Power transformer fault early warning system and method based on intelligent spectrum sensing

Abstract

The invention relates to the field of power transformer fault monitoring, and discloses a power transformer fault early warning system and method based on intelligent spectrum sensing, and the system comprises a transformer data collection module which comprises a spectrum sensing unit which collects an electromagnetic spectrum signal in real time during the operation of a power transformer, the state sensor unit collects temperature, humidity and vibration data of the transformer; the transformer feature extraction module extracts spectrum features and state features from the preprocessed data, analyzes to obtain an evaluation value of each feature parameter, establishes a transformer health evaluation model according to the evaluation value of each feature parameter, and obtains a transformer health evaluation value through the transformer health evaluation model; and the fault early warning module compares the transformer health assessment value with a preset health threshold range, starts an early warning mechanism when the transformer health assessment value exceeds the health threshold range, and judges the severity and development situation of the fault according to the degree and trend of exceeding the health threshold range.

Description

Technical Field

[0001] The present invention relates to the field of power transformer fault monitoring, and in particular to a power transformer fault early warning system and method based on intelligent spectrum sensing. Background Art

[0002] With the continuous growth of electricity demand and the increasing scale of power systems, power transformers, as key equipment in power systems, are also increasing in number and capacity. Once a power transformer fails, it may cause a large-scale power outage, causing huge losses to the society and economy.

[0003] There are many deficiencies in the existing fault monitoring methods for power transformers. Among them, the traditional monitoring methods mainly rely on regular inspections and simple electrical parameter measurements, such as measuring winding resistance and insulation resistance, which can only detect relatively obvious faults and are difficult to detect early potential faults. Moreover, the regular inspection cycle is long, and between two inspections, the transformer may have potential faults but has not been discovered in time. In addition, a single electrical parameter measurement cannot fully reflect the operating status of the transformer, and it is difficult to accurately determine the type and severity of the fault. At the same time, with the demand for the intelligent development of power systems, traditional monitoring methods cannot meet the requirements of real-time monitoring, data analysis and intelligent early warning.

[0004] In summary, the current power transformer fault monitoring faces severe challenges, and there is an urgent need for a more advanced and reliable fault warning system and method to improve the operating reliability of power transformers and ensure the safe and stable operation of the power system. Summary of the invention

[0005] The purpose of the present invention is to provide a power transformer fault early warning system and method based on intelligent spectrum sensing to solve the above technical problems.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] Power transformer fault early warning system based on intelligent spectrum sensing includes:

[0008] The transformer data acquisition module includes a spectrum sensing unit and a state sensor unit. The spectrum sensing unit is used to collect electromagnetic spectrum signals of the power transformer in real time when the power transformer is running. The state sensor unit is used to collect temperature, humidity and vibration data of the transformer.

[0009] Transformer data preprocessing module, used for signal filtering, data normalization and data compression processing of collected electromagnetic spectrum signals, temperature, humidity and vibration data;

[0010] A transformer feature extraction module is connected to the data preprocessing module to extract spectrum features and state features from the preprocessed data, wherein the spectrum features include spectrum peak, frequency bandwidth, and spectrum center of gravity, and the state features include temperature change rate, vibration amplitude, and vibration frequency;

[0011] The transformer analysis module analyzes the spectrum characteristics and state characteristics, that is, by comparing and analyzing the change curve of the characteristic parameter in multiple time periods with the reference change curve, the evaluation value of each characteristic parameter is obtained, and then a transformer health assessment model is established based on the evaluation value of each characteristic parameter, and the transformer health assessment value is obtained through the transformer health assessment model;

[0012] A fault warning module is used to receive the transformer health assessment value and compare it with a preset health threshold range;

[0013] When the transformer health assessment value exceeds the health threshold range, the early warning mechanism is activated, and the severity and development trend of the fault are judged based on the degree and trend of exceeding the health threshold range.

[0014] In a further solution, the transformer health assessment model is expressed as:

[0015] ; In the formula, represents the transformer health assessment value, Indicates The evaluation value of the feature parameter, Indicates The weight coefficients corresponding to the characteristic parameters are formulated based on experimental data and historical data. Represents the number of terms of feature parameters.

[0016] Further plans, The expression for the evaluation value of the item feature parameter is: ;

[0017] In the formula, Indicates collection time period, A curve showing the change of the current characteristic parameters over time in each time period. Indicates time period, , Respectively represent The start and end time of a time period. Indicates the reference change curve of the current characteristic parameters, Indicates the weight factor for each time period.

[0018] A further solution is to obtain the weight factor for each time period The process is:

[0019] First, through the formula: , respectively calculate the actual average change rate of the current characteristic parameters in each time period , reference average change rate ;

[0020] Substituting into the formula again: , calculate the weight factor for each time period ,and ;

[0021] In the formula, , Respectively indicate the end time , start time The corresponding current characteristic parameter value; , Respectively indicate the end time , start time The corresponding reference value of the current characteristic parameter, Indicates the relative rate of change of the current characteristic parameters in each time period, Represents the adjustment factor, which is determined based on analysis of historical data.

[0022] A further solution is to determine the severity and development trend of the fault based on the degree and trend of exceeding the healthy threshold range:

[0023] Continuously obtain M sampling time points Corresponding transformer health assessment value ;

[0024] By formula: Calculate the fault trend index ;

[0025] By formula: Calculate the fault severity index ;

[0026] In the formula, , is the preset scale factor, , represents the average difference between the transformer health assessment values ​​corresponding to two adjacent sampling time points, is the fluctuation coefficient of the transformer health assessment value, Indicates the preset health threshold range.

[0027] A further solution is the fluctuation coefficient of the transformer health assessment value. The expression is: ; In the formula, Indicates The difference between the transformer health assessment values ​​corresponding to two adjacent sampling time points.

[0028] A further solution is to use the fault severity index Compare with the preset fault index interval;

[0029] like If the fault index range is exceeded, the transformer fault level is judged to be severe;

[0030] like If it falls within the fault index range, the transformer fault level is judged to be moderate;

[0031] like If it is lower than the fault index range, the transformer fault level is judged to be low.

[0032] A further solution is to use the fault trend indicator Preset fault trend warning value Make comparisons;

[0033] like , it is judged that the transformer fault is in a worsening trend;

[0034] like , it is judged that the transformer fault is in a stable trend.

[0035] The power transformer fault early warning method based on intelligent spectrum sensing includes the following steps:

[0036] S100, real-time collection of electromagnetic spectrum signals, temperature, humidity and vibration data of power transformers during operation;

[0037] S200, performing signal filtering, data normalization and data compression processing on the collected electromagnetic spectrum signals, temperature, humidity and vibration data;

[0038] S300, extracting spectrum features and state features from the preprocessed data;

[0039] S400, analyzing the spectrum characteristics and state characteristics, that is, obtaining an evaluation value of each characteristic parameter by comparing and analyzing the change curve of the characteristic parameter in multiple time periods with the reference change curve, and then establishing a transformer health assessment model according to the evaluation values ​​of each characteristic parameter, and obtaining the transformer health assessment value through the transformer health assessment model;

[0040] S500, receiving the transformer health assessment value and comparing it with the preset health threshold range; when the transformer health assessment value exceeds the health threshold range, the early warning mechanism is activated, and the severity and development trend of the fault are judged according to the degree and trend of exceeding the health threshold range.

[0041] Beneficial effects of the present invention:

[0042] (1) The spectrum sensing unit and the status sensor unit collect a variety of data, including electromagnetic spectrum signals, temperature, humidity and vibration data, which can reflect the operating status of the power transformer from multiple dimensions. Compared with single parameter monitoring, it greatly improves the accuracy and comprehensiveness of fault warning;

[0043] (2) The transformer health assessment model comprehensively considers multiple characteristic parameters and their weights to calculate the health assessment value, which can more accurately assess the health status of the transformer; the fault warning module judges the severity and development trend of the fault based on the comparison results between the assessment value and the health threshold range, as well as the degree and trend of exceeding the health threshold range, and issues a warning in advance, so as to buy sufficient time for the operation and maintenance personnel to take measures, effectively avoid the occurrence of large-scale power outages, and reduce social and economic losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The present invention will be further described below in conjunction with the accompanying drawings.

[0045] Figure 1 It is a system logic block diagram of the present invention;

[0046] Figure 2 It is a diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] See also Figure 1 As shown, the present invention is a power transformer fault early warning system based on intelligent spectrum sensing, comprising:

[0049] The transformer data acquisition module includes a spectrum sensing unit and a state sensor unit. The spectrum sensing unit is used to collect electromagnetic spectrum signals of the power transformer in real time when the power transformer is running. The state sensor unit is used to collect temperature, humidity and vibration data of the transformer.

[0050] It should be noted that: the spectrum sensing unit can use a high-sensitivity electromagnetic spectrum sensor and install it in the key parts of the power transformer to ensure that the electromagnetic spectrum signal during operation can be collected in real time and accurately; the temperature sensor of the state sensor unit can use a high-precision optical fiber temperature sensor, the humidity sensor can use a capacitive humidity sensor, and the vibration sensor can use a piezoelectric vibration sensor. These sensors have the characteristics of strong anti-interference ability and high precision, and can reliably collect the temperature, humidity and vibration data of the transformer;

[0051] Transformer data preprocessing module, used for signal filtering, data normalization and data compression processing of collected electromagnetic spectrum signals, temperature, humidity and vibration data;

[0052] It should be noted that the wavelet transform filtering algorithm can be used in the signal filtering stage, which can effectively remove the noise in the signal while retaining the detailed characteristics of the signal. The Z-score normalization method can be used for data normalization to map data with different characteristics to the same scale for subsequent analysis; lossless compression algorithms, such as the Huffman coding algorithm, can be used for data compression to reduce the amount of data without losing data information;

[0053] The transformer feature extraction module is connected to the data preprocessing module to extract spectrum features and state features from the preprocessed data, wherein the spectrum features include spectrum peak, frequency bandwidth, and spectrum center of gravity, and the state features include temperature change rate, vibration amplitude, and vibration frequency. It should be noted that the spectrum features are extracted by a pre-trained convolutional neural network model CNN, while the state features are extracted by a pre-trained recurrent neural network model RNN and its variant LSTM. The above-mentioned neural network models are all prior arts directly obtained by skilled technicians, so they are not described in detail.

[0054] The transformer analysis module analyzes the spectrum characteristics and state characteristics, that is, by comparing and analyzing the change curve of the characteristic parameter in multiple time periods with the reference change curve, the evaluation value of each characteristic parameter is obtained, and then a transformer health assessment model is established based on the evaluation value of each characteristic parameter, and the transformer health assessment value is obtained through the transformer health assessment model;

[0055] A fault warning module is used to receive the transformer health assessment value and compare it with a preset health threshold range;

[0056] When the transformer health assessment value exceeds the health threshold range, the early warning mechanism is activated, and the severity and development trend of the fault are judged based on the degree and trend of exceeding the health threshold range.

[0057] In the present invention, a variety of data are collected through the spectrum sensing unit and the state sensor unit, covering electromagnetic spectrum signals, temperature, humidity and vibration data, which can reflect the operating status of the power transformer from multiple dimensions. Compared with single parameter monitoring, the accuracy and comprehensiveness of fault warning are greatly improved; for example, when a partial discharge fault occurs inside the transformer, the spectrum peak, frequency bandwidth and other characteristics of the electromagnetic spectrum signal will change, and the temperature and vibration data may also be abnormal. Combining these parameters can more accurately judge the fault; the data preprocessing module filters, normalizes and compresses the collected data, effectively removes noise interference, unifies the data scale, and reduces the burden of data storage and transmission. Subsequent analysis provides high-quality data; the extraction of spectral features and state features can mine key information reflecting the operating status of the transformer, such as the spectral center of gravity, temperature change rate, etc., providing strong support for health assessment and fault warning; the transformer health assessment model comprehensively considers multiple characteristic parameters and their weights, calculates the health assessment value, and can more accurately assess the health status of the transformer; the fault warning module judges the severity and development trend of the fault based on the comparison result of the assessment value with the health threshold range, as well as the degree and trend of exceeding the health threshold range, and issues a warning in advance, so as to buy sufficient time for the operation and maintenance personnel to take measures, effectively avoid the occurrence of large-scale power outages, and reduce social and economic losses.

[0058] The expression of the transformer health assessment model is:

[0059] ; In the formula, represents the transformer health assessment value, Indicates The evaluation value of the feature parameter, Indicates The weight coefficients corresponding to the characteristic parameters are formulated based on experimental data and historical data. Represents the number of terms of feature parameters.

[0060] No. The expression for the evaluation value of the item feature parameter is: ; In the formula, Indicates collection time period, A curve showing the change of the current characteristic parameters over time in each time period. Indicates time period, , Respectively represent The start and end time of a time period. Indicates the reference change curve of the current characteristic parameters, Indicates the weight factor for each time period.

[0061] In the present invention, a specific calculation formula of the transformer health assessment model is provided. Get transformer health assessment value ,It can be seen from the above expression that within the health threshold range, the larger the evaluation value of each characteristic parameter, the healthier the transformer is. The slight changes in the evaluation value of each characteristic parameter are amplified by exponential means to improve the accuracy of transformer health assessment, so that the health assessment value can more accurately reflect the actual health status of the transformer, provide a quantitative basis for subsequent fault warning, and help to timely discover the potential fault risks of the transformer;

[0062] The evaluation value of each characteristic parameter is Obviously, by analyzing the current characteristic parameter change curves in multiple time periods Change curve with reference Perform integral operations and compare, and combine weights , can comprehensively consider the changes of characteristic parameters in different time periods, through The average processing of data from multiple time periods reduces the impact of data fluctuations in a single time period, making the evaluation value more stable and reliable; weight The results can be adjusted according to the importance of different time periods, and greater weights can be given to key time periods to more accurately reflect the impact of characteristic parameters on the health status of the transformer in different time periods, thereby obtaining characteristic parameter evaluation values ​​that are more in line with the actual situation.

[0063] Get the weight factor for each time period The process is:

[0064] First, through the formula: , respectively calculate the actual average change rate of the current characteristic parameters in each time period , reference average change rate ;

[0065] Substituting into the formula again: , calculate the weight factor for each time period ,and ;

[0066] In the formula, , Respectively indicate the end time , start time The corresponding current characteristic parameter value; , Respectively indicate the end time , start time The corresponding reference value of the current characteristic parameter, Indicates the relative rate of change of the current characteristic parameters in each time period, Represents the adjustment factor, which is determined based on analysis of historical data.

[0067] In the present invention, by calculating the actual average rate of change and the reference average rate of change , and taking their ratio as part of the relative rate of change can intuitively reflect the degree of change of the current characteristic parameter relative to the reference value in each time period; the existence of the adjustment factor This provides greater flexibility for adjusting weights, and the weights can be further optimized according to actual conditions. Calculating the weight factor makes the determination of the weight more scientific and reasonable, and can dynamically adjust the weight according to the actual changes in the characteristic parameters, thereby more accurately reflecting the importance of different time periods when calculating the characteristic parameter evaluation value and improving the accuracy of the evaluation results.

[0068] Based on the degree and trend of exceeding the healthy threshold range, the process of judging the severity and development trend of the fault is as follows:

[0069] Continuously obtain M sampling time points Corresponding transformer health assessment value ;

[0070] By formula: Calculate the fault trend index ;

[0071] By formula: Calculate the fault severity index ;

[0072] In the formula, , is the preset scale factor, , represents the average difference between the transformer health assessment values ​​corresponding to two adjacent sampling time points, is the fluctuation coefficient of the transformer health assessment value, Indicates the preset health threshold range.

[0073] In the present invention, by formula: Calculate the fault trend index , comprehensively considering the average difference of transformer health assessment values ​​at adjacent sampling time points and the fluctuation coefficient of the transformer health assessment value , by presetting the proportionality factor , To adjust the relative importance of the two in fault trend judgment; the fault trend indicator can reflect the changing trend of the transformer health assessment value. When compared with the preset fault trend warning value, it can be judged whether the transformer fault is in a deteriorating trend or a stable trend, providing the operation and maintenance personnel with important information on the development direction of the transformer fault, which helps to formulate corresponding maintenance strategies in advance;

[0074] By formula: Calculate the fault severity index ; By comparing the transformer health assessment value With preset health threshold range The fault severity index is calculated based on the relationship between the health assessment value and the threshold value. When the health assessment value exceeds the threshold range, the degree of excess can be intuitively reflected. By comparing it with the preset fault indicator range, the severity of the transformer fault can be judged, providing the operation and maintenance personnel with a quantitative fault severity index, which helps to take corresponding measures to deal with the fault in time and ensure the safe and stable operation of the power transformer.

[0075] Fluctuation coefficient of transformer health assessment value The expression is: ; In the formula, Indicates The difference between the transformer health assessment values ​​corresponding to two adjacent sampling time points.

[0076] In the present invention, the difference of health assessment values ​​at adjacent sampling time points is calculated. Difference from the mean The square root of the average value of the sum of squared deviations can quantify the degree of fluctuation of the transformer health assessment value. The fluctuation coefficient reflects the stability of the transformer operating state and is used in calculating the fault trend index. As an important parameter, the average difference between the health assessment values ​​at adjacent sampling time points Together, it can more comprehensively reflect the changes in transformer health assessment values, thereby more accurately determining fault trends.

[0077] Fault severity index Compare with the preset fault index interval;

[0078] like If the fault index range is exceeded, the transformer fault level is judged to be severe;

[0079] like If it falls within the fault index range, the transformer fault level is judged to be moderate;

[0080] like If it is lower than the fault index range, the transformer fault level is judged to be low.

[0081] Failure Trend Indicator Preset fault trend warning value Make comparisons;

[0082] like , it is judged that the transformer fault is in a worsening trend;

[0083] like , it is judged that the transformer fault is in a stable trend.

[0084] See also Figure 2 As shown, the power transformer fault early warning method based on intelligent spectrum sensing includes the following steps:

[0085] S100, real-time collection of electromagnetic spectrum signals, temperature, humidity and vibration data of power transformers during operation;

[0086] S200, performing signal filtering, data normalization and data compression processing on the collected electromagnetic spectrum signals, temperature, humidity and vibration data;

[0087] S300, extracting spectrum features and state features from the preprocessed data;

[0088] S400, analyzing the spectrum characteristics and state characteristics, that is, obtaining an evaluation value of each characteristic parameter by comparing and analyzing the change curve of the characteristic parameter in multiple time periods with the reference change curve, and then establishing a transformer health assessment model according to the evaluation values ​​of each characteristic parameter, and obtaining the transformer health assessment value through the transformer health assessment model;

[0089] S500, receiving the transformer health assessment value and comparing it with the preset health threshold range; when the transformer health assessment value exceeds the health threshold range, the early warning mechanism is activated, and the severity and development trend of the fault are judged according to the degree and trend of exceeding the health threshold range.

[0090] It should be noted that the calculation formulas and various parameters involved in the calculation in the present invention have been dimensionally processed in advance, and the process of dimensionless processing is well known in the industry and will not be described here.

[0091] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. The power transformer fault early warning system based on intelligent spectrum sensing is characterized by: include: The transformer data acquisition module includes a spectrum sensing unit and a state sensor unit. The spectrum sensing unit is used to collect electromagnetic spectrum signals of the power transformer in real time when the power transformer is running. The state sensor unit is used to collect temperature, humidity and vibration data of the transformer. Transformer data preprocessing module, used for signal filtering, data normalization and data compression processing of collected electromagnetic spectrum signals, temperature, humidity and vibration data; A transformer feature extraction module is connected to the data preprocessing module to extract spectrum features and state features from the preprocessed data, wherein the spectrum features include spectrum peak, frequency bandwidth, and spectrum center of gravity, and the state features include temperature change rate, vibration amplitude, and vibration frequency; The transformer analysis module analyzes the spectrum characteristics and state characteristics, that is, by comparing and analyzing the change curve of the characteristic parameter in multiple time periods with the reference change curve, the evaluation value of each characteristic parameter is obtained, and then a transformer health assessment model is established based on the evaluation value of each characteristic parameter, and the transformer health assessment value is obtained through the transformer health assessment model; A fault warning module is used to receive the transformer health assessment value and compare it with a preset health threshold range; When the transformer health assessment value exceeds the health threshold range, the early warning mechanism is activated, and the severity and development trend of the fault are judged based on the degree and trend of exceeding the health threshold range.

2. The power transformer fault early warning system based on intelligent spectrum sensing according to claim 1 is characterized in that: The expression of the transformer health assessment model is: ; In the formula, represents the transformer health assessment value, Indicates The evaluation value of the feature parameter, Indicates The weight coefficients corresponding to the characteristic parameters are formulated based on experimental data and historical data. The number of terms representing the feature parameters.

3. The power transformer fault early warning system based on intelligent spectrum sensing according to claim 2 is characterized in that: No. The expression for the evaluation value of the item feature parameter is: ; In the formula, Indicates collection time period, A curve showing the change of the current characteristic parameters over time in each time period. Indicates time period, , Respectively represent The start and end time of a time period. Indicates the reference change curve of the current characteristic parameters, Indicates the weight factor for each time period.

4. The power transformer fault early warning system based on intelligent spectrum sensing according to claim 3 is characterized in that: Get the weight factor for each time period The process is: First, through the formula: , respectively calculate the actual average change rate of the current characteristic parameters in each time period , reference average change rate ; Substituting into the formula again: , calculate the weight factor for each time period ,and ; In the formula, , Respectively indicate the end time , start time The corresponding current characteristic parameter value; , Respectively indicate the end time , start time The corresponding reference value of the current characteristic parameter, Indicates the relative rate of change of the current characteristic parameters in each time period, Represents the adjustment factor, which is determined based on analysis of historical data.

5. The power transformer fault early warning system based on intelligent spectrum sensing according to claim 2 or 4, characterized in that: Based on the degree and trend of exceeding the healthy threshold range, the process of judging the severity and development trend of the fault is as follows: Continuously obtain M sampling time points Corresponding transformer health assessment value ; By formula: Calculate the fault trend index ; By formula: Calculate the fault severity index ; In the formula, , is the preset scale factor, , represents the average difference between the transformer health assessment values ​​corresponding to two adjacent sampling time points, is the fluctuation coefficient of the transformer health assessment value, Indicates the preset health threshold range.

6. The power transformer fault early warning system based on intelligent spectrum sensing according to claim 5 is characterized in that: Fluctuation coefficient of transformer health assessment value The expression is: ; In the formula, Indicates The difference between the transformer health assessment values ​​corresponding to two adjacent sampling time points.

7. The power transformer fault early warning system based on intelligent spectrum sensing according to claim 5 is characterized in that: Fault severity index Compare with the preset fault index interval; like If the fault index range is exceeded, the transformer fault level is judged to be severe; like If it falls within the fault index range, the transformer fault level is judged to be moderate; like If it is lower than the fault index range, the transformer fault level is judged to be low.

8. The power transformer fault early warning system based on intelligent spectrum sensing according to claim 5 is characterized in that: Failure Trend Indicator Preset fault trend warning value Make comparisons; like , it is judged that the transformer fault is in a worsening trend; like , it is judged that the transformer fault is in a stable trend.

9. A power transformer fault early warning method based on intelligent spectrum sensing, characterized in that: The method is implemented based on the power transformer fault early warning system based on intelligent spectrum sensing as described in claim 1, and comprises the following steps: S100, real-time collection of electromagnetic spectrum signals, temperature, humidity and vibration data during operation of power transformers; S200, performing signal filtering, data normalization and data compression processing on the collected electromagnetic spectrum signals, temperature, humidity and vibration data; S300, extracting spectrum features and state features from the preprocessed data; S400, analyzing the spectrum characteristics and state characteristics, that is, obtaining an evaluation value of each characteristic parameter by comparing and analyzing the change curve of the characteristic parameter in multiple time periods with the reference change curve, and then establishing a transformer health assessment model according to the evaluation values ​​of each characteristic parameter, and obtaining the transformer health assessment value through the transformer health assessment model; S500, receiving a transformer health assessment value and comparing it with a preset health threshold range; When the transformer health assessment value exceeds the health threshold range, the early warning mechanism is activated, and the severity and development trend of the fault are judged based on the degree and trend of exceeding the health threshold range.

Images