Power equipment state temperature monitoring method and system

Abstract

The invention provides a power equipment state temperature monitoring method and system. The method comprises the following steps: acquiring heat indexes and energy consumption data of to-be-monitored equipment at a plurality of specified moments in a preset time window; the heat index is a weighted value according to the equipment temperature, the equipment energy consumption and the equipment operation state; arranging the popularity index and the energy consumption data at the specified moment according to a time sequence, and constructing a popularity time sequence and an energy consumption time sequence; performing dynamic analysis on the energy consumption time sequence to obtain an energy consumption time sequence feature set; analyzing the popularity time sequence, and extracting a popularity time sequence feature set; according to the energy consumption time sequence feature set, carrying out weight adjustment and combination on the extracted heat time sequence features to form a combined heat time sequence feature set; and according to the merged heat time sequence characteristics, whether the working condition of the to-be-monitored equipment is abnormal or not is identified through the preset heat critical value and the preset change interval, so that the reliability of state temperature monitoring of the power equipment is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of power equipment status monitoring, and in particular to a power equipment status temperature monitoring method and system. Background Art

[0002] With the gradual evolution of science and technology and the increasing maturity of intelligent technology, we increasingly need to conduct more sophisticated real-time monitoring and abnormal identification of the working status of power equipment. In the past, power equipment monitoring usually relied on regular inspections by staff or setting simple thresholds for judgment. However, these traditional methods may sometimes be difficult to capture subtle changes or potential problems in equipment in a timely manner, and there are also certain limitations in efficiency.

[0003] The temperature monitoring technology for power equipment that is currently widely used has the following shortcomings: (1) Inaccurate temperature measurement. In actual operation, many transformers fail due to overtemperature, which has a great impact on the power system; (2) Some monitoring systems are prone to failure when running for a long time or facing harsh environments, such as sensor failure and data transmission interruption; (3) Many systems can only simply display and store temperature data and cannot perform intelligent analysis of the data, such as trend analysis of historical data and alarm for abnormal data; (4) In temperature monitoring, only the equipment temperature is considered, and the impact of factors such as equipment energy consumption and equipment operating status on the equipment temperature is not considered, resulting in low reliability of power equipment status temperature monitoring.

[0004] In view of this problem, the present invention provides a method and system for monitoring the temperature of an electric power device to solve the above problem. Summary of the invention

[0005] In order to solve the problems existing in the prior art, the present invention innovatively proposes a method and system for monitoring the state temperature of electric equipment, which effectively solves the problem of low reliability of the state temperature monitoring of electric equipment caused by the prior art, and effectively improves the reliability of the state temperature monitoring of electric equipment.

[0006] A first aspect of the present invention provides a method for monitoring the temperature of an electric power device, comprising:

[0007] Obtain the heat index and energy consumption data of the monitored device at multiple specified times within a preset time window; wherein the heat index is a weighted value based on the device temperature, device energy consumption and device operating status;

[0008] Arrange the heat index and energy consumption data at the specified time in time series to construct heat time series and energy consumption time series;

[0009] Perform dynamic analysis on the energy consumption time series, evaluate the change rate and amplitude of energy consumption data at each specified time, and obtain the energy consumption time series feature set;

[0010] Analyze the heat time series, identify the heat change rate, the highest heat point and the lowest heat point at each specified time, and extract the heat time series feature set;

[0011] According to the energy consumption time series feature set, the extracted heat time series features are weighted and merged to form a merged heat time series feature set;

[0012] According to the merged heat time series characteristics, through the preset heat critical value and the preset change range, it is identified whether the working condition of the monitored equipment is abnormal. If the merged heat time series characteristics are greater than the sum of the preset heat critical value and the upper limit fluctuation value of the preset change range, or less than the difference between the preset heat critical value and the lower limit fluctuation value of the preset change range, it is considered to be an abnormal condition.

[0013] A second aspect of the present invention provides a power equipment state temperature monitoring system, comprising:

[0014] An acquisition module is used to acquire the heat index and energy consumption data of the device to be monitored at multiple specified times within a preset time window; wherein the heat index is a weighted value based on the device temperature, device energy consumption and device operating status;

[0015] The construction module arranges the heat index and energy consumption data at a specified time in time series to construct the heat time series and energy consumption time series;

[0016] The analysis module dynamically analyzes the energy consumption time series, evaluates the change rate and amplitude of the energy consumption data at each specified time, and obtains the energy consumption time series feature set;

[0017] The parsing module parses the heat time series, identifies the heat change rate, the highest heat point and the lowest heat point at each specified time, and extracts the heat time series feature set;

[0018] The adjustment and merging module adjusts and merges the extracted heat time series features according to the energy consumption time series feature set to form a merged heat time series feature set;

[0019] The identification module identifies whether the working condition of the monitored equipment is abnormal based on the merged heat time series feature set, through the preset heat critical value and the preset change interval. If the merged heat time series feature is greater than the sum of the preset heat critical value and the upper limit fluctuation value of the preset change interval, or is less than the difference between the preset heat critical value and the lower limit fluctuation value of the preset change interval, it is considered to be an abnormal condition.

[0020] The technical solution adopted by the present invention includes the following technical effects:

[0021] 1. The present invention obtains the heat index and energy consumption data of the monitored equipment at multiple specified moments within a preset time window; wherein the heat index is a weighted value based on the equipment temperature, equipment energy consumption and equipment operating status; the heat index and energy consumption data at the specified moment are arranged in time series to construct a heat time series and an energy consumption time series; the energy consumption time series is dynamically analyzed to obtain an energy consumption time series feature set; the heat time series is parsed to extract a heat time series feature set; according to the energy consumption time series feature set, the extracted heat time series features are weighted and merged to form a merged heat time series feature set; according to the merged heat time series features, by presetting the heat critical value and the preset variation range, it is identified whether the working condition of the monitored equipment is abnormal, which effectively solves the problem of low reliability of power equipment status temperature monitoring caused by the prior art, and effectively improves the reliability of power equipment status temperature monitoring.

[0022] 2. The thermal index is used in the technical solution of the present invention. Compared with using only temperature to monitor the status of the device, the thermal index can comprehensively consider multiple factors related to the heat of the device, rather than just the temperature, thereby providing a more comprehensive evaluation of the thermal performance of the device; by integrating multiple parameters, the thermal index can more keenly capture possible problems with the device, such as failure or overload of the cooling system, thereby issuing an early warning; the thermal index can be customized and optimized according to specific needs, and by adjusting the parameters or weights in the algorithm, it can be more adapted to the monitoring needs of specific equipment; for some complex systems or equipment, a single temperature measurement may not be sufficient to fully reflect its thermal performance; the thermal index can more accurately evaluate the status of these systems by integrating multiple related factors.

[0023] 3. The energy consumption time series features in the technical solution of the present invention include the peak value, valley value, average change rate, maximum change amount of energy consumption at all time points in the energy consumption time series, and the change rate and maximum change amount of energy consumption at adjacent time points in the energy consumption time series; the heat time series features in the heat time series feature set include the average change rate, maximum change rate, minimum change rate of the entire heat change rate sequence, the maximum heat value and the minimum heat value in the entire heat time series, the time point corresponding to the maximum heat value in the entire heat time series, the time point corresponding to the minimum heat value, and the difference between the maximum heat value and the minimum heat value in the entire heat time series, which not only includes the numerical information of energy consumption and heat data, but also includes the dynamic characteristics of energy consumption and heat data changing over time, etc., providing a rich data basis for subsequent anomaly detection.

[0024] 4. In the technical solution of the present invention, through the optimization of genetic algorithm, the optimal weight combination corresponding to all thermal time series features in a set of thermal time series feature sets is obtained, so that the correlation or similarity between the merged thermal time series feature set and the energy consumption time series feature set is the highest; further improving the reliability of power equipment status temperature monitoring.

[0025] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0027] Figure 1 This is a schematic diagram of a process of a method in Example 1 of the present invention;

[0028] Figure 2 Schematic diagram of the process of step S2 in the method of embodiment 1 of the present invention;

[0029] Figure 3 Schematic diagram of the process of step S3 in the method of embodiment 1 of the present invention;

[0030] Figure 4 Schematic diagram of the process of step S4 in the method of embodiment 1 of the present invention;

[0031] Figure 5 Schematic diagram of the process of step S5 in the method of embodiment 1 of the present invention;

[0032] Figure 6 Schematic diagram of the process of step S51 in the method of embodiment 1 of the present invention;

[0033] Figure 7 Schematic diagram of the process of step S52 in the method of embodiment 1 of the present invention;

[0034] Figure 8 It is a schematic diagram of the structure of the system of Example 2 in the solution of the present invention. DETAILED DESCRIPTION

[0035] In order to clearly illustrate the technical features of the present solution, the present invention is described in detail below through specific implementation methods and in conjunction with the accompanying drawings. The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. In addition, the present invention may repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed. It should be noted that the components illustrated in the accompanying drawings are not necessarily drawn to scale. The present invention omits the description of known components and processing techniques and processes to avoid unnecessary limitations on the present invention.

[0036] Embodiment 1

[0037] The temperature monitoring technology for power equipment currently in widespread use has the following disadvantages: (1) Inaccurate temperature measurement. In actual operation, many transformers fail due to overtemperature, which has a great impact on the power system; (2) Some monitoring systems are prone to failure when running for a long time or facing harsh environments, such as sensor failure, data transmission interruption, etc.; (3) Many systems can only simply display and store temperature data and cannot perform intelligent analysis of the data, such as trend analysis of historical data, alarm for abnormal data, etc.; (4) The installation and debugging process of some systems is relatively cumbersome and requires professional technicians to perform on-site operations, which undoubtedly increases the use cost and time cost of the system; (5) Some systems also have the risk of damaging the original equipment or affecting the normal operation of the transformer during the installation and debugging process.

[0038] To solve some or all of the shortcomings of existing power equipment temperature monitoring technology, such as Figure 1 As shown, the present invention provides a method for monitoring the temperature of an electric power device, comprising:

[0039] S1, obtaining the heat index and energy consumption data of the device to be monitored at multiple specified times within a preset time window; wherein the heat index is a weighted value based on the device temperature, device energy consumption and device operating status;

[0040] S2, arrange the heat index and energy consumption data at the specified time in time series to construct the heat time series and energy consumption time series;

[0041] S3, dynamically analyze the energy consumption time series, evaluate the change rate and change range of the energy consumption data at each specified time, and obtain the energy consumption time series feature set;

[0042] S4, analyzing the heat time series, identifying the heat change rate, the highest heat point and the lowest heat point at each specified time, and extracting the heat time series feature set;

[0043] S5, according to the energy consumption time series feature set, weight adjustment and merging of the extracted heat time series features to form a merged heat time series feature set;

[0044] S6, according to the combined heat time series characteristics, by using a preset heat critical value and a preset change range, identifying whether the working condition of the monitored equipment is abnormal;

[0045] S7, judging whether the combined heat time series feature is greater than the sum of the preset heat critical value and the upper limit fluctuation value of the preset change interval, or less than the difference between the preset heat critical value and the lower limit fluctuation value of the preset change interval. If the judgment result is yes, executing step S8, if the judgment result is no, executing step S9;

[0046] S8, it is determined that the working condition of the equipment to be monitored is abnormal;

[0047] S9, it is determined that there is no abnormality in the working condition of the equipment to be monitored.

[0048] Wherein, in step S1, the heat index and energy consumption data of the monitored equipment at multiple specified moments within a preset time window are captured through the power data center;

[0049] Heat index and temperature are two different concepts, although they are both related to "heat". Temperature is a physical concept used to quantify the thermal state of an object or environment, usually expressed in degrees Celsius, Fahrenheit or Kelvin. It is a specific, measurable physical quantity that reflects the average kinetic energy of the movement of microscopic particles inside an object.

[0050] The thermal index is a relatively broader and more comprehensive indicator. It may include temperature information, but it may also include other factors related to "heat", such as the operating status of the device, power consumption, heat dissipation, etc. The thermal index can be a value calculated by combining multiple parameters to reflect the overall "heat" status of the device or system. This index may not be a specific physical quantity, but a value calculated based on a specific algorithm or model, which is used to evaluate or monitor the thermal performance of the device.

[0051] When defining the heat index, the following factors can be considered:

[0052] Temperature: The real-time temperature of the device or environment is an important component of the thermal index.

[0053] Energy consumption: The energy consumption of a device can also reflect its thermal status. High energy consumption often means more heat generation.

[0054] Heat dissipation performance: The device's heat dissipation design, fan speed, heat sink efficiency, etc. will affect the heat index.

[0055] Operating time: The longer the device is continuously running, the more heat it is likely to generate.

[0056] Taking all the above factors into consideration, the heat index can be calculated using the following formula to more comprehensively monitor and evaluate the thermal performance of the equipment. For example, H = α·T+β·P+γ·R+δ·S; where H is the heat index; T is the real-time temperature of the equipment (power equipment), which can be obtained through a temperature sensor; P is the energy consumption of the equipment, which can be obtained through power monitoring equipment; R is the heat dissipation performance score of the equipment, which can be estimated based on the design and operating status of the heat dissipation system; S is the continuous operating time of the equipment, which can be recorded through an internal timer or an external monitoring system; α, β, γ, and δ are weight coefficients. Heat dissipation performance score of the equipment Where k is the thermal conductivity coefficient of the heat dissipation system, which represents the ability of the heat dissipation system to conduct heat; T max It is the highest temperature reached by the equipment cooling system during operation. This temperature should be measured by temperature sensors at key locations on the cooling system. env is the ambient temperature of the equipment, that is, the temperature of the air around the cooling system, which can be measured by an ambient temperature sensor; c is the temperature fluctuation coefficient; ΔT is the temperature fluctuation range of the cooling system during stable operation, that is, the difference between the highest temperature and the lowest temperature.

[0057] Advantages of heat index over temperature:

[0058] Using the thermal index has the following advantages over using temperature alone to monitor device status:

[0059] The thermal index can comprehensively consider multiple factors related to device heat, not just temperature, to provide a more comprehensive evaluation of device thermal performance; by integrating multiple parameters, the thermal index can more keenly capture possible problems with the device, such as cooling system failure or overload, and issue early warnings; the thermal index can be customized and optimized according to specific needs, and by adjusting the parameters or weights in the algorithm, it can be more suitable for the monitoring needs of specific devices; for some complex systems or devices, a single temperature measurement may not be enough to fully reflect its thermal performance. By integrating multiple related factors, the thermal index can more accurately evaluate the status of these systems.

[0060] Among them, Figure 2 As shown, step S2 specifically includes:

[0061] S21, determining the initial sliding time point of the sliding window of the sliding average filter and the size of the sliding window;

[0062] Specifically, determine the size of the sliding window of the sliding average filter and the initial sliding time point (data point), for example, select 5 time points as a window; the heat index and energy consumption data obtained from the data center are arranged in chronological order, so starting from the first data point (time point), the first 5 data points can be used as the content of the initial sliding window;

[0063] S22, respectively calculating the average values ​​of the heat index and energy consumption data in the current sliding window;

[0064] S23, moving the entire sliding window to the right by one time point, and calculating the average values ​​of the heat index and energy consumption data in the new sliding window; until the sliding window covers all time points in the preset time window;

[0065] Specifically, the sliding window is moved to the right by one time point, that is, the leftmost data point is removed, and the next new data point is added to the rightmost side of the window. This step is repeated for the new sliding window, and the average values ​​of the heat index and energy consumption data in the new sliding window are calculated. This step is repeated until the sliding window covers all data points, and the average values ​​of the heat index and energy consumption data in each window are calculated.

[0066] S24, the average values ​​of the heat index and energy consumption data in each sliding window after sliding average processing are arranged in time series (which can be the initial time point, middle time point, or end time point of each sliding window, etc.) to construct a heat time series and an energy consumption time series.

[0067] Among them, Figure 3 As shown, step S3 specifically includes:

[0068] S31, calculating the change amount and change rate of energy consumption data at two adjacent time points in the energy consumption time series;

[0069] Specifically, the change in energy consumption data between two adjacent time points in the energy consumption time series can be calculated by subtracting the energy consumption value at the previous time point from the energy consumption value at the later time point; the change rate of energy consumption data between two adjacent time points can be calculated by dividing the change in energy consumption between the two adjacent time points by the time interval. This change rate reflects how fast the energy consumption changes over time.

[0070] S32, calculating the maximum change amount and change rate of energy consumption data at any two time points in the energy consumption time series;

[0071] Specifically, the change in energy consumption data between any two time points in the energy consumption time series can be calculated by subtracting the energy consumption value at the previous time point from the energy consumption value at the later time point; the change rate of energy consumption data between any two time points can be calculated by dividing the change in energy consumption at any two time points by the time interval. This change rate reflects how fast the energy consumption changes over time.

[0072] S33, extracting energy consumption time series characteristics, wherein the energy consumption time series characteristics include the peak value, valley value, average change rate, and maximum change amount of energy consumption at all time points of the energy consumption time series, and the change rate and maximum change amount of energy consumption at adjacent time points of the energy consumption time series;

[0073] Specifically, the variation range (amount of change) of the energy consumption time series is analyzed. The variation range can be obtained by calculating the difference between the maximum and minimum values ​​of the energy consumption data in the entire series, or by calculating the difference between each data point and its adjacent data points to evaluate the local variation range.

[0074] S34, merging the extracted energy consumption time series features into an energy consumption time series feature set.

[0075] Specifically, the energy consumption time series features are integrated into a feature set, which will comprehensively describe the energy consumption behavior pattern of the monitored equipment within a preset time window. This feature set not only contains the numerical information of energy consumption, but also includes the dynamic characteristics of energy consumption changing over time.

[0076] Among them, Figure 4 As shown, step S4 specifically includes:

[0077] S41, obtaining a heat change rate sequence for the heat change rates of the previous and next time points in the heat time series;

[0078] Specifically, for each time point in the heat time series, calculate its heat change rate with the previous time point. The formula is: heat change rate = (heat at the current time point - heat at the previous time point) / time interval, and you can get a heat change rate sequence, which reflects the speed of heat change over time.

[0079] S42, traversing the entire heat time series, and finding the maximum heat value and the minimum heat value in the entire heat time series;

[0080] Specifically, traverse the entire heat time series, find the maximum and minimum values ​​in the series, which correspond to the highest and lowest points of heat, respectively, and record the time points and heat values ​​corresponding to these extreme points, because they may be key indicators of changes in device status.

[0081] S43, based on the heat change rate sequence, extract the heat time series feature set, the heat time series features in the heat time series feature set include the average change rate, maximum change rate, minimum change rate of the entire heat change rate sequence, the maximum heat value and the minimum heat value in the entire heat time series, the time point corresponding to the maximum heat value in the entire heat time series, the time point corresponding to the minimum heat value, and the difference between the maximum heat value and the minimum heat value in the entire heat time series.

[0082] Based on the calculated heat change rate sequence, heat time series features such as average change rate, maximum change rate, minimum change rate, etc. are extracted. Combined with the highest and lowest heat points, features such as the time when the peak value appears, the time when the valley value appears, and the gap between the peak value and the valley value can be further extracted. The extracted features are integrated to form a comprehensive heat time series feature set, which will include statistical information on the heat change rate, extreme heat values ​​and their occurrence time, etc., providing a rich data foundation for subsequent anomaly detection.

[0083] Among them, Figure 5 As shown, step S5 specifically includes:

[0084] S51, obtaining an optimal weight combination corresponding to all heat time series features in a set of heat time series feature sets through optimization of a genetic algorithm, so that the correlation or similarity between the combined heat time series feature set and the energy consumption time series feature set is the highest;

[0085] S52, using an optimal weight combination to weightedly merge the heat time series features in the heat time series feature set and the energy consumption time series features in the energy consumption time series feature set to form a merged heat time series feature set.

[0086] like Figure 6 As shown, step S51 specifically includes:

[0087] S511, setting a corresponding initial weight for each heat time series feature in the heat time series feature set;

[0088] S512, based on a genetic algorithm, determining a fitness function for evaluating the weight according to the correlation between the heat time series characteristics and the energy consumption time series characteristics;

[0089] The calculation formula of the fitness function is:

[0090]

[0091] Among them, F(w) is the fitness function; w i is a vector containing n weights, where each w i Corresponding to the i-th heat time series feature h in the heat time series feature set H i The weight of h iis the value of the i-th heat time series feature at a certain moment or within a certain period of time; e j is the value of the jth energy consumption time series feature at a certain moment or within a certain period of time; n is the number of heat time series features in the heat time series feature set; m is the number of energy consumption time series features in the energy consumption time series feature set E.

[0092] Weight w i Indicates the importance of the i-th heat feature when merging heat features; H = [h1,h2,…,h n ] is a vector containing n thermal feature values, representing the thermal time series characteristics of the device. Each h i is a scalar, representing the value of the ith heat feature at a certain moment or within a certain period of time; E = [e1, e2, …, e m ] is a vector containing m energy consumption characteristic values, representing the energy consumption time series characteristic set of the device. Each e j is a scalar representing the value of the jth energy consumption feature at a certain moment or within a certain period of time; n is the number of heat time series features, that is, the dimension of vector H and weight vector w; m is the number of energy consumption time series feature sets, that is, the dimension of vector E.

[0093] The fitness function plays a vital role in the genetic algorithm. It is used to evaluate the quality of each individual (in this case, the weight combination). In step S5, the fitness function is designed to measure the correlation between the combined heat time series features and the energy consumption time series feature set. Specifically, the value of the fitness function F(w) reflects the degree of match between the heat time series features and the energy consumption time series features under a given weight combination w.

[0094] The calculation formula of the fitness function F(w) takes into account the difference between the heat time series characteristics and the energy consumption time series characteristics. This difference is calculated by calculating the combined heat time series characteristics (i.e., w i ×h i The cumulative sum of each energy consumption characteristic value e j The smaller the sum of squares, the closer the combined heat time series characteristics are to the energy consumption time series characteristics, that is, the better the weight combination is.

[0095] The fitness function F(w) is designed so that when the difference is minimized, the function value reaches the maximum, because the function form is When the difference is 0, F(w) reaches its maximum value of 1, indicating a perfect match. As the difference increases, the function value gradually decreases, indicating a decrease in the degree of match.

[0096] S513, taking each weight combination as an individual, and selecting individuals whose fitness function value is greater than a preset first fitness threshold (weight setting is better) to enter the next generation according to the fitness function;

[0097] S514, randomly selecting two individuals, performing a crossover operation with a preset crossover probability, and generating a new weight combination;

[0098] The crossover operation can be a single-point crossover, multi-point crossover, or uniform crossover, etc. The crossover probability refers to the probability of two individuals performing a crossover operation in a genetic algorithm. This probability is usually a preset value between 0 and 1. In practical applications, the setting of the crossover probability often depends on the characteristics of the problem and empirical judgment. A common practice is to conduct experiments to determine a crossover probability value that balances exploration and exploitation. For example, the crossover probability can be set to 0.7 or 0.8, which means that in each iteration, approximately 70% or 80% of the individuals will perform a crossover operation.

[0099] There is no specific formula for the crossover operation because it is a random process used to combine the genetic information of two parent individuals to produce new offspring individuals. Taking single-point crossover as an example, its operation can be described as:

[0100] ① Randomly select a crossover point, usually an integer between 1 and the individual length (i.e. the number of weights).

[0101] ② Keep the part of the two parent individuals (assumed to be Parent1 and Parent2) from the starting position to the intersection point unchanged.

[0102] ③Exchange the parts of the two parent individuals after the crossover point to generate two new offspring individuals.

[0103] For example, if Parent1 = [w1, w2, w3, w4, w5] and Parent2 = [v1, v2, v3, v4, v5], and the crossover point is 3, then the offspring after crossover may be Child1 = [w1, w2, w3, v4, v5] and Child2 = [v1, v2, v3, w4, w5].

[0104] S515, randomly adjusting a certain weight combination in the individuals with a preset mutation probability to increase the diversity of the population and avoid falling into a local optimum;

[0105] The mutation probability refers to the probability that the genes (here, weights) of an individual in a genetic algorithm will change randomly. Like the crossover probability, the mutation probability is also a value between 0 and 1, usually small, such as 0.01 or 0.001. This means that in each iteration, about 1% or 0.1% of the genes will mutate. The setting of the mutation probability is also to find a balance between exploration and exploitation, while increasing the diversity of the population to prevent the algorithm from converging to a local optimal solution too early.

[0106] The mutation operation is usually a stochastic process used to introduce small random changes in the genes of an individual. In genetic algorithms, mutation can be achieved by replacing, flipping, or swapping genes. For continuous-valued weights, the mutation operation may involve adding a small random perturbation to the current weight value. For example:

[0107] ① Select a gene to be mutated (i.e., weight), which is usually determined randomly based on the probability of mutation;

[0108] Generate a small random number (can be a normal or other distributed random number) and add it to the selected weight.

[0109] ③ Ensure that the mutated weight value is within a reasonable range. If necessary, operations such as truncation or modulus operation can be performed.

[0110] S516, calculating the fitness function of the adjusted individual;

[0111] S517, judging whether the preset termination condition is reached, if the judgment result is yes, executing step S518, if the judgment result is no, executing step S513; the preset termination condition includes reaching the maximum number of iterations or the fitness function value reaching the preset second fitness threshold; the preset second fitness threshold is greater than the preset first fitness threshold;

[0112] S518, obtaining a set of optimal weight combinations so that the correlation or similarity between the combined heat time series feature set and the energy consumption time series feature set is the highest.

[0113] like Figure 7 As shown, step S52 specifically includes:

[0114] S521, setting the optimal weight combination to a one-dimensional list or a one-dimensional array, wherein the one-dimensional list or the one-dimensional array includes weights corresponding to n heat time series features respectively;

[0115] Through genetic algorithms or other optimization methods, the optimal weight combination has been found, recorded as optimal_weights. This is a one-dimensional list or one-dimensional array containing n weights, where n is the number of heat time series features.

[0116] S522, setting the heat time series feature data to a two-dimensional array or a two-dimensional list, wherein the row data elements in the two-dimensional array or the two-dimensional list represent each time step in a preset time period, and the column data elements in the two-dimensional array or the two-dimensional list represent the heat time series feature in each time step in the preset time period;

[0117] The heat time series feature data is a two-dimensional array or two-dimensional list with a shape of (timesteps, features), where timesteps represents the number of time steps in a preset time period, and features represents the number of heat time series features in each time step in the preset time period. This data is called heat_features.

[0118] S523, create an empty list to store the weighted merged heat time series features; for the two-dimensional array or two-dimensional list, multiply each heat time series feature value in each time step with the corresponding weight in the one-dimensional list or one-dimensional array and perform weighted summation to obtain the merged heat time series feature value of the time step; add the merged heat time series feature value calculated in each time step to the empty list.

[0119] Create an empty list merged_heat_features to store the weighted merged heat time series features. For each time step in heat_features, use optimal_weights to perform a weighted merge. The specific approach is to multiply each feature value by its corresponding weight, and then add all the weighted feature values ​​to obtain the merged heat time series feature value of the time step; add the merged heat time series feature value calculated for each time step to the merged_heat_features list. Finally, the merged_heat_features list will contain the merged heat time series feature values ​​of all time steps.

[0120] In steps S6-S7, the merged heat time series features are obtained from step S5, which are recorded as merged_heat_features. This is a list or array containing heat features at each time point.

[0121] Set the heat threshold and change range:

[0122] According to historical data and the heat characteristics of the device when it is working normally, a heat critical value critical_heat_value and a variation range heat_variation_range are set. The variation range can be a specific value range, such as [lower_bound, upper_bound], which indicates the range of fluctuation allowed for the heat characteristics when it is working normally.

[0123] Specifically, the setting of the critical heat value (critical_heat_value) and the heat_variation_range is based on the understanding of the normal working status of the monitored equipment and the analysis of historical data. The setting of these values ​​is not purely empirical, but should be dynamically set according to multiple parameters to ensure accuracy and flexibility. For example, the setting of the critical heat value:

[0124] The critical heat value is a threshold used to determine whether the device may be abnormal. The following factors should be considered when setting this value:

[0125] Historical data: By analyzing the thermal data of the device during normal operation, a baseline thermal value can be determined. This baseline value can be the average, median, or other statistical value of the historical data, representing the thermal level of the device under normal operating conditions.

[0126] Equipment specifications and working environment: The equipment specifications (such as power, model, etc.) and working environment (such as temperature, humidity, ventilation conditions, etc.) will affect its normal working temperature. These factors should be taken into consideration to adjust the heat threshold.

[0127] Safety margin: To avoid false positives, a safety margin can be added to the baseline heat value to determine the critical value. This margin can be set based on the fluctuation range of the heat value in the historical data.

[0128] Taking all the above factors into consideration, the heat threshold can be dynamically set using the following formula:

[0129] critical_heat_value = baseline heat value + safety margin.

[0130] Setting of the change range:

[0131] The heat_variation_range defines the range of fluctuations allowed for the heat feature when it is working properly. This range should also be set based on multiple parameters:

[0132] Fluctuation range of historical data: By analyzing the fluctuation of heat values ​​in historical data, a normal range of variation can be determined. This can be achieved by calculating the standard deviation, interquartile range (IQR) or other statistics of historical data.

[0133] Working mode and load changes: The working mode and load changes of the equipment will affect its thermal fluctuation. For example, the temperature of the equipment may rise when it is running under high load, and it may fall when it is under low load or in standby state. These working modes should be taken into account to set a reasonable variation range.

[0134] External environmental factors: Changes in the external environment (such as seasonal changes, climate change, etc.) may also affect the thermal fluctuations of the equipment. These factors can be monitored and recorded to identify their impact on the thermals and adjust the range accordingly.

[0135] Based on the above considerations, the change range can be dynamically set in the following ways:

[0136] The preset change range is determined as follows:

[0137] heat_variation_range = [baseline heat value - lower limit fluctuation maximum value, baseline heat value + upper limit fluctuation range] = [lower_bound, upper_bound],

[0138] Among them, heat_variation_range is the preset variation range; the lower limit fluctuation value is the difference between the baseline heat value and the lower limit fluctuation maximum value; the upper limit fluctuation value is the sum of the baseline heat value and the upper limit fluctuation maximum value, and the lower limit fluctuation maximum value and the upper limit fluctuation maximum value are empirical values.

[0139] Traverse the combined heat time series features:

[0140] For each element in merged_heat_features (i.e. the merged heat feature value at each time point), check whether the heat feature is abnormal:

[0141] For each time point, the combined heat characteristic value current_heat_value is judged whether it exceeds the preset stable range. The specific judgment logic is as follows:

[0142] If current_heat_value is greater than critical_heat_value+upper_bound or less than critical_heat_value-lower_bound, it is considered an abnormal condition.

[0143] Once an abnormal situation is found, the heat characteristic value and the corresponding timestamp at that time point are recorded, and further actions can be taken, such as issuing an alarm, notifying management personnel, or performing automatic repairs.

[0144] After checking the combined heat feature values ​​of all time points, an anomaly report is generated, listing all detected abnormal time points, the corresponding heat feature values, possible causes and recommended measures.

[0145] By setting the critical value and variation range of heat, abnormalities in the working condition of the equipment can be discovered in time, so that corresponding measures can be taken to intervene and prevent equipment failure or performance degradation; through the automated abnormal detection process, the cost and time of manual inspections can be reduced, and the efficiency and response speed of equipment maintenance can be improved. At the same time, the generated abnormal report can provide valuable reference information for maintenance personnel, helping them to locate and solve problems more quickly.

[0146] The present invention can effectively reduce the noise and fluctuation in the data by performing sliding average processing on the heat index and energy consumption data, making the subsequent feature extraction and anomaly detection more accurate. At the same time, dynamic analysis and extraction of time series features can more finely capture the state changes of the equipment.

[0147] In the embodiment of the present invention, the abnormal detection of the working condition of the equipment can be reflected by the heat characteristics and energy consumption characteristics, mainly based on the following reasons:

[0148] (I) Thermal characteristics and device anomaly detection

[0149] Heat is associated with device status: The heat of a device is usually closely related to its working status. When an abnormality occurs in the internal components of the device, such as overload, short circuit, or poor lubrication, it often causes the local or overall temperature of the device to rise. Therefore, the heat feature can be used as an important indicator of the health of the device.

[0150] Fault early warning: Abnormal heat is often an early sign of equipment failure. By monitoring thermal changes, potential problems can be discovered before equipment failure occurs, allowing preventive maintenance to be performed in a timely manner.

[0151] Safety considerations: Excessive temperatures may cause equipment damage or even fire and other safety accidents. Therefore, monitoring thermal characteristics is critical to ensure safe operation of equipment.

[0152] 2. Energy consumption characteristics and equipment anomaly detection

[0153] Energy consumption and work efficiency: The energy consumption of equipment is closely related to its work efficiency and working status. When the equipment is worn, blocked or has other faults, its energy consumption pattern may change. By monitoring the energy consumption characteristics, the working status of the equipment can be indirectly inferred.

[0154] Cost-benefit analysis: Abnormal energy consumption patterns may mean that the energy efficiency of the equipment is reduced, which not only increases operating costs but may also have a negative impact on the environment. By detecting abnormal energy consumption, energy efficiency issues can be discovered and resolved in a timely manner.

[0155] Fault location: Certain types of faults (such as leakage, blockage, etc.) may cause a significant increase in equipment energy consumption. Analysis of energy consumption data helps locate these fault points.

[0156] (III) Relationship with equipment temperature monitoring

[0157] Equipment temperature monitoring usually directly measures the temperature of the equipment through temperature sensors installed on the equipment. This monitoring method provides real-time data of the equipment temperature, while thermal features are higher-level information extracted based on these real-time data, which are used to reflect the overall thermal state of the equipment and its changing trends. Thermal features can be regarded as the further processing and analysis results of temperature monitoring data, which can help staff better understand the thermal behavior of the equipment and judge the working status of the equipment accordingly.

[0158] (IV) Relationship with equipment working condition monitoring

[0159] Equipment working condition monitoring is a broader concept, which includes monitoring of temperature, pressure, vibration, sound and other aspects. Thermal characteristics and energy consumption characteristics are two important aspects of equipment working condition monitoring. They provide intuitive information about the equipment's operating status and working efficiency, and help evaluate the overall performance and health of the equipment. By comprehensively analyzing these characteristics, we can have a more comprehensive understanding of the equipment's working conditions and promptly discover and deal with potential problems.

[0160] In summary, thermal characteristics and energy consumption characteristics play an important role in equipment anomaly detection, and they are closely related to equipment temperature monitoring and equipment working condition monitoring.

[0161] The present invention obtains the heat index and energy consumption data of the monitored equipment at multiple specified moments within a preset time window; wherein the heat index is a weighted value according to the equipment temperature, the equipment energy consumption and the equipment operating status; the heat index and energy consumption data at the specified moment are arranged in time series to construct a heat time series and an energy consumption time series; the energy consumption time series is dynamically analyzed to obtain an energy consumption time series feature set; the heat time series is parsed to extract a heat time series feature set; according to the energy consumption time series feature set, the extracted heat time series features are weighted and merged to form a merged heat time series feature set; according to the merged heat time series features, by presetting a heat critical value and a preset variation interval, it is identified whether the working condition of the monitored equipment is abnormal, thereby effectively solving the problem of low reliability of power equipment status temperature monitoring caused by the prior art, and effectively improving the reliability of power equipment status temperature monitoring.

[0162] The thermal index is used in the technical solution of the present invention. Compared with using only temperature to monitor the status of the device, the thermal index can comprehensively consider multiple factors related to the heat of the device, rather than just the temperature, thereby providing a more comprehensive evaluation of the thermal performance of the device; by integrating multiple parameters, the thermal index can more keenly capture possible problems with the device, such as failure or overload of the cooling system, thereby issuing an early warning; the thermal index can be customized and optimized according to specific needs, and by adjusting the parameters or weights in the algorithm, it can be more adapted to the monitoring needs of specific equipment; for some complex systems or equipment, a single temperature measurement may not be sufficient to fully reflect its thermal performance; the thermal index can more accurately evaluate the status of these systems by integrating multiple related factors.

[0163] The energy consumption time series features in the technical solution of the present invention include the peak value, valley value, average change rate, maximum change amount of energy consumption at all time points in the energy consumption time series, and the change rate and maximum change amount of energy consumption at adjacent time points in the energy consumption time series; the heat time series features in the heat time series feature set include the average change rate, maximum change rate, minimum change rate of the entire heat change rate sequence, the maximum heat value and the minimum heat value in the entire heat time series, the time point corresponding to the maximum heat value in the entire heat time series, the time point corresponding to the minimum heat value, and the difference between the maximum heat value and the minimum heat value in the entire heat time series, which not only includes the numerical information of energy consumption and heat data, but also includes the dynamic characteristics of energy consumption and heat data changing over time, etc., providing a rich data basis for subsequent anomaly detection.

[0164] In the technical solution of the present invention, through the optimization of the genetic algorithm, the optimal weight combination corresponding to all the thermal time series features in a set of thermal time series feature sets is obtained, so that the correlation or similarity between the merged thermal time series feature set and the energy consumption time series feature set is the highest; the reliability of the state temperature monitoring of the power equipment is further improved, the actual working condition of the equipment can be more accurately reflected, and the sensitivity of abnormality detection is improved.

[0165] Embodiment 2

[0166] like Figure 8 As shown, the technical solution of the present invention also provides a power equipment state temperature monitoring system, comprising:

[0167] The acquisition module 101 acquires the heat index and energy consumption data of the device to be monitored at multiple specified moments within a preset time window; wherein the heat index is a weighted value based on the device temperature, the device energy consumption and the device operating status;

[0168] The construction module 102 arranges the heat index and energy consumption data at a specified time in time series to construct a heat time series and an energy consumption time series;

[0169] The analysis module 103 performs dynamic analysis on the energy consumption time series, evaluates the change rate and change range of the energy consumption data at each specified time, and obtains the energy consumption time series feature set;

[0170] The analysis module 104 analyzes the heat time series, identifies the heat change rate, the highest heat point and the lowest heat point at each specified time, and extracts the heat time series feature set;

[0171] The adjustment and merging module 105 adjusts and merges the extracted heat time series features according to the energy consumption time series feature set to form a merged heat time series feature set;

[0172] The identification module 106 identifies whether the working condition of the monitored equipment is abnormal based on the merged heat time series feature set, through the preset heat critical value and the preset change interval. If the merged heat time series feature is greater than the sum of the preset heat critical value and the upper limit fluctuation value of the preset change interval, or is less than the difference between the preset heat critical value and the lower limit fluctuation value of the preset change interval, it is deemed to be an abnormal condition.

[0173] The implementation process of the monitoring and acquisition module 101, the construction module 102, the analysis module 103, the parsing module 104, the adjustment and merging module 105, and the identification module 106 in the second embodiment corresponds to the method steps in the first embodiment, and will not be repeated in this second embodiment.

[0174] The present invention obtains the heat index and energy consumption data of the monitored equipment at multiple specified moments within a preset time window; wherein the heat index is a weighted value according to the equipment temperature, the equipment energy consumption and the equipment operating status; the heat index and energy consumption data at the specified moment are arranged in time series to construct a heat time series and an energy consumption time series; the energy consumption time series is dynamically analyzed to obtain an energy consumption time series feature set; the heat time series is parsed to extract a heat time series feature set; according to the energy consumption time series feature set, the extracted heat time series features are weighted and merged to form a merged heat time series feature set; according to the merged heat time series features, by presetting a heat critical value and a preset variation interval, it is identified whether the working condition of the monitored equipment is abnormal, thereby effectively solving the problem of low reliability of power equipment status temperature monitoring caused by the prior art, and effectively improving the reliability of power equipment status temperature monitoring.

[0175] The thermal index is used in the technical solution of the present invention. Compared with using only temperature to monitor the status of the device, the thermal index can comprehensively consider multiple factors related to the heat of the device, rather than just the temperature, thereby providing a more comprehensive evaluation of the thermal performance of the device; by integrating multiple parameters, the thermal index can more keenly capture possible problems with the device, such as failure or overload of the cooling system, thereby issuing an early warning; the thermal index can be customized and optimized according to specific needs, and by adjusting the parameters or weights in the algorithm, it can be more adapted to the monitoring needs of specific equipment; for some complex systems or equipment, a single temperature measurement may not be sufficient to fully reflect its thermal performance; the thermal index can more accurately evaluate the status of these systems by integrating multiple related factors.

[0176] The energy consumption time series features in the technical solution of the present invention include the peak value, valley value, average change rate, maximum change amount of energy consumption at all time points in the energy consumption time series, and the change rate and maximum change amount of energy consumption at adjacent time points in the energy consumption time series; the heat time series features in the heat time series feature set include the average change rate, maximum change rate, minimum change rate of the entire heat change rate sequence, the maximum heat value and the minimum heat value in the entire heat time series, the time point corresponding to the maximum heat value in the entire heat time series, the time point corresponding to the minimum heat value, and the difference between the maximum heat value and the minimum heat value in the entire heat time series, which not only includes the numerical information of energy consumption and heat data, but also includes the dynamic characteristics of energy consumption and heat data changing over time, etc., providing a rich data basis for subsequent anomaly detection.

[0177] In the technical solution of the present invention, through the optimization of genetic algorithm, the optimal weight combination corresponding to all thermal time series features in a set of thermal time series feature sets is obtained, so that the correlation or similarity between the merged thermal time series feature set and the energy consumption time series feature set is the highest, further improving the reliability of power equipment status temperature monitoring.

[0178] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A method for monitoring the temperature of an electric power device, characterized in that: include: Obtain the heat index and energy consumption data of the monitored device at multiple specified times within a preset time window; wherein the heat index is a weighted value based on the device temperature, device energy consumption and device operating status; Arrange the heat index and energy consumption data at the specified time in time series to construct heat time series and energy consumption time series; Perform dynamic analysis on the energy consumption time series, evaluate the change rate and amplitude of energy consumption data at each specified time, and obtain the energy consumption time series feature set; Analyze the heat time series, identify the heat change rate, the highest heat point and the lowest heat point at each specified time, and extract the heat time series feature set; According to the energy consumption time series feature set, the extracted heat time series features are weighted and merged to form a merged heat time series feature set; According to the merged heat time series characteristics, through the preset heat critical value and the preset change range, it is identified whether the working condition of the monitored equipment is abnormal. If the merged heat time series characteristics are greater than the sum of the preset heat critical value and the upper limit fluctuation value of the preset change range, or less than the difference between the preset heat critical value and the lower limit fluctuation value of the preset change range, it is considered to be an abnormal condition.

2. A method for monitoring the state temperature of electric equipment according to claim 1, characterized in that: The heat index is a weighted value based on the device temperature, device energy consumption, and device operating status. Specifically: H=α·T+β·P+γ·R+δ·S; Where H is the heat index; T is the real-time temperature of the device; P is the energy consumption of the device; R is the heat dissipation performance score of the device; S is the continuous operation time of the device; α, β, γ, and δ are weight coefficients; The thermal performance score of a device is calculated as follows: Where k is the thermal conductivity coefficient of the equipment cooling system; T max It is the highest temperature reached by the equipment cooling system during operation; T env is the ambient temperature of the equipment; c is the temperature fluctuation coefficient; ΔT is the temperature fluctuation range of the equipment cooling system when it is in stable operation.

3. A method for monitoring the temperature of an electric power device according to claim 1, characterized in that: Arrange the heat index and energy consumption data at a specified time in time series to construct the heat time series and energy consumption time series. Specifically, the following are included: Determine the initial sliding time point of the sliding window of the sliding average filter and the size of the sliding window; Calculate the average values ​​of heat index and energy consumption data in the current sliding window respectively; Move the sliding window rightward by one time point as a whole, and calculate the average value of the heat index and energy consumption data in the new sliding window; until the sliding window covers all time points in the preset time window; The average values ​​of the heat index and energy consumption data in each sliding window after sliding average processing are arranged in time series to construct heat time series and energy consumption time series.

4. A method for monitoring the temperature of an electric power device according to claim 1, characterized in that: Dynamic analysis of energy consumption time series is performed to evaluate the change rate and amplitude of energy consumption data at each specified time, and the energy consumption time series feature set is obtained, which specifically includes: Calculate the change amount and change rate of energy consumption data at two adjacent time points in the energy consumption time series; Calculate the maximum change and change rate of energy consumption data at any two time points in the energy consumption time series; Extracting energy consumption time series features, wherein the energy consumption time series features include the peak value, valley value, average change rate, and maximum change amount of energy consumption at all time points in the energy consumption time series, and the change rate and maximum change amount of energy consumption at adjacent time points in the energy consumption time series; The extracted energy consumption time series features are merged into an energy consumption time series feature set.

5. A method for monitoring the state temperature of electric equipment according to claim 1, characterized in that: Analyze the heat time series, identify the heat change rate, the highest heat point and the lowest heat point at each specified time, and extract the heat time series feature set, including: For the heat change rates of the previous and next time points in the heat time series, a heat change rate sequence is obtained; Traverse the entire heat time series and find the maximum and minimum heat values ​​in the entire heat time series; Based on the heat change rate sequence, a heat time series feature set is extracted, wherein the heat time series features in the heat time series feature set include the average change rate, maximum change rate, minimum change rate of the entire heat change rate sequence, the maximum heat value and the minimum heat value in the entire heat time series, the time point corresponding to the maximum heat value in the entire heat time series, the time point corresponding to the minimum heat value, and the difference between the maximum heat value and the minimum heat value in the entire heat time series.

6. A method for monitoring the temperature of an electric power device according to claim 1, characterized in that: According to the energy consumption time series feature set, the extracted heat time series features are weighted and merged to form a merged heat time series feature set, which specifically includes: Through the optimization of the genetic algorithm, the optimal weight combination corresponding to all the heat time series features in a set of heat time series feature sets is obtained, so that the correlation or similarity between the combined heat time series feature set and the energy consumption time series feature set is the highest; The heat time series features in the heat time series feature set and the energy consumption time series features in the energy consumption time series feature set are weightedly merged using the optimal weight combination to form a merged heat time series feature set.

7. A method for monitoring the state temperature of electric equipment according to claim 6, characterized in that: Through the optimization of the genetic algorithm, the optimal weight combination corresponding to all the heat time series features in a set of heat time series features is obtained, which specifically includes: Set the corresponding initial weight for each heat time series feature in the heat time series feature set; Based on the genetic algorithm, the fitness function used to evaluate the weights is determined according to the correlation between the heat time series characteristics and the energy consumption time series characteristics; Take each weight combination as an individual, and select individuals whose fitness function value is greater than a preset first fitness threshold to enter the next generation according to the fitness function; Randomly select two individuals and perform crossover operation with a preset crossover probability to generate a new weight combination; Randomly adjust a certain weight combination in the individual with a preset mutation probability to increase the diversity of the population and avoid falling into the local optimum; Iteratively calculate the fitness function of the adjusted individuals until the preset termination condition is met, and obtain a set of optimal weight combinations so that the correlation or similarity between the combined thermal time series characteristics and the energy consumption time series characteristic set is the highest; the preset termination condition includes reaching the maximum number of iterations or the fitness function value reaching a preset second fitness threshold.

8. A method for monitoring the state temperature of electric equipment according to claim 7, characterized in that: The calculation formula of the fitness function is: Among them, F(w) is the fitness function; w i is a vector containing n weights, where each w i Corresponding to the i-th heat time series feature h in the heat time series feature set H i The weight of h i is the value of the i-th heat time series feature at a certain moment or within a certain period of time; e j is the value of the jth energy consumption time series feature at a certain moment or within a certain period of time; n is the number of heat time series features in the heat time series feature set; m is the number of energy consumption time series features in the energy consumption time series feature set E.

9. A method for monitoring the state temperature of electric equipment according to claim 6, characterized in that: The heat time series features in the heat time series feature set and the energy consumption time series features in the energy consumption time series feature set are weightedly merged using the optimal weight combination to form a merged heat time series feature set, which specifically includes: The optimal weight combination is set to a one-dimensional list or a one-dimensional array, wherein the one-dimensional list or the one-dimensional array includes weights corresponding to the n heat time series features respectively; The heat time series feature data is set as a two-dimensional array or a two-dimensional list, wherein the row data elements in the two-dimensional array or the two-dimensional list represent each time step in a preset time period, and the column data elements in the two-dimensional array or the two-dimensional list represent the heat time series feature in each time step in the preset time period; Create an empty list to store the weighted merged heat time series features; for a two-dimensional array or a two-dimensional list, multiply each heat time series feature value in each time step by the corresponding weight in the one-dimensional list or the one-dimensional array and perform weighted summation to obtain the merged heat time series feature value of the time step; add the merged heat time series feature value calculated in each time step to the empty list.

10. A power equipment status temperature monitoring system, characterized in that: include: An acquisition module is used to acquire the heat index and energy consumption data of the device to be monitored at multiple specified times within a preset time window; wherein the heat index is a weighted value based on the device temperature, device energy consumption and device operating status; The construction module arranges the heat index and energy consumption data at a specified time in time series to construct the heat time series and energy consumption time series; The analysis module dynamically analyzes the energy consumption time series, evaluates the change rate and amplitude of the energy consumption data at each specified time, and obtains the energy consumption time series feature set; The parsing module parses the heat time series, identifies the heat change rate, the highest heat point and the lowest heat point at each specified time, and extracts the heat time series feature set; The adjustment and merging module adjusts and merges the extracted heat time series features according to the energy consumption time series feature set to form a merged heat time series feature set; The identification module identifies whether the working condition of the monitored equipment is abnormal based on the merged heat time series feature set, through the preset heat critical value and the preset change interval. If the merged heat time series feature is greater than the sum of the preset heat critical value and the upper limit fluctuation value of the preset change interval, or is less than the difference between the preset heat critical value and the lower limit fluctuation value of the preset change interval, it is considered to be an abnormal condition.

Images