and the device response lags, and the temperature and humidity fluctuations may be relatively severe; if the reference value is small, it indicates that the load is balanced and the temperature and humidity fluctuations are relatively stable.

[0029] The larger the reference value of the instantaneous load response time generated after analyzing the response speed of the equipment in the power distribution cabinet to load changes under the detection window, the more pronounced the trend of the power load in the power distribution cabinet, which in turn leads to rapid changes in temperature and humidity. The instantaneous load response time refers to the response speed of the equipment in the power distribution cabinet to load changes, reflecting the lag in heat release and heat dissipation capacity of the equipment when the load changes. When the power load intensifies, the load-bearing capacity of the equipment approaches the limit, and the lag effect of temperature rise and heat accumulation caused by load changes is obvious. Therefore, the response speed of the equipment to load changes will slow down, and the measured value of the instantaneous load response time will increase. This means that although the load change has occurred, the fluctuation of temperature and humidity will still be delayed. However, once the fluctuation starts, the change will be relatively rapid and intense, resulting in a sharp increase in the temperature and humidity inside the equipment. When the measured value of the instantaneous load response time is small, it usually means that the load in the power distribution cabinet is in a balanced state, the equipment can quickly respond to load changes, and the temperature and humidity fluctuations are relatively stable, without causing sharp changes in temperature and humidity.

[0030] The distortion of the current signal waveform usually indicates that the power load in the power distribution cabinet is intensifying, leading to rapid changes in the temperature and humidity of the power distribution cabinet. When the power load suddenly increases or shows unstable fluctuations, the current signal in the power distribution cabinet will be distorted, manifested as waveform distortion or abnormality. This distortion is usually caused by the following factors: current pulses caused by sudden load increases, instantaneous overloads when the equipment load changes, and harmonic generation in the power system. When the load intensifies, the electrical equipment inside the power distribution cabinet needs to consume more electrical energy, resulting in more heat being released and the temperature rising sharply. At the same time, due to the limited sealing and ventilation conditions of the power distribution cabinet, the humidity will also change with the increase in temperature. Especially in a relatively humid environment, the sharp fluctuation of humidity may pose a threat to the safety of the equipment. As an early signal of the intensification of the power load, the distortion of the current waveform can be used to predict the abnormal temperature and humidity caused by load fluctuations by monitoring this change, and adjust the sensitivity of the temperature and humidity monitoring system in advance to prevent equipment failures due to overheating or abnormal humidity. Therefore, the distortion of the current signal waveform is not only an indicator of the intensification of the power load, but also indirectly reflects that the temperature and humidity may change rapidly, affecting the safe operation of the equipment.

[0031] The specific steps for analyzing the distortion degree of the current signal under the detection window to generate the current waveform distortion reference value are as follows: First, decompose the current signal into high-frequency and low-frequency components, deeply analyze the distortion characteristics in the signal. Through the fast Fourier transform, convert the current signal into a frequency-domain signal, and pay special attention to the high-frequency components (such as harmonics, transient fluctuations) and low-frequency components (such as fundamental current) in the signal. The formula for generating the frequency-domain representation of the current signal is: , In the formula, Frequency-domain representation of the signal, which represents the amplitude and phase information of the current signal in the frequency domain and covers all frequency components. is the frequency. is the original current signal, which describes the intensity and waveform of the current at any time in time. is the natural base. is the imaginary unit. is the frequency of the signal. is the time variable; To extract the high-frequency and low-frequency components from the frequency-domain signal, the frequency-domain signal is segmented by a frequency threshold as follows: If the frequency range represents the stable and periodic part of the signal, it is labeled as the low-frequency component of the frequency-domain signal ; If the frequency range corresponds to the fast-changing, noise, and non-linear effects in the signal, it is labeled as the high-frequency component of the frequency-domain signal ; The high-frequency component reflects the instantaneous fluctuations of the current waveform, while the low-frequency component represents the smooth changes of the load. The high-frequency component is usually related to power load fluctuations, overload phenomena, and harmonics, while the low-frequency component corresponds to normal load fluctuations.

[0032] After segmenting the high-frequency and low-frequency components of the frequency-domain signal, energy analysis is used to measure the distortion degree of the current waveform. To accurately measure the amplitude of the distortion, the distortion energy index of the current signal is calculated, and the calculation formula is as follows: , where, is the distortion energy index of the current waveform, which represents the cumulative energy of the high-frequency distorted components in the current signal, is the high-frequency frequency range of the frequency-domain signal, is the lower limit of the high-frequency range, is the upper limit of the high-frequency range, is the frequency weighting factor, which is used to assign different weights to high-frequency components of different frequencies; By summing the energies of the high-frequency components, the distortion energy of the current signal is quantified. The greater the distortion energy, the more significant the distortion of the current waveform, which may indicate an increase in the load. This step captures the non-linear fluctuations and harmonic interference of the current waveform when the load increases by emphasizing the energy of the high-frequency components.

[0033] Based on the energy of the high-frequency components and the low-frequency stable part, through weighted synthesis, a reference value for the distortion of the current waveform is generated to quantify the overall distortion degree of the current waveform. The generation formula is as follows: , wherein, is the low - frequency frequency range of the frequency - domain signal, is the lower limit of the low - frequency range, is the upper limit of the low - frequency range, is the weight coefficient, which is used to balance the influence of the high - frequency and low - frequency energy ratio on the reference value of the overall current waveform distortion, is the reference value of the current waveform distortion.

[0034] In this step, the numerator is the energy of the high - frequency signal, and the denominator is the energy of the low - frequency signal. After forming the ratio, it reflects the intensity of the signal distortion. If this ratio is large, it indicates that the energy of the high - frequency component has increased significantly, which may mean that the load fluctuation has intensified, resulting in serious distortion of the current waveform. Finally, can be generated by quantifying the ratio of the high - frequency energy to the low - frequency stable part, accurately reflecting the distortion degree of the current waveform, and providing an accurate alarm - triggering basis for the temperature - humidity monitoring system.

[0035] The larger the reference value of the current waveform distortion generated after analyzing the distortion degree of the current signal under the detection window, usually means that there is an unstable or sudden increase trend in the power load. This change is often accompanied by phenomena such as instantaneous overload of the equipment load and harmonic fluctuations in the power system, indicating that the load state of the electrical equipment is intensifying, resulting in a sharp rise in the heat inside the switchgear cabinet. As the temperature rises, the humidity may also change accordingly, especially in a humid environment, and these changes may quickly affect the safety of the equipment. Therefore, the increase in the reference value of the current waveform distortion is not only an early signal of the intensification of the power load, but also means that the rapid change of temperature and humidity may cause equipment failures. When the reference value of the current waveform distortion is small and stable, it indicates that the power load is maintained in a balanced state, the load change is small, and the temperature - humidity fluctuation is relatively stable.

[0036] Input the quantified load - change trend characteristics into a pre - trained deep - learning model, and use the intelligent ability of the model to make an intelligent prediction of the future load - change trend of the substation; Input the generated instantaneous load response - time reference value and the reference value of the current waveform distortion after analysis into a pre - learned deep - learning model, generate a load abnormal - fluctuation index based on the deep - learning model, and make an intelligent prediction of the future load - change trend of the substation through the abnormal - fluctuation index.

[0037] A pre-trained deep learning model refers to a deep learning model that has been trained with a large amount of historical data and can effectively identify and predict load change patterns. During the training phase, this model has learned the complex patterns of load changes in the switchgear cabinet, especially those load fluctuation characteristics that may lead to abnormal temperature and humidity. Specifically, the model is trained with a large amount of historical load data (including instantaneous load response time, current waveform distortion, etc.) to optimize its parameters, enabling it to accurately predict future load fluctuations when real-time data is input. During the training process, deep learning algorithms (such as convolutional neural network CNN, long short-term memory network LSTM, or other regression models) will identify the complex relationships in load changes and learn how to infer future load fluctuations based on features such as instantaneous load response time and current waveform distortion.

[0038] The key advantage of this pre-trained deep learning model is its ability to adapt to the complex dynamic patterns and non-linear characteristics in the operation of the switchgear cabinet. The load of the switchgear cabinet does not show simple linear fluctuations. In many cases, it is affected by multiple factors, including grid load changes, the working state of electrical equipment, sudden increases or decreases in load, etc. By learning the patterns in this historical data, the deep learning model can capture the potential factors behind load fluctuations, even abnormal changes that cannot be accurately captured by traditional methods. For example, when there are mutations in current waveform distortion or instantaneous load response time, the deep learning model can quickly identify these minor changes and predict possible abnormal fluctuations in future loads. By generating a load abnormal fluctuation index, this model can not only monitor the load status in real-time but also give early warnings of possible load overloading or power quality problems, enabling the temperature and humidity monitoring system to timely adjust the monitoring sensitivity and avoid equipment overheating, corrosion, or other failures caused by abnormal loads. This intelligent prediction ability greatly improves the accuracy and response speed of switchgear cabinet monitoring. Especially when the load changes violently, it can provide more powerful guarantees for the safe operation of the equipment.

[0039] The deep learning model is not limited here. Any deep learning model that can realize comprehensive analysis of the reference value of instantaneous load response time and the reference value of current waveform distortion to generate a load abnormal fluctuation index is acceptable. To implement the technical solution of the present invention, the present invention provides a specific implementation method; Load abnormal fluctuation index The generation formula is as follows: , In the formula, , are the preset proportionality coefficients of the reference value of instantaneous load response time and the reference value of current waveform distortion respectively, and , are both greater than 0.

[0040] From the load abnormal fluctuation index, the larger the instantaneous load response time reference value generated by analyzing the response speed of the equipment in the switch cabinet to load changes under the detection window, and the larger the current waveform distortion reference value generated by analyzing the distortion degree of the current signal under the detection window, it indicates that the larger the load abnormal fluctuation index generated by the intelligent ability of the pre-trained deep learning model to predict the future load change trend of the substation, which means the greater the probability that the load increase in the switch cabinet causes a large change in temperature and humidity. On the contrary, it means the smaller the probability that the load increase in the switch cabinet causes a large change in temperature and humidity.

[0041] According to the prediction result of the deep learning model, the load change trend of the switch cabinet is divided into balanced power load and intensified power load; Compare and analyze the load abnormal fluctuation index generated by the intelligent ability of the pre-trained deep learning model to predict the future load change trend of the substation with the preset load abnormal fluctuation index reference threshold, and divide the load change trend of the switch cabinet. The division steps are as follows: If the load abnormal fluctuation index is greater than the preset load abnormal fluctuation index reference threshold, then divide the load change trend of the switch cabinet into intensified power load; If the load abnormal fluctuation index is less than or equal to the preset load abnormal fluctuation index reference threshold, then divide the load change trend of the switch cabinet into balanced power load.

[0042] Balanced power load means that the power load in the switch cabinet remains relatively stable for a period of time, that is, the fluctuation range of the load is small, and the operation of electrical equipment will not generate excessive heat accumulation or cause overload of the electrical system. Intensified power load means that the load fluctuates violently or suddenly increases, usually resulting in more heat generated by electrical equipment, which will cause a sharp rise in the temperature inside the switch cabinet.

[0043] For the case of balanced power load, the temperature and humidity monitoring system continues to monitor with the preset sensitivity to ensure that the temperature and humidity are maintained within the safe range and prevent equipment failures caused by too high temperature and humidity; In the case of balanced power load, the temperature and humidity monitoring system continues to monitor with a preset sensitivity, aiming to maintain the stability of the system and ensure that there are no abnormal temperature and humidity conditions for the equipment under normal operating conditions. Monitoring with a fixed sensitivity can effectively avoid excessive false alarms, ensuring that the system triggers an alarm only when the real temperature and humidity fluctuations exceed the safe range. This helps to reduce the intervention burden on the operation and maintenance personnel. At the same time, due to the small load fluctuations, the temperature and humidity changes in the switchgear are usually within a controllable range. Therefore, using a fixed sensitivity is sufficient to ensure that the temperature and humidity are always maintained at a safe value, thereby preventing equipment failures caused by excessive temperature or humidity, such as overheating, electrical failures or corrosion caused by moisture, etc., and ensuring the normal operation of the equipment and extending its service life.

[0044] For the case of increased power load, based on the prediction results of the deep learning model, dynamically increase the sensitivity of temperature and humidity monitoring, keenly capture the changes in temperature and humidity, and trigger an alarm in a timely manner; For the case of increased power load, based on the prediction results of the deep learning model, the specific steps to dynamically increase the sensitivity of temperature and humidity monitoring, keenly capture the changes in temperature and humidity, and trigger an alarm in a timely manner are as follows: After the load change trend of the switchgear is classified as increased power load, at this time, adjust the sensitivity of the temperature and humidity monitoring system, and determine how to increase the sensitivity by calculating the dynamic sensitivity adjustment coefficient, so as to ensure that the sensitivity of the temperature and humidity monitoring system is effectively adjusted according to the increased load situation. The calculation expression is as follows: , In the formula, is the load abnormal fluctuation index, is the reference threshold of the load abnormal fluctuation index, is the maximum value of the load abnormal fluctuation index, based on historical data or the maximum fluctuation value preset by the system, and are both non-linear adjustment parameters, used to control the load abnormal fluctuation index for the non-linear amplification degree of the reference threshold , is used to control the non-linear influence of the load abnormal fluctuation index on the maximum value of the load abnormal fluctuation index, is the dynamic sensitivity adjustment coefficient; This step calculates the adjustment coefficient through the deviation between the load abnormal fluctuation index and the reference threshold of the load abnormal fluctuation index. When the deviation of the load abnormal fluctuation index is larger, the sensitivity of the system will be increased accordingly. At the same time, Further enhanced the non-linear characteristics of sensitivity adjustment, making the improvement of sensitivity more significant under the condition of aggravated extreme load.

[0045] According to the dynamic sensitivity adjustment coefficient Adjust the sensitivity of the temperature and humidity monitoring system to capture the changing trend of the temperature and humidity in the substation cabinet with higher sensitivity. When the rapid rise of temperature and humidity is captured, the temperature and humidity monitoring system immediately issues a warning to remind the operator that the temperature and humidity are in a rapid change trend. In this way, when the load intensifies, the temperature and humidity monitoring system not only provides higher-sensitivity monitoring but also effectively issues a warning in advance, thus avoiding equipment damage caused by reaction lag. The calculation expression is as follows: , In the formula, is the adjusted temperature and humidity monitoring sensitivity, representing the sensitivity actually applied by the system when the load intensifies, is the preset basic sensitivity, is the adjustment coefficient, controlling the influence degree of the load change speed on the sensitivity adjustment, is the load abnormal fluctuation index Change rate, reflecting the acceleration of the load change, is the adjustment coefficient, controlling the amplification ratio of the sensitivity with the aggravation of the load change.

[0046] The increase in sensitivity is based on the degree of load aggravation and the change of the load abnormal fluctuation index . When the load intensifies, the sensitivity will gradually increase, enabling the temperature and humidity monitoring system to make a more sensitive response to the temperature and humidity fluctuations and promptly discover and respond to the abnormal changes in temperature and humidity. At the same time, when the adjusted sensitivity reaches or exceeds the preset warning threshold, the system will immediately trigger a warning to notify the operator of the possible risk of excessive temperature and humidity and prompt the adoption of necessary intervention measures. Through this mechanism, not only can it ensure that the substation cabinet equipment can continuously operate safely in the dynamic load environment, but also it can provide sufficient response time for the operation and maintenance personnel to effectively prevent equipment failures or potential safety hazards caused by abnormal temperature and humidity.

[0047] In the case of increased power load, the sensitivity of temperature and humidity monitoring is dynamically improved based on the prediction results of the deep learning model. The purpose is to ensure that the temperature and humidity monitoring system can promptly identify and respond to drastic changes in temperature and humidity in an environment with drastic load fluctuations. The increase in load is usually accompanied by a sharp increase in the heat generated by electrical equipment. The temperature in the substation cabinet may rise rapidly, and the humidity may also fluctuate greatly due to local temperature differences or condensation. In this case, the traditional fixed sensitivity monitoring system may not be able to sensitively capture such rapid changes in temperature and humidity, resulting in delayed alarms and missing the best time to respond, which in turn causes equipment overheating, corrosion or other failures, which may seriously affect the normal operation of the substation cabinet.

[0048] Through the intelligent prediction of the deep learning model, the system can determine in advance whether the load will increase according to the load change trend, and dynamically adjust the temperature and humidity monitoring sensitivity. When the load increases, the deep learning model predicts the change trend of the power load by analyzing historical data and real-time monitoring information, intelligently evaluates the possibility of load increase, and then decides whether the sensitivity needs to be increased. In this way, the system can respond in advance when the load increases sharply, and increase the sensitivity to a level sufficient to cope with rapid changes, so as to detect abnormal fluctuations in temperature and humidity more promptly and trigger alarms. This dynamic adjustment mechanism not only improves the accuracy and timeliness of the alarm system, but also avoids the problems of missed reports and false alarms caused by traditional fixed sensitivity settings, ensuring that the substation cabinet can still operate stably when the load increases, preventing equipment failures and accidents caused by temperature and humidity, and greatly improving the safety and operational reliability of substation equipment.

[0049] The present invention optimizes load management and monitoring of equipment operating status by accurately predicting the load change trend of the substation through a deep learning model. By collecting operating parameters in real time and extracting key features, the deep learning model can intelligently analyze the load change trend and predict future load changes based on historical data and real-time parameters. This intelligent prediction can not only identify whether the load is increasing, but also quantify the change pattern of the substation load, so as to rationally plan and manage the equipment load. When the load changes greatly, the system will dynamically adjust the alarm threshold or monitoring strategy to ensure that the equipment is still in a safe working state during load fluctuations. This optimization measure can greatly improve the operating stability of the equipment, reduce the risk of failures caused by excessive or uneven loads, and at the same time improve the energy efficiency and long-term stable operation capability of the substation.

[0050] The present invention provides Figure 2 A remote monitoring and alarm system for temperature and humidity of a transformer cabinet is shown, including a temperature and humidity monitoring module, an operation parameter acquisition module, a data preprocessing and feature extraction module, a deep learning prediction module, a load change trend classification module, a load balancing monitoring module, and a dynamic sensitivity adjustment and alarm module: Temperature and humidity monitoring module. The temperature and humidity monitoring system continuously monitors the temperature and humidity inside the switchgear cabinet with a preset sensitivity, monitors the fluctuations of temperature and humidity, and promptly discovers and alarms any abnormal conditions beyond the safe range. Operating parameter acquisition module. While monitoring the temperature and humidity, it real-time acquires the operating parameter information of the switchgear cabinet, reflects the working state and load changes of the switchgear cabinet through the acquired data, and provides a data basis for subsequent load trend analysis. Data preprocessing and feature extraction module. It preprocesses the real-time acquired operating parameter data, extracts key features reflecting the load change trend of the switchgear cabinet from the preprocessed operating parameter data, and analyzes the extracted features under the detection window to quantify and understand the specific trends and patterns of the load change of the switchgear cabinet. Deep learning prediction module. It inputs the quantified load change trend features into a pre-trained deep learning model, and uses the intelligent ability of the model to intelligently predict the future load change trend of the substation. Load change trend classification module. According to the prediction results of the deep learning model, it classifies the load change trend of the switchgear cabinet into balanced power load and aggravated power load. Load balance monitoring module. For the case of balanced power load, the temperature and humidity monitoring system continues to monitor with the preset sensitivity to ensure that the temperature and humidity are maintained within the safe range and prevent equipment failures caused by too high temperature and humidity. Dynamic sensitivity adjustment and alarm module. For the case of aggravated power load, based on the prediction results of the deep learning model, it dynamically increases the sensitivity of temperature and humidity monitoring, keenly captures the changes in temperature and humidity, and promptly triggers an alarm.

[0051] A method for remote monitoring and alarming of the temperature and humidity of a switchgear cabinet provided by an embodiment of the present invention is implemented through the above-mentioned remote monitoring and alarming system for the temperature and humidity of a switchgear cabinet. The specific methods and processes of the remote monitoring and alarming system for the temperature and humidity of a switchgear cabinet are detailed in the embodiments of the above-mentioned method for remote monitoring and alarming of the temperature and humidity of a switchgear cabinet, and will not be elaborated here.

[0052] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by software simulation of a large amount of collected data to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0053] Only some exemplary embodiments of the present invention have been described by way of illustration above. Undoubtedly, for those of ordinary skill in the art, without departing from the spirit and scope of the present invention, the described embodiments can be modified in various different ways. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the protection scope of the claims of the present invention.

[0054] It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0055] It should be understood that in various embodiments of the present application, the magnitudes of the sequence numbers of the above processes do not imply the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0056] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Skilled professionals can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present application.

[0057] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0058] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0059] In addition, the functional units in various embodiments of the present application can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0060] As described above, it is only the specific implementation manner of the present application. However, the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described above.

[0061] Only some exemplary embodiments of the present invention have been described by way of illustration above. Without doubt, for those of ordinary skill in the art, the described embodiments can be modified in various different ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the protection scope of the claims of the present invention.

Claims

1. A method for remotely monitoring and alarming the temperature and humidity of a substation cabinet, characterized in that, It includes the following steps: Continuously monitor the temperature and humidity inside the switchgear cabinet with a preset sensitivity, monitor the fluctuations of temperature and humidity, and promptly detect and alarm any abnormal conditions beyond the safe range; While monitoring the temperature and humidity, collect the operating parameter information of the switchgear cabinet in real time, and reflect the working state and load changes of the switchgear cabinet through the collected data, providing a data basis for subsequent load trend analysis; Preprocess the real-time collected operating parameter data, extract the key features reflecting the load change trend of the switchgear cabinet from the preprocessed operating parameter data, and analyze the extracted features under the detection window to quantify and understand the specific trends and patterns of the load change of the switchgear cabinet; Input the quantified load change trend features into a pre-trained deep learning model, and use the intelligent ability of the model to intelligently predict the future load change trend of the substation; According to the prediction results of the deep learning model, divide the load change trend of the switchgear cabinet into balanced power load and increased power load; For the case of balanced power load, the temperature and humidity monitoring system continues to monitor with a preset sensitivity to ensure that the temperature and humidity are maintained within the safe range; For the case of increased power load, based on the prediction results of the deep learning model, dynamically increase the sensitivity of temperature and humidity monitoring, keenly capture the changes in temperature and humidity, and promptly trigger an alarm.

2. A method for remote monitoring and alarming of temperature and humidity in a power transformation cabinet according to claim 1, characterized in that, Extract the key features reflecting the load change trend of the switchgear cabinet from the preprocessed operating parameter data. The extracted parameters include the response speed of the equipment in the switchgear cabinet to load changes and the distortion degree of the current signal. Under the detection window, analyze the response speed of the equipment in the switchgear cabinet to load changes and the distortion degree of the current signal, and generate the instantaneous load response time reference value and the current waveform distortion reference value respectively. Quantify and understand the specific trends and patterns of the load change of the switchgear cabinet through the instantaneous load response time reference value and the current waveform distortion reference value.

3. A method for remote monitoring and alarming of temperature and humidity of a substation cabinet according to claim 2, characterized in that, The specific steps for analyzing the response speed of the equipment in the switchgear cabinet to load changes under the detection window to generate the instantaneous load response time reference value are as follows: Collect the operating parameters of various equipment in the switchgear cabinet in real time, and preprocess the obtained data. Through the preprocessed data, evaluate the rate and amplitude of load changes, and analyze the impact of load changes on equipment performance; After obtaining the load change rate, establish an instantaneous load response model of the equipment. By analyzing the relationship between the load change rate and the equipment state, establish a dynamic response function, and calculate the instantaneous load response time of the equipment based on the characteristics of load changes; Generate the instantaneous load response time reference value based on the modeling results of the load change rate and the instantaneous load response time of the equipment.

4. A method for remote monitoring and alarming of temperature and humidity of a power distribution cabinet according to claim 2, characterized in that, The specific steps for analyzing the distortion degree of the current signal under the detection window to generate the current waveform distortion reference value are as follows: Decompose the current signal into high-frequency and low-frequency components, deeply analyze the distortion characteristics in the signal, and transform the current signal into a frequency-domain signal through fast Fourier transform; Segment the frequency-domain signal through a frequency threshold; After segmenting the high-frequency and low-frequency components of the frequency-domain signal, use energy analysis to measure the distortion degree of the current waveform; Generate a reference value for current waveform distortion based on the energy of high-frequency components and the low-frequency stable part through weighted synthesis.

5. A method for remote monitoring and alarming of temperature and humidity of a substation cabinet according to claim 2, characterized in that, Input the generated reference value of instantaneous load response time and the reference value of current waveform distortion after analysis into a pre-trained deep learning model. Generate a load abnormal fluctuation index based on the deep learning model, and intelligently predict the future load change trend of the substation through the abnormal fluctuation index.

6. A method for remote monitoring and alarming of temperature and humidity in a power distribution cabinet according to claim 5, characterized in that, Compare and analyze the load abnormal fluctuation index generated when intelligently predicting the future load change trend of the substation through the intelligent ability of the pre-trained deep learning model with the pre-set reference threshold of the load abnormal fluctuation index, and classify the load change trend of the switch cabinet. The classification steps are as follows: If the load abnormal fluctuation index is greater than the pre-set reference threshold of the load abnormal fluctuation index, classify the load change trend of the switch cabinet as increased power load; If the load abnormal fluctuation index is less than or equal to the pre-set reference threshold of the load abnormal fluctuation index, classify the load change trend of the switch cabinet as balanced power load.

7. A method for remote monitoring and alarming of temperature and humidity of a power distribution cabinet according to claim 6, characterized in that, For the case of increased power load, based on the prediction result of the deep learning model, dynamically improve the sensitivity of temperature and humidity monitoring. The specific steps are as follows: After the load change trend of the switch cabinet is classified as increased power load, adjust the sensitivity of the temperature and humidity monitoring system, and improve the sensitivity by calculating the dynamic sensitivity adjustment coefficient to ensure that the sensitivity of the temperature and humidity monitoring system is effectively adjusted according to the increased load situation. The calculation expression of the dynamic sensitivity adjustment coefficient is as follows: ， In the formula, is the load abnormal fluctuation index, is the reference threshold of the load abnormal fluctuation index, is the maximum value of the load abnormal fluctuation index, and are both non-linear adjustment parameters, used to control the load abnormal fluctuation index for the non-linear amplification degree of the reference threshold of the load abnormal fluctuation index ; is used to control the non-linear influence of the load abnormal fluctuation index on the maximum value of the load abnormal fluctuation index ; is the dynamic sensitivity adjustment coefficient; According to the dynamic sensitivity adjustment coefficient Adjust the sensitivity of the temperature and humidity monitoring system to capture the changing trend of the temperature and humidity in the switch cabinet with higher sensitivity. The adjustment expression is as follows: ， Wherein, is the adjusted temperature and humidity monitoring sensitivity, is the preset basic sensitivity, is the adjustment coefficient, is the load abnormal fluctuation index change rate, is the adjustment factor.

8. A remote temperature and humidity monitoring and alarming system for a power transformation cabinet, which is used to implement the remote temperature and humidity monitoring and alarming method for the power transformation cabinet described in any one of the above claims 1-7, is characterized in that, Including a temperature and humidity monitoring module, an operating parameter acquisition module, a data preprocessing and feature extraction module, a deep learning prediction module, a load change trend classification module, a load balance monitoring module, and a dynamic sensitivity adjustment and alarm module: Temperature and humidity monitoring module: The temperature and humidity monitoring system continuously monitors the temperature and humidity inside the switch cabinet with a preset sensitivity, monitors the temperature and humidity fluctuations, and promptly discovers and alarms any abnormal conditions beyond the safe range. Operating parameter acquisition module: While monitoring the temperature and humidity, it real-time acquires the operating parameter information of the switch cabinet, and reflects the working state and load change of the switch cabinet through the acquired data, providing a data basis for subsequent load trend analysis. Data preprocessing and feature extraction module: Preprocess the real-time acquired operating parameter data, and extract key features reflecting the load change trend of the switch cabinet from the preprocessed operating parameter data. Analyze the extracted features under the detection window to quantify and understand the specific trend and pattern of the load change of the switch cabinet. Deep learning prediction module: Input the quantified load change trend features into a pre-trained deep learning model, and utilize the intelligent ability of the model to intelligently predict the future load change trend of the substation. Load change trend classification module: According to the prediction result of the deep learning model, classify the load change trend of the switch cabinet into balanced power load and increased power load. Load balance monitoring module: For the case of balanced power load, the temperature and humidity monitoring system continues to monitor with the preset sensitivity to ensure that the temperature and humidity are maintained within the safe range and prevent equipment failures caused by excessive temperature and humidity. Dynamic sensitivity adjustment and alarm module. In case of increasing power load, based on the prediction results of the deep learning model, it dynamically improves the sensitivity of temperature and humidity monitoring, keenly captures the changes in temperature and humidity, and triggers an alarm in a timely manner.

Images