Alarm method, device and equipment of switch cabinet, medium and product

Abstract

The embodiment of the invention provides an alarm method, device and equipment of a switch cabinet, a medium and a product. The method comprises the following steps: firstly, acquiring real-time load data of a target switch cabinet; then, a real-time temperature sequence and a temperature threshold value are determined according to the real-time load data; then, inputting the real-time load data into the temperature prediction model to obtain a predicted temperature sequence corresponding to the real-time load data; and finally, inputting the real-time temperature sequence, the predicted temperature sequence and the temperature threshold value into a false alarm judgment model to obtain a judgment result about whether the alarm signal is a false alarm signal or not. Through the method, the accuracy of judging the false alarm of the switch cabinet is improved.

Description

Technical Field

[0001] This application relates to the field of switchgear alarm technology, and in particular to an alarm method, device, equipment, medium and product for switchgear. Background Technology

[0002] High-current switchgear, as a critical power distribution device in power systems, directly impacts the stability and security of the power supply system. With the continuous expansion of new energy integration and the increasing complexity of power system operating conditions, the electrical loads borne by switchgear exhibit high volatility, high nonlinearity, and frequent time-varying characteristics. Conductive components such as switchgear contacts are prone to overheating under high load and frequent fluctuations. If abnormal temperatures are not detected in time, it can lead to increased contact resistance, aging or even burnout of insulation materials, resulting in equipment failure or large-scale power outages. Temperature alarms provide early warnings before thermal damage occurs, guiding maintenance personnel to take measures such as load reduction and maintenance, thereby ensuring the safe and stable operation of the power supply system and reducing the risk of power outages and economic losses.

[0003] In existing technologies, a static threshold judgment mechanism is generally used for switchgear temperature monitoring and alarm. This method sets a fixed temperature threshold, and an alarm is triggered when the operating temperature of the switchgear exceeds the threshold. Its main implementation involves installing temperature sensors on key parts of the switchgear to collect temperature data in real time and comparing it with a preset static threshold. Once the threshold is exceeded, an alarm signal is issued.

[0004] However, existing switchgear alarm methods are difficult to accurately generate alarms, and are prone to false alarms or missed alarms, resulting in low fault diagnosis efficiency. Summary of the Invention

[0005] This application provides alarm methods, devices, equipment, media, and products for switchgear, in order to solve the problems in the prior art that make it difficult to accurately generate alarms, easily generate false alarms or missed alarms, and result in low fault diagnosis efficiency.

[0006] In a first aspect, embodiments of this application provide an alarm method for a switchgear, comprising:

[0007] Obtain the real-time load data of the target switchgear;

[0008] The real-time temperature sequence and temperature threshold are determined based on the real-time load data.

[0009] The real-time load data is input into the temperature prediction model to obtain the predicted temperature sequence corresponding to the real-time load data;

[0010] The real-time temperature sequence, the predicted temperature sequence, and the temperature threshold are input into the false alarm discrimination model to obtain the result of whether the target switchgear needs to generate an alarm.

[0011] In one possible implementation, determining the real-time temperature sequence and temperature threshold based on the real-time load data includes:

[0012] The real-time load data is analyzed to obtain the load average, load fluctuation degree, instantaneous change rate, and historical trend change rate corresponding to the real-time load data;

[0013] The real-time temperature sequence is determined based on the average load, the degree of load fluctuation, the instantaneous rate of change, and the historical trend rate of change.

[0014] The temperature threshold is determined based on the real-time temperature sequence.

[0015] In one possible implementation, the training process of the temperature prediction model includes:

[0016] Obtain a set of operating states for the sample switchgear, the set of operating states including fault state operating data and normal state operating data of the sample switchgear;

[0017] A dataset mapping the load to the contact temperature of the sample switchgear is constructed based on the set of operating states.

[0018] The temperature prediction model is trained based on the mapping dataset; the temperature prediction model is constructed based on a long short-term memory network.

[0019] In one possible implementation, the step of inputting the real-time temperature sequence, the predicted temperature sequence, and the temperature threshold into the false alarm discrimination model to obtain a result on whether the target switchgear needs to generate an alarm includes:

[0020] The residual value between the real-time temperature sequence and the predicted temperature sequence is calculated using the false alarm discrimination model.

[0021] When the false alarm discrimination model determines that the residual value meets the temperature threshold, it outputs the result that the target switchgear needs to generate an alarm.

[0022] When the false alarm discrimination model determines that the residual value does not meet the temperature threshold, it outputs the result that the target switchgear does not need to generate an alarm.

[0023] In one possible implementation, after outputting the result that the target switchgear needs to generate an alarm, the method further includes:

[0024] Acquire the retest load data within a preset time window, as well as the process load data before receiving the alarm signal;

[0025] An extended dataset is formed based on the retested load data and the process load data;

[0026] The temperature prediction model is trained based on the extended dataset to update the network parameters of the temperature prediction model.

[0027] In one possible implementation, after training the temperature prediction model based on the extended dataset, the method further includes:

[0028] The real-time load data is input into the updated temperature prediction model to obtain the updated predicted temperature sequence corresponding to the real-time load data;

[0029] Based on the updated predicted temperature sequence and the determination result that the alarm signal is not a false alarm signal, the parameters of the false alarm determination model are optimized.

[0030] Secondly, embodiments of this application provide an alarm device for a switchgear, comprising:

[0031] The acquisition module is used to acquire the real-time load data of the target switchgear;

[0032] The determination module is used to determine the real-time temperature sequence and temperature threshold based on the real-time load data;

[0033] The first processing module is used to input the real-time load data into the temperature prediction model to obtain a predicted temperature sequence corresponding to the real-time load data.

[0034] The second processing module is used to input the real-time temperature sequence, the predicted temperature sequence, and the temperature threshold into the false alarm discrimination model to obtain the result of whether the target switch cabinet needs to generate an alarm.

[0035] In one possible implementation, the determining module is specifically used for:

[0036] The real-time load data is analyzed to obtain the load average, load fluctuation degree, instantaneous change rate, and historical trend change rate corresponding to the real-time load data;

[0037] The real-time temperature sequence is determined based on the average load, the degree of load fluctuation, the instantaneous rate of change, and the historical trend rate of change.

[0038] The temperature threshold is determined based on the real-time temperature sequence.

[0039] In one possible implementation, the false alarm detection device for the switchgear further includes a training module, specifically used for:

[0040] Obtain a set of operating states for the sample switchgear, the set of operating states including fault state operating data and normal state operating data of the sample switchgear;

[0041] A dataset mapping the load to the contact temperature of the sample switchgear is constructed based on the set of operating states.

[0042] The temperature prediction model is trained based on the mapping dataset; the temperature prediction model is constructed based on a long short-term memory network.

[0043] In one possible implementation, the second processing module is specifically used for:

[0044] The residual value between the real-time temperature sequence and the predicted temperature sequence is calculated using the false alarm discrimination model.

[0045] When the false alarm discrimination model determines that the residual value meets the temperature threshold, it outputs the result that the target switchgear needs to generate an alarm.

[0046] When the false alarm discrimination model determines that the residual value does not meet the temperature threshold, it outputs the result that the target switchgear does not need to generate an alarm.

[0047] In one possible implementation, after the output of the result indicating that the target switchgear needs to generate an alarm, the second processing module is further configured to:

[0048] Acquire the retest load data within a preset time window, as well as the process load data before receiving the alarm signal;

[0049] An extended dataset is formed based on the retested load data and the process load data;

[0050] The temperature prediction model is trained based on the extended dataset to update the network parameters of the temperature prediction model.

[0051] In one possible implementation, after training the temperature prediction model based on the extended dataset, the second processing module is further configured to:

[0052] The real-time load data is input into the updated temperature prediction model to obtain the updated predicted temperature sequence corresponding to the real-time load data;

[0053] Based on the updated predicted temperature sequence and the determination result that the alarm signal is not a false alarm signal, the parameters of the false alarm determination model are optimized.

[0054] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0055] The memory stores computer-executed instructions;

[0056] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0057] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0058] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0059] The alarm method, apparatus, equipment, medium, and product for switchgear provided in this application first acquire real-time load data of the target switchgear; then, the temperature observation window length and threshold size are adaptively determined using the real-time load data, so that the threshold is dynamically adjusted according to the operating conditions (high load, low load, gradual change, sudden change), thus avoiding the use of fixed thresholds for alarm judgment in traditional methods and reducing false alarms and missed alarms; then, the real-time load data is input into a temperature prediction model to obtain a predicted temperature sequence corresponding to the real-time load data, generating a desired temperature sequence that matches the current operating conditions; finally, the real-time temperature sequence, the predicted temperature sequence, and the temperature threshold are input into a false alarm discrimination model to obtain a result on whether the target switchgear needs to generate an alarm, which can effectively distinguish between normal temperature rise caused by changes in operating conditions and true abnormal thermal runaway, reducing invalid alarms and avoiding missed alarms; at the same time, the result can also be used as a monitoring signal for closed-loop optimization for subsequent model and threshold parameter correction. Attached Figure Description

[0060] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0061] Figure 1 A flowchart illustrating the alarm method for switchgear provided in this application embodiment. Figure 1 ;

[0062] Figure 2 A flowchart illustrating the alarm method for switchgear provided in this application embodiment. Figure 2 ;

[0063] Figure 3 A schematic diagram of the alarm device for the switchgear provided in this application embodiment;

[0064] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0065] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0066] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0067] High-current switchgear, as a critical power distribution device in power systems, directly impacts the stability and security of the power supply system. With the continuous expansion of new energy integration and the increasing complexity of power system operating conditions, the electrical loads borne by switchgear exhibit high volatility, high nonlinearity, and frequent time-varying characteristics. Conductive components such as switchgear contacts are prone to overheating under high load and frequent fluctuations. If abnormal temperatures are not detected in time, it can lead to increased contact resistance, aging or even burnout of insulation materials, resulting in equipment failure or large-scale power outages. Temperature alarms provide early warnings before thermal damage occurs, guiding maintenance personnel to take measures such as load reduction and maintenance, thereby ensuring the safe and stable operation of the power supply system and reducing the risk of power outages and economic losses.

[0068] In existing technologies, a static threshold judgment mechanism is generally used for switchgear temperature monitoring and alarm. This method sets a fixed temperature threshold, and an alarm is triggered when the operating temperature of the switchgear exceeds the threshold. Its main implementation involves installing temperature sensors on key parts of the switchgear to collect temperature data in real time and comparing it with a preset static threshold. Once the threshold is exceeded, an alarm signal is issued.

[0069] However, existing technologies use fixed thresholds, which cannot dynamically adjust alarm thresholds according to load changes. In scenarios with gradual changes in high load or sudden changes in low load, false alarms or missed alarms are easily generated, and the technologies cannot adapt to the dynamic evolution process under complex load backgrounds. At the same time, existing technologies do not fully explore the temporal coupling between load and temperature, and ignore the modulation effect of load fluctuations on temperature trends, resulting in a lack of adaptability and specificity in temperature alarm strategies.

[0070] Based on this, this application proposes an alarm method for switchgear. Addressing the inaccuracy of alarms in existing technologies, this application introduces a load sensing mechanism to dynamically quantify characteristics such as load average, volatility, instantaneous rate of change, and historical trend rate of change. Combining a regression model with a Long Short-Term Memory (LSTM) time-series prediction model, a dynamic mapping relationship between load and temperature is established. This allows for adaptive adjustment of the temperature alarm threshold under different load conditions. After an alarm is triggered, the model and discrimination parameters are continuously optimized through closed-loop correction using retest data. This approach captures the dynamic characteristics of temperature response under complex operating conditions such as gradual changes in high load and sudden changes in low load, while reducing false alarms and missed alarms caused by fixed thresholds, significantly improving the adaptability and accuracy of the alarm strategy.

[0071] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0072] Figure 1 A flowchart illustrating the alarm method for switchgear provided in this application embodiment. Figure 1 ;like Figure 1 As shown, the method includes:

[0073] S101. Obtain the real-time load data of the target switchgear.

[0074] S102. Determine the real-time temperature sequence and temperature threshold based on the real-time load data.

[0075] In one feasible approach, the real-time load data is first analyzed to obtain the load average, load fluctuation degree, instantaneous change rate, and historical trend change rate corresponding to the real-time load data; then, the real-time temperature sequence is determined based on the load average, load fluctuation degree, instantaneous change rate, and historical trend change rate; finally, the temperature threshold is determined based on the real-time temperature sequence.

[0076] It should be noted that before analyzing the real-time load data, the set of switchgear operating states must first be defined. ,in Indicates a fault state. This indicates the normal state. Constructing a current load. Temperature of switchgear contacts Mapping relationship dataset By mapping the dataset to real-time load data, the real-time temperature sequence can be determined.

[0077]

[0078] In the formula, For status identifiers ( ), For state The number of samples, For discrete time points, This represents the total number of samples. Representing state No. Samples in Current load value at time t, This represents the corresponding contact temperature value. This dataset was obtained through actual measurements during offline operation and covers the temperature response characteristics under different load conditions.

[0079] Understandably, by establishing the aforementioned mapping dataset, it is possible to cover temperature responses under different load conditions and operating states, providing a data foundation for subsequent load-based temperature prediction models and threshold adjustments.

[0080] Furthermore, to achieve dynamic monitoring of the switchgear operating temperature and adaptive adjustment of alarm values, a load sensing mechanism needs to be introduced to quantify and analyze load characteristics, thereby constructing a temperature data sequence based on load sensing. Four key characteristics are extracted:

[0081]

[0082] In the formula, This is the load average, used to quantify the current load level. The degree of load fluctuation is used to measure load instability (high fluctuation indicates that the temperature may change drastically in a short period of time). The instantaneous rate of change represents the deviation of the current load from the near historical average. The historical trend change rate is represented by an angular average to reflect the recent upward / downward trend of the load. Window size for calculating the mean. Indexed by the current time. The sampling interval is... Historical offset in instantaneous change calculation ( ), Historical offset in trend analysis ( ), For the time-sliding window index.

[0083] Furthermore, based on the load average The temperature threshold is determined dynamically (i.e., the interval length used in the temperature series). Specifically:

[0084]

[0085] In the formula, As the benchmark scaling factor, and This is the weighting adjustment parameter. The actual acquired temperature sequence is... ,in If the length of the sequence is the number of units rounded down, then the lower bound of the interval is... The final interval definition is... .

[0086] For example, calculated If the value is 2.23343, then the length of the sequence after rounding is... Assuming the measurement is Therefore, the interval length for the temperature series is set to... .

[0087] It should be noted that the temperature threshold is dynamically determined using the average load because the normal fluctuation range of switchgear contact temperature is closely related to the current operating load level: the higher the load, the greater the absolute value and amplitude of the temperature response; the lower the load, the relatively smaller the temperature fluctuation range. By dynamically calculating the range length based on the average load, the temperature threshold can be adjusted in real time according to the operating conditions. This avoids oversensitivity leading to false alarms under high loads and avoids excessively wide thresholds leading to missed alarms under low loads, thus achieving adaptive temperature monitoring for different load backgrounds.

[0088] S103. Input the real-time load data into the temperature prediction model to obtain the predicted temperature sequence corresponding to the real-time load data.

[0089] In one feasible approach, the training process of the temperature prediction model includes: first, obtaining a set of operating states of sample switchgear; then, constructing a mapping relationship dataset between load and contact temperature of sample switchgear based on the set of operating states; and finally, training the temperature prediction model based on the mapping dataset.

[0090] The set of operating states includes fault-state operating data and normal-state operating data of the sample switchgear; the temperature prediction model is built based on a long short-term memory network.

[0091] It should be noted that the process of constructing the switchgear operating status set has been described in embodiment S102, and will not be repeated here; while the process of constructing the temperature prediction sequence based on the mapping dataset and load characteristics is as follows:

[0092]

[0093] In the formula, For model parameters, For the first The weight coefficients of each sample, The width parameter of the Gaussian kernel function. This is the Gaussian kernel function.

[0094] Understandably, by utilizing a trained temperature prediction model and real-time load data, a predicted temperature sequence can be obtained. This allows for the generation of dynamic expected values ​​for equipment contact temperatures under current operating conditions, establishing a time-series benchmark relationship between load and temperature. By comparing actual temperature changes with predicted values, the difference between temperature rises caused by normal load fluctuations and abnormal temperature rises caused by potential faults can be accurately identified. This not only avoids misjudgments caused by relying solely on fixed temperature thresholds but also improves the accuracy and specificity of alarm determination, providing a reliable reference baseline for false alarm identification.

[0095] S104. Input the real-time temperature sequence, the predicted temperature sequence, and the temperature threshold into the false alarm discrimination model to obtain the result of whether the target switchgear needs to generate an alarm.

[0096] It should be noted that the false alarm discrimination model is defined as follows:

[0097]

[0098] In the formula, For indicator functions, For the threshold, For model parameters, when If the alarm is true, it is considered a genuine alarm; otherwise, it is considered a false alarm, meaning no alarm result is needed. This is because the model determines that although the current temperature deviation does not meet the temperature threshold, the actual equipment operation is normal. The temperature anomaly may be a temporary fluctuation or sensor error, and no emergency handling is required.

[0099] Understandably, by combining a temperature prediction model and a false alarm discrimination model, accurate judgment of switchgear temperature anomalies is achieved. The temperature prediction model accurately predicts the expected temperature change trend of the equipment based on load characteristics, while the false alarm discrimination model determines whether an alarm should be issued by comparing residuals with thresholds. This effectively reduces false alarms caused by load fluctuations or temporary disturbances, while ensuring that genuine fault alarms are not missed, thus significantly improving the accuracy and reliability of temperature monitoring.

[0100] The alarm method for switchgear provided in this application first acquires the real-time load data of the target switchgear; then, it adaptively determines the length of the temperature observation window and the threshold size using the real-time load data, so that the threshold is dynamically adjusted according to the operating conditions (high load, low load, gradual change, sudden change), thus avoiding the use of a fixed threshold for alarm judgment in traditional methods and reducing false alarms and missed alarms; then, it inputs the real-time load data into a temperature prediction model to obtain a predicted temperature sequence corresponding to the real-time load data, generating a desired temperature sequence that matches the current operating conditions; finally, it inputs the real-time temperature sequence, the predicted temperature sequence, and the temperature threshold into a false alarm discrimination model to obtain the result of whether the target switchgear needs to generate an alarm, which can effectively distinguish between normal temperature rise caused by changes in operating conditions and true abnormal thermal runaway, reduce invalid alarms and avoid missed alarms; at the same time, the result can also be used as a monitoring signal for closed-loop optimization for subsequent model and threshold parameter correction.

[0101] Figure 2 A flowchart illustrating the alarm method for switchgear provided in this application embodiment. Figure 2 ;like Figure 2 As shown, in this embodiment... Figure 1 Based on the implementation examples, this paper details how to use a false alarm discrimination model to determine whether a target switchgear needs to generate an alarm. The method includes:

[0102] S201. Calculate the residual value between the real-time temperature sequence and the predicted temperature sequence using the false alarm discrimination model.

[0103] Understandably, this method quantifies the deviation between the actual temperature and the predicted temperature derived from the predictive temperature model, thus distinguishing between temperature differences caused by normal load variations and those caused by potential equipment anomalies. A larger residual value indicates a more significant deviation between the actual temperature and the predicted value, suggesting a more likely real fault; a smaller residual value indicates that the temperature change is consistent with load characteristics, suggesting a more likely false alarm. This effectively reduces the false alarm rate caused by fixed thresholds under complex load conditions, significantly improving the accuracy and reliability of alarms.

[0104] S202. When the false alarm discrimination model determines that the residual value meets the temperature threshold, the output should show the result that the target switchgear needs to generate an alarm.

[0105] It should be noted that after the target switchgear generates an alarm, a field retest process will be triggered. This process involves collecting retest data (current load and temperature sequence) to correct or update the model parameters of the temperature prediction model, ensuring its accuracy. Specifically:

[0106] First, acquire the retest load data within a preset time window, as well as the process load data before receiving the alarm signal; then, form an extended dataset based on the retest load data and the process load data; finally, train a temperature prediction model based on the extended dataset to update the network parameters of the temperature prediction model.

[0107] For the alarm signal output by the discrimination model ( (At the alarm time), the actual current load sequence was collected on-site within a certain time window. and temperature sequence :

[0108]

[0109] In the formula, Extend the time window. For alarm occurrence time index, This is the index for the retest time.

[0110] Understandably, by collecting actual current load and temperature data on-site when an alarm is triggered, a retest dataset is formed, providing reliable input for subsequent model correction. This method accurately obtains the actual operating conditions at the time of the alarm, distinguishes between false alarms and real faults, and provides a data foundation for closed-loop model optimization, thereby improving the accuracy and reliability of temperature prediction and false alarm identification models.

[0111] A temperature prediction model based on LSTM is constructed, and its input-output relationship is expressed as follows:

[0112]

[0113] In the formula, Represents the mapping function of the LSTM network. For the network weight parameter set, The input feature vector contains the load sequence and temperature sequence.

[0114] The intermediate process load sequence can be represented as:

[0115]

[0116] In the formula, Index of alarm occurrence time For retest time index; This is the intermediate process load sequence.

[0117] Based on retest data and intermediate process data Create an extended dataset The temperature prediction model is retrained, and the gradient descent algorithm is used to correct the weight parameters of the LSTM network during training. The goal is to minimize the mean squared error loss function, i.e.:

[0118]

[0119] In the formula, For learning rate, To expand the number of samples in the dataset.

[0120] Understandably, by constructing an expanded dataset using intermediate process data before alarm triggering and on-site retest data, and then incrementally updating and adjusting the weights of the LSTM temperature prediction model, the model can quickly adapt to changes in the latest operating conditions and the relationship between load and temperature. This method significantly improves the prediction accuracy and generalization ability of the temperature prediction model under complex operating conditions, and reduces prediction errors caused by changes in model parameters or data distribution.

[0121] It should also be noted that after the temperature prediction model is updated, the output predicted temperature sequence will change, causing the residual distribution and amplitude characteristics between the predicted value and the real-time temperature to change accordingly. If the old false alarm discrimination model parameters are still used, the threshold and discrimination rules may not match the new residual characteristics, thereby reducing the discrimination accuracy. By synchronously updating the parameters of the false alarm discrimination model, its threshold and judgment logic can dynamically adapt to the latest prediction accuracy and residual pattern, ensuring that the false alarm discrimination maintains a low false alarm rate and a low false negative rate even after the model update.

[0122] In one possible approach, firstly, real-time load data is input into an updated temperature prediction model to obtain an updated predicted temperature sequence corresponding to the real-time load data; then, based on the updated predicted temperature sequence and the determination result that the alarm signal is not a false alarm signal, the parameters of the false alarm discrimination model are optimized.

[0123] It should be understood that the revised temperature prediction model is used to generate an updated predicted temperature sequence based on the new load sequence. Optimize the parameters of the false alarm discrimination model to minimize the error of the discrimination result in the retest confirmation:

[0124]

[0125] In the formula, To optimize step size, The loss function (e.g., cross-entropy or mean squared error) is used to measure the difference between the discriminant output and the true label. The true label determined for retesting ( when (otherwise it is 0). This is the temperature deviation threshold.

[0126] For example, at a certain moment An alarm was triggered, and the discrimination model output y(100)=1; the time window for retest data collection. If the value is 2, then the data collected on-site is:

[0127]

[0128] However, due to the large prediction error of the previous alarm discrimination model in this interval, false alarms occurred; therefore, the data collected on-site was compared with the load sequence before the alarm occurred. Merge into an expanded training set And retrain the LSTM network parameters using an expanded dataset. This allows the temperature prediction model to better fit the actual temperature curve; furthermore, the modified temperature prediction model is used to generate new predicted temperatures. Finally, the parameters of the false alarm discrimination model are determined based on the results of whether an alarm is needed, using the retest data. Make fine adjustments to reduce the probability of false alarms.

[0129] Understandably, by using the modified temperature prediction model to generate a new predicted temperature sequence and combining it with the alarm results obtained from the retest, the parameters of the false alarm discrimination model are optimized, which effectively reduces the false alarm rate and the missed alarm rate, realizes the continuous adaptive optimization of the alarm discrimination model, and thus significantly improves the reliability and stability of the overall alarm system.

[0130] S203. When the false alarm discrimination model determines that the residual value does not meet the temperature threshold, the output result is that the target switchgear does not need to generate an alarm.

[0131] It is understandable that when the false alarm discrimination model determines that the residual value does not meet the temperature threshold, it means that the status of each device in the target switch cabinet is normal and no equipment maintenance is required.

[0132] Figure 3 This is a schematic diagram of the structure of the alarm device for the switchgear provided in the embodiments of this application; as shown. Figure 3 As shown, the device includes:

[0133] The acquisition module 301 is used to acquire real-time load data of the target switchgear;

[0134] The determination module 302 is used to determine the real-time temperature sequence and temperature threshold based on the real-time load data;

[0135] The first processing module 303 is used to input real-time load data into the temperature prediction model to obtain a predicted temperature sequence corresponding to the real-time load data.

[0136] The second processing module 304 is used to input the real-time temperature sequence, the predicted temperature sequence, and the temperature threshold into the false alarm discrimination model to obtain the result of whether the target switch cabinet needs to generate an alarm.

[0137] In one possible implementation, the determining module 302 is specifically used for:

[0138] Analyze real-time load data to obtain the load average, load fluctuation degree, instantaneous change rate, and historical trend change rate corresponding to the real-time load data;

[0139] The real-time temperature sequence is determined based on the average load, the degree of load fluctuation, the instantaneous rate of change, and the historical trend rate of change.

[0140] The temperature threshold is determined based on the real-time temperature sequence.

[0141] In one possible implementation, the false alarm detection device for the switchgear further includes a training module, specifically used for:

[0142] Obtain the set of operating states of the sample switchgear, which includes fault-state operating data and normal-state operating data of the sample switchgear;

[0143] A dataset mapping the load to the contact temperature of sample switchgear was constructed based on the set of operating states.

[0144] The temperature prediction model is trained based on the mapping dataset; the temperature prediction model is built based on a long short-term memory network.

[0145] In one possible implementation, the second processing module 304 is specifically used for:

[0146] The residual value between the real-time temperature sequence and the predicted temperature sequence is calculated using a false alarm discrimination model;

[0147] When the false alarm discrimination model determines that the residual value meets the temperature threshold, it outputs the result that the target switchgear needs to generate an alarm.

[0148] When the false alarm discrimination model determines that the residual value does not meet the temperature threshold, it outputs a result indicating that the target switchgear does not need to generate an alarm.

[0149] In one possible implementation, after the target switchgear needs to generate an alarm result, the second processing module 304 is further configured to:

[0150] Acquire retest load data within a preset time window, as well as process load data before receiving an alarm signal;

[0151] An extended dataset is formed based on the re-measured load data and the process load data;

[0152] The temperature prediction model was trained using an expanded dataset to update its network parameters.

[0153] In one possible implementation, after training the temperature prediction model based on the extended dataset, the second processing module 304 is further configured to:

[0154] The real-time load data is input into the updated temperature prediction model to obtain the updated predicted temperature sequence corresponding to the real-time load data.

[0155] Based on the updated predicted temperature sequence and the judgment results that the alarm signal is not a false alarm signal, the parameters of the false alarm discrimination model are optimized.

[0156] The alarm device for the switchgear provided in this application embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0157] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4 As shown, the electronic device 40 provided in this embodiment includes at least one processor 401 and a memory 402. Optionally, the device 40 further includes a communication component 403. The processor 401, memory 402, and communication component 403 are connected via a bus 404.

[0158] In a specific implementation, at least one processor 401 executes computer execution instructions stored in memory 402, causing at least one processor 401 to perform the above-described method.

[0159] The specific implementation process of processor 401 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0160] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0161] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0162] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0163] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0164] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0165] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0166] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0167] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0168] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0169] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0170] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0171] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0172] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. An alarm method for a switchgear, characterized in that, include: Obtain real-time load data for the target switchgear; The real-time temperature sequence and temperature threshold are determined based on the real-time load data. The real-time load data is input into the temperature prediction model to obtain the predicted temperature sequence corresponding to the real-time load data; The real-time temperature sequence, the predicted temperature sequence, and the temperature threshold are input into the false alarm discrimination model to obtain the result of whether the target switchgear needs to generate an alarm.

2. The method according to claim 1, characterized in that, The step of determining the real-time temperature sequence and temperature threshold based on the real-time load data includes: The real-time load data is analyzed to obtain the load average, load fluctuation degree, instantaneous change rate, and historical trend change rate corresponding to the real-time load data; The real-time temperature sequence is determined based on the average load, the degree of load fluctuation, the instantaneous rate of change, and the historical trend rate of change. The temperature threshold is determined based on the real-time temperature sequence.

3. The method according to claim 1, characterized in that, The training process of the temperature prediction model includes: Obtain a set of operating states for the sample switchgear, the set of operating states including fault state operating data and normal state operating data of the sample switchgear; A dataset mapping the load to the contact temperature of the sample switchgear is constructed based on the set of operating states. The temperature prediction model is trained based on the mapping dataset; the temperature prediction model is constructed based on a long short-term memory network.

4. The method according to claim 3, characterized in that, The step of inputting the real-time temperature sequence, the predicted temperature sequence, and the temperature threshold into the false alarm discrimination model to obtain a result on whether the target switchgear needs to generate an alarm includes: The residual value between the real-time temperature sequence and the predicted temperature sequence is calculated using the false alarm discrimination model. When the false alarm discrimination model determines that the residual value meets the temperature threshold, it outputs the result that the target switchgear needs to generate an alarm. When the false alarm discrimination model determines that the residual value does not meet the temperature threshold, it outputs the result that the target switchgear does not need to generate an alarm.

5. The method according to claim 4, characterized in that, After outputting the result that the target switchgear needs to generate an alarm, the method further includes: Acquire the retest load data within a preset time window, as well as the process load data before receiving the alarm signal; An extended dataset is formed based on the retested load data and the process load data; The temperature prediction model is trained based on the extended dataset to update the network parameters of the temperature prediction model.

6. The method according to claim 5, characterized in that, After training the temperature prediction model based on the extended dataset, the method further includes: The real-time load data is input into the updated temperature prediction model to obtain an updated predicted temperature sequence corresponding to the real-time load data. Based on the updated predicted temperature sequence and the determination result that the alarm signal is not a false alarm signal, the parameters of the false alarm determination model are optimized.

7. A false alarm detection device for switchgear, characterized in that, include: The acquisition module is used to acquire the real-time load data of the target switchgear; The determination module is used to determine the real-time temperature sequence and temperature threshold based on the real-time load data; The first processing module is used to input the real-time load data into the temperature prediction model to obtain a predicted temperature sequence corresponding to the real-time load data. The second processing module is used to input the real-time temperature sequence, the predicted temperature sequence, and the temperature threshold into the false alarm discrimination model to obtain the discrimination result of whether the alarm signal is a false alarm signal.

8. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.

10. A computer program, characterized in that, The computer program includes a computer program stored in a computer-readable storage medium, which at least one processor can read from the computer-readable storage medium, and which, when executing the computer program, can implement the method described in any one of claims 1 to 6.

Images