A high-sensitivity power distribution cabinet power failure protection device and method

Abstract

The present invention discloses a high-sensitivity power-off protection device and method for power distribution cabinets, and relates to the technical field of power distribution. The device comprises: a monitoring parameter acquisition module: acquiring data units and sensitivity evaluation coefficients of high-sensitivity monitoring equipment of power distribution cabinets; a rule configuration module: configuring load environment adaptive feedback rules according to acquisition data units, sensitivity evaluation coefficients and load environment characteristics; an identification relationship construction module: establishing a decision identification relationship between monitoring parameters and power-off protection identification monitoring channels based on adaptive feedback rules; a data import module: connecting monitoring equipment, importing real-time cabinet monitoring data into identification monitoring channels, and judging power-off protection requirements; a protection instruction generation module: matching power-off protection equipment according to power-off protection requirements, and generating corresponding protection instructions. Thus, the technical effect of improving protection reliability and environmental adaptability and ensuring stable power distribution is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution, and in particular to a high-sensitivity power-off protection device and method for a power distribution cabinet. Background Art

[0002] Power-off protection is one of the important functions of the power distribution cabinet, which can cut off the power supply in time when a fault or abnormality occurs to avoid equipment damage and personal injury. However, with the increasing complexity of the power distribution system and the diversification of environmental conditions, the power-off protection setting of the power distribution cabinet power-off protection device faces greater challenges.

[0003] In existing power-off protection technologies, the sensitivity of the power-off protection device is usually set based on a single electrical parameter (such as current, power) to ensure a quick response when a fault occurs. In actual applications, when faced with instantaneous current changes, the power-off protection device may malfunction, resulting in frequent power outages. There are technical problems such as high false-touch frequency, low environmental adaptability, and impact on power distribution stability. Summary of the invention

[0004] The present invention provides a high-sensitivity power-off protection device and method for a power distribution cabinet, so as to solve the technical problems in the prior art of high false touch frequency, low environmental adaptability and influence on power distribution stability, and achieve the technical effect of improving protection reliability and environmental adaptability and ensuring power distribution stability.

[0005] In a first aspect, the present invention provides a high-sensitivity power-off protection device for a power distribution cabinet, wherein the device comprises:

[0006] The monitoring parameter acquisition module is used to collect data from the high-sensitivity monitoring equipment of the power distribution cabinet and obtain the acquisition data unit and sensitivity evaluation coefficient of the monitoring equipment.

[0007] A rule configuration module is used to configure load environment adaptive feedback rules according to the collected data unit and its sensitive evaluation coefficient of the monitoring device and in combination with load environment characteristics.

[0008] An identification relationship building module is used to establish a decision identification relationship between the monitoring parameters of the monitoring equipment and the adaptive feedback rules, and to build a power-off protection identification monitoring channel according to the decision identification relationship. The power-off protection identification monitoring channel performs protection mechanism identification processing.

[0009] A data import module is used to connect to the high-sensitivity monitoring device, obtain real-time electrical cabinet monitoring data and import it into the identification monitoring channel to determine the power-off protection requirements.

[0010] A protection instruction generation module is used to match the power-off protection processing device according to the power-off protection requirement and generate a power-off protection instruction.

[0011] In a second aspect, the present invention further provides a high-sensitivity power distribution cabinet power failure protection method, wherein the method comprises:

[0012] A high-sensitivity monitoring device for collecting data in the power distribution cabinet is used to obtain the collected data unit and its sensitivity evaluation coefficient of the monitoring device; according to the collected data unit and its sensitivity evaluation coefficient of the monitoring device and in combination with the load environment characteristics, a load environment adaptive feedback rule is configured; a decision identification relationship between the monitoring parameters of the monitoring device and the adaptive feedback rule is established, and a power-off protection identification monitoring channel is constructed according to the decision identification relationship, and the power-off protection identification monitoring channel performs protection mechanism identification processing; the high-sensitivity monitoring device is connected to obtain real-time cabinet monitoring data and import it into the identification monitoring channel to determine the power-off protection demand; the power-off protection processing device is matched according to the power-off protection demand to generate a power-off protection instruction.

[0013] The present invention discloses a high-sensitivity power-off protection device and method for a power distribution cabinet, comprising: a monitoring parameter acquisition module acquires the acquisition data of a high-sensitivity monitoring device deployed in a power distribution cabinet, and obtains the acquisition data unit of each monitoring device and its corresponding sensitivity evaluation coefficient; a rule configuration module sets and configures an adaptive feedback rule applicable to the load environment according to the acquisition data unit and the sensitivity evaluation coefficient of the monitoring device, combined with the characteristics of the load environment; an identification relationship construction module establishes a decision-making identification relationship between the monitoring parameters of the monitoring device and the adaptive feedback rule, and based on this, constructs a power-off protection identification monitoring channel, and performs related protection mechanism identification processing; a data import module connects to a high-sensitivity monitoring device, imports the real-time monitoring data of the power distribution cabinet into the identification monitoring channel, analyzes and determines whether there is a power-off protection demand; a protection instruction generation module matches the corresponding power-off protection processing device according to the identified power-off protection demand, generates and issues a power-off protection instruction, and ensures that the safety protection strategy of the power distribution cabinet is effectively executed. A high-sensitivity power distribution cabinet power-off protection device and method disclosed by the present invention solves the technical problems of high false touch frequency, low environmental adaptability, and influence on the stability of power distribution, and achieves the technical effect of improving protection reliability and environmental adaptability and ensuring the stability of power distribution. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a structural schematic diagram of a high-sensitivity power-off protection device for a power distribution cabinet according to the present invention;

[0015] Figure 2 The present invention is a flowchart of a high-sensitivity power distribution cabinet power failure protection method.

[0016] Explanation of the accompanying drawings: monitoring parameter acquisition module 11, rule configuration module 12, identification relationship construction module 13, data import module 14, protection instruction generation module 15. DETAILED DESCRIPTION

[0017] The technical solution provided in the embodiments of the present invention is to solve the technical problems of high false touch frequency, low environmental adaptability and influence on power distribution stability in the prior art. The overall idea adopted is as follows:

[0018] First, the monitoring parameter acquisition module is used to collect the high-sensitivity monitoring device data of the distribution cabinet, and obtain the acquisition data unit and the sensitivity evaluation coefficient of the monitoring device; then, the rule configuration module configures the load environment adaptive feedback rule according to the acquisition data unit and the sensitivity evaluation coefficient of the monitoring device, combined with the load environment characteristics; then, the identification relationship construction module is used to establish the decision identification relationship between the monitoring parameters of the monitoring device and the adaptive feedback rule, and construct a power-off protection identification monitoring channel according to the decision identification relationship, which is used for protection mechanism identification processing; then, the data import module connects the high-sensitivity monitoring device, obtains the real-time cabinet monitoring data and imports it into the identification monitoring channel to determine the power-off protection demand; then, the protection instruction generation module matches the power-off protection processing device according to the power-off protection demand and generates a power-off protection instruction.

[0019] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings, rather than all of them.

[0020] Embodiment 1

[0021] Figure 1 The present invention is a schematic diagram of the structure of a high-sensitivity power distribution cabinet power failure protection device, wherein the device comprises:

[0022] The monitoring parameter acquisition module 11 is used to collect data from the high-sensitivity monitoring equipment of the power distribution cabinet and obtain the acquisition data unit and the sensitivity evaluation coefficient of the monitoring equipment.

[0023] Specifically, high-sensitivity monitoring equipment refers to equipment that can accurately and real-time collect various data in the power distribution cabinet (such as current, voltage, temperature, humidity, etc.) and quickly respond to abnormal conditions. Common high-sensitivity monitoring equipment includes voltage sensors, current transformers, temperature sensors, wireless sensor network nodes, etc.

[0024] Specifically, the data collection unit refers to the standardized measurement value unit used in the monitoring process, such as ampere (A) for current collection, volt (V) for voltage collection, Celsius (℃) for temperature collection, ppm (parts per million) for gas concentration collection, etc. The data collection unit should be divided according to the performance of the monitoring equipment and the different types and characteristics of the monitored objects. In other words, the data collection unit of the monitoring equipment is different in different application environments.

[0025] Specifically, the sensitivity evaluation coefficient is a coefficient used to quantify the monitoring device's ability to respond to specific data changes, and is used to describe the monitoring device's sensitivity to small data fluctuations (such as current / voltage fluctuations, temperature surges). By defining and obtaining the sensitivity evaluation coefficient, it is helpful to dynamically adjust the power-off protection strategy so that it can adapt to the precise sampling requirements in different scenarios. Optionally, the definition of the evaluation coefficient includes the ratio of the data change amount to the response speed of the monitoring device, the ratio of the monitoring error to the sensitivity threshold, etc.

[0026] In some embodiments, the acquisition of the data collection unit and the sensitivity evaluation coefficient of the monitoring device comprises the following steps:

[0027] Obtain the input change and output change of the monitoring device.

[0028] A minimum monitoring amount is obtained according to the input variation and the output variation, and the minimum monitoring amount is used as the data collection unit.

[0029] The sensitivity evaluation coefficient is calculated based on the ratio of the output change to the input change.

[0030] Specifically, first, obtain the input change and output change of the monitoring device, where the input change of the monitoring device refers to the change value of the external physical quantity (such as temperature, pressure, current, vibration, etc.) received by the device within a specific time period, such as a voltage increase of 5V, a current increase of 2A, etc. The output change is the signal change value output by the monitoring device after the input signal is collected and processed or converted by the sensor within the corresponding time period. For example, if the voltage sensor converts a 1V input voltage change into a corresponding 4mV output electrical signal change, the output change is 4mV.

[0031] Specifically, the minimum monitoring amount refers to the smallest signal change that the monitoring device can distinguish. This parameter represents the sensitivity or resolution of the monitoring device. For example, if the minimum temperature change that a temperature sensor can detect is 0.01°C, then its minimum monitoring amount is 0.01°C. Optionally, by obtaining the monitoring device's ability to resolve output changes and combining the proportional relationship between output changes and input changes, the minimum monitoring amount is determined as the data collection unit, and then the sensitivity level of the monitoring device is quantified, thereby providing an accurate benchmark unit for subsequent data collection.

[0032] Specifically, the sensitivity evaluation coefficient refers to the ratio of the change in the output signal of the monitoring device to the change in the input signal, which is used to quantify the sensitivity of the device. The calculation of the sensitivity evaluation coefficient can effectively measure the sensitivity of the monitoring device and provide a quantitative basis for subsequent abnormal analysis and power-off protection decision-making.

[0033] The rule configuration module 12 is used to configure the load environment adaptive feedback rule according to the acquisition data unit and the sensitivity evaluation coefficient of the monitoring device and the load environment characteristics.

[0034] In some embodiments, according to the data collection unit and the sensitivity evaluation coefficient of the monitoring device, combined with the load environment characteristics, the load environment adaptive feedback rule is configured, and the execution steps include:

[0035] Based on historical sample data, the relationship between load characteristics, environmental characteristics, collected data units and their sensitive evaluation coefficients and power-off identification of abnormal protection is established.

[0036] A power-off protection mechanism compensation analysis is performed based on the power-off identification relationship of the abnormal protection to determine a compensation feedback strategy.

[0037] The compensation feedback strategy is added to the power-off protection mechanism, and a compensation mapping relationship among the load characteristics, environmental characteristics, acquisition data units and their sensitive evaluation coefficients is established to obtain the load environment adaptive feedback rule.

[0038] Specifically, feature engineering is performed based on historical data samples to extract load features such as current and voltage fluctuation characteristics, load change rate, start / stop, etc. Exemplarily, the feature extraction method includes frequency feature analysis through Fourier transform (FFT) and time-frequency feature extraction through short-time Fourier transform (STFT). Among them, the historical data sample set is formed by continuously collecting current, voltage, temperature and other data under different load and environmental conditions through monitoring equipment.

[0039] Specifically, environmental characteristics include external environmental data such as temperature, humidity, and air velocity recorded in historical data samples, and include fluctuation characteristics of collected data under different environmental conditions.

[0040] Specifically, based on mathematical models (such as regression analysis models) or machine learning models (such as decision trees, support vector machines, etc.), the relationship between load characteristics, environmental characteristics, collected data units and their sensitive evaluation coefficients and power outage identification of abnormal protection is established, which will help to subsequently optimize the accuracy and sensitivity of power outage identification.

[0041] Optionally, the historical data samples are divided into positive samples and negative samples according to the accuracy of the power-off protection operation, wherein the positive samples refer to the sample data in the historical data that successfully triggers the power-off protection and avoids the occurrence of faults, reflecting the normal response characteristics of the monitoring equipment under abnormal load or environmental conditions, including the abnormal fluctuation characteristics of current, voltage, and temperature. Negative samples refer to the sample data in the historical data that fails to correctly trigger the power-off protection or erroneously triggers the power-off protection, which is manifested as false alarms, missed alarms, etc., reflecting the erroneous response characteristics of the monitoring equipment under abnormal conditions. Then, feature extraction is performed on the feature data of the positive samples and negative samples (such as current fluctuations, voltage changes, temperature surges, etc.) to form a standardized sample vector. PCA (principal component analysis) is involved for dimensionality reduction, and time series feature extraction algorithms (such as autocorrelation analysis) are used to extract time series features. Then, all sample data are normalized (such as normalizing the current fluctuation amplitude to the [0,1] interval) to eliminate the magnitude differences between features. Through sample classification and labeling, a high-quality training data set is provided for subsequent model training, ensuring that the machine learning model can identify the correct power-off triggering conditions in different scenarios, thereby more accurately learning the mapping relationship between features and power-off protection conditions.

[0042] Specifically, the compensation analysis of the power-off protection mechanism is used to identify potential problems in the power-off protection strategy (such as false alarms and missed alarms), and to make compensatory improvements to the power-off strategy by optimizing parameters, adjusting thresholds, etc. This analysis is based on the power-off identification relationship of abnormal protection. Among them, the compensation feedback strategy is a set of strategies that dynamically adjusts and optimizes the problems in the power-off protection strategy (such as unreasonable thresholds, improper sensitivity coefficient settings, etc.) based on the results of the aforementioned compensation analysis. This strategy introduces an adaptive feedback mechanism to enable the power-off protection strategy to maintain the optimal state under different load and environmental conditions.

[0043] Exemplarily, the compensation feedback strategy is determined, including extracting the main characteristic dimensions of false alarm and missed alarm samples through methods such as principal component analysis (PCA) or cluster analysis (K-Means), identifying key factors that may cause false alarms or missed alarms (such as excessive sensitivity evaluation coefficients or unreasonable power-off trigger threshold settings), and then, based on the analysis results of false alarm and missed alarm characteristics, using a threshold dynamic adjustment algorithm (such as an adaptive threshold adjustment method, a fuzzy logic adjustment method, etc.) to make compensatory adjustments to the power-off protection trigger threshold. For example, if the false alarm rate is high under a certain environmental feature (such as high humidity), the coefficient of the trigger threshold (such as the current fluctuation amplitude threshold) can be appropriately increased to reduce false alarms caused by environmental factors.

[0044] Optionally, determining the compensation feedback strategy also includes dynamically adjusting the sensitivity coefficient of the monitoring device according to the false alarm and missed alarm situations in different load scenarios. For example, when the load fluctuates violently, the sensitivity coefficient can be reduced to avoid false alarms; and when the load is stable, the sensitivity coefficient can be increased to improve the accuracy of abnormality identification.

[0045] Specifically, the compensation feedback strategy includes input variables and output variables. Among them, the input variables include load characteristics (such as current fluctuation characteristics, load change rate), environmental characteristics (such as temperature, humidity, external electromagnetic interference, etc.), data acquisition units (such as minimum monitoring quantity) and their sensitivity evaluation coefficients. Output variables include power-off protection triggering thresholds (such as fluctuation amplitude thresholds of current, voltage, temperature, etc.), power-off response time (such as the delay time for triggering power-off protection), recovery strategies (such as recovery strategies after power-off protection), etc.

[0046] Specifically, the compensation mapping relationship refers to establishing a multi-dimensional feature mapping model by analyzing the nonlinear relationship between load characteristics, environmental characteristics, acquisition data units and their sensitive evaluation coefficients and power-off protection strategies. This mapping relationship is used to accurately determine when the power-off protection compensation strategy needs to be triggered and adjusted under different environmental conditions, thereby realizing an adaptive power-off protection strategy.

[0047] By introducing the "compensation feedback strategy" into the power-off protection mechanism and establishing a mapping relationship between load characteristics, environmental characteristics, collected data units and their sensitive evaluation coefficients, the power-off protection strategy can be automatically adjusted according to the dynamic changes of the actual operating environment, thereby improving the adaptability to complex working conditions and the accuracy and stability of power-off protection.

[0048] The identification relationship building module 13 is used to establish a decision identification relationship between the monitoring parameters of the monitoring equipment and the adaptive feedback rules, and to build a power failure protection identification monitoring channel according to the decision identification relationship, and the power failure protection identification monitoring channel performs protection mechanism identification processing.

[0049] In some embodiments, a decision recognition relationship between monitoring parameters of a monitoring device and an adaptive feedback rule is established, and a power failure protection recognition monitoring channel is constructed according to the decision recognition relationship. The execution steps include:

[0050] A decision-making identification relationship between load characteristics and feedback mechanism is established, and a primary feedback rule layer is constructed. The load characteristics include load fluctuation type, load fluctuation amplitude, and load trend characteristics.

[0051] A decision recognition relationship between environmental characteristics and feedback mechanisms is established, and a secondary feedback rule layer is constructed. The environmental characteristics include temperature / humidity characteristics.

[0052] The decision-making identification relationship between the collected data unit and its sensitive evaluation coefficient and the dynamic time window is established, and a three-level feedback rule layer is constructed.

[0053] Establish a decision-making recognition relationship between the dynamic time window and the feedback warning threshold, and construct the top feedback rule layer.

[0054] A feedback closed-loop mechanism is introduced to perform feature recognition connection from the primary feedback rule layer to the top feedback rule layer to obtain a decision recognition relationship of an adaptive feedback rule.

[0055] An identification decision tree is established according to the decision identification relationship, and the power failure protection identification monitoring channel is constructed based on the identification decision tree.

[0056] Specifically, based on the relationship between load characteristics and feedback mechanisms, by analyzing different load fluctuation types, fluctuation amplitudes and trend characteristics, a primary decision-making recognition relationship applicable to various load conditions is established, so as to make a preliminary judgment on abnormal load changes. For example, if the load fluctuation type is "periodic fluctuation" and the fluctuation amplitude is less than a certain threshold (such as 5%), the primary feedback rule outputs "normal". If the load fluctuation type is "sudden fluctuation" and the fluctuation amplitude exceeds a certain threshold (such as 20%), the primary feedback rule outputs "abnormal" and triggers the next level of rule layer.

[0057] The load fluctuation types include: long-term stable load, intermittent load, and high-fluctuation load. Different load types correspond to different response strategies. For example, stable load corresponds to low acquisition frequency and sensitivity. Intermittent load corresponds to increasing the acquisition frequency before the load starts. High-fluctuation load corresponds to high-frequency acquisition and sensitivity adjustment.

[0058] Among them, the load fluctuation amplitude reflects the magnitude of the load fluctuation, including clearly defined fluctuation amplitude classifications, such as small fluctuations, medium fluctuations, and large fluctuations. The feedback strategy under different fluctuation amplitudes should have clear acquisition frequency adjustment and threshold settings. The load trend characteristics describe the overall trend of load fluctuations (such as gradual increase, continuous decrease, rapid fluctuations), and are identified through trend analysis (such as linear fitting, moving average method, etc.) to trigger the protection mechanism in advance.

[0059] Specifically, the secondary feedback rule layer is based on the relationship between environmental characteristics (such as temperature, humidity, etc.) and the feedback mechanism. Environmental characteristics have a direct impact on the stability of the load and the sensitivity of the monitoring equipment. Therefore, environmental characteristic recognition rules are introduced on the basis of the primary rule layer to further improve the accuracy of abnormal recognition. For example, high humidity or high temperature environment may affect the sensitivity and stability of the monitoring equipment. The data collection unit and response strategy need to be adjusted according to the humidity characteristics to avoid misjudgment.

[0060] Specifically, the three-level feedback rule layer combines the acquisition data unit, sensitivity evaluation coefficient and dynamic time window to form a time series-based recognition model for dynamic recognition and feedback adjustment of complex time features (such as frequent fluctuations, slow trend changes, etc.). Exemplarily, a corresponding feedback mechanism is established for the acquisition units of different sensors (such as current, voltage, temperature, etc.). For example, when the sensitivity of the acquisition unit is low, the data acquisition frequency should be increased to avoid missing important load changes. When the sensitivity of the acquisition unit is high, the time window can be appropriately extended to reduce false alarms.

[0061] Optionally, the adjustment of the dynamic time window needs to be combined with the actual operating environment and adjusted in real time according to the environment and load. For example, when the current fluctuates frequently, the sensor's acquisition time window should be shortened and the sensitivity should be improved.

[0062] Specifically, the top feedback rule layer is the comprehensive result of the rules of each layer, which is used to judge and execute the final power-off protection strategy. By establishing the identification relationship between the dynamic time window and the feedback warning threshold, the global control and strategy optimization of the overall monitoring can be achieved. Optionally, the top feedback rule layer is constructed based on a decision tree model or a hierarchical analysis model. By comprehensively judging the identification results of the rule layers at each level and combining the changes in the dynamic time window, the final power-off protection warning threshold (such as the current fluctuation amplitude threshold, power-off trigger time, etc.) is adjusted to form the top feedback rule and output the global power-off protection strategy.

[0063] Furthermore, a feedback closed-loop mechanism is introduced to adaptively connect the judgment results of each level of rule layers to form an adaptive feedback rule decision-making recognition relationship. In other words, there are not only recognition results transmitted upward between each level of rule layers (bottom-up feedback transmission), but also downward reverse adjustments (top-down feedback corrections) are introduced to form a two-way closed-loop mechanism. For example, when the top rule layer determines that the overall environment is abnormal, the dynamic time window of the third-level rule layer can be adjusted in reverse, thereby realizing adaptive recognition strategy adjustment, ensuring that strategy adjustment and optimization can be performed based on real-time data under complex environmental conditions.

[0064] Furthermore, the feature recognition results of each level of rule layers are integrated into the node relationship of the overall decision tree model. By constructing a decision recognition relationship with a hierarchical structure, the complex feature determination process can be transformed into a structured decision path, thereby improving the efficiency and accuracy of decision-making.

[0065] In some implementations, a decision recognition relationship between a dynamic time window and a feedback warning threshold is established, and a top feedback rule layer is constructed. The execution steps include:

[0066] According to the dynamic time window, a first warning threshold and a second protection threshold are set, and the first warning threshold is smaller than the second protection threshold.

[0067] In the dynamic time window, when the first warning threshold is reached, a warning message is sent, and when the second protection threshold is reached, the power-off protection requirement is obtained.

[0068] Specifically, a dual threshold mechanism is adopted, and the warning threshold (first warning threshold) and the protection threshold (second protection threshold) are set respectively. Among them, the warning threshold is used to identify slight or early abnormal fluctuations. When it is detected that the load or environmental parameters reach the warning threshold, a warning message is sent (such as prompting that the current load fluctuation exceeds the normal range), but the power-off protection operation will not be triggered immediately. The protection threshold is used to deal with serious faults. When it is detected that the load or environmental parameters reach the protection threshold, it is determined that power-off protection measures need to be taken immediately, and the power-off operation is triggered.

[0069] Specifically, different dynamic time windows correspond to different first warning thresholds and second protection thresholds to monitor and respond to input features in real time at different time scales. By setting dynamic time windows of different lengths, short-term fluctuations and long-term trends can be effectively distinguished, thereby improving the accuracy of threshold determination.

[0070] The data import module 14 is used to connect to the high-sensitivity monitoring device, obtain real-time cabinet monitoring data and import it into the identification monitoring channel to determine the power-off protection demand.

[0071] Specifically, high-sensitivity monitoring equipment is used to monitor and collect real-time cabinet monitoring data, and the real-time cabinet monitoring data is used as input data. Power-off protection judgment is performed by identifying the monitoring channel to generate power-off protection requirements.

[0072] Optionally, the power-off protection requirement includes the object to be powered off and the location of the object, the power-off and reset requirement parameters, etc. For example, when it is determined that there is an irreversible or potentially destructive fault (such as the current fluctuation amplitude exceeds 30% and lasts for more than 2 seconds), the power-off protection requirement is to immediately trigger the power-off protection. When it is determined that the fault may be caused by sudden fluctuations (such as the current fluctuation amplitude exceeds 30%, but the duration is less than 1 second), the power-off protection requirement is to delay the power-off and continue to monitor the fluctuation trend to prevent the wrong power-off caused by instantaneous fluctuations.

[0073] The protection instruction generating module 15 is used to match the power-off protection processing device according to the power-off protection requirement and generate a power-off protection instruction.

[0074] Specifically, after the identification monitoring channel determines the power-off protection requirement, the power-off protection requirement is matched with the specific power-off protection processing equipment, and the corresponding power-off protection instruction is generated according to the matching result. Among them, the power-off protection processing equipment includes circuit breakers, switchgear, relay protection devices, etc. The generated power-off protection instruction is a program instruction that can be understood by the corresponding power-off protection processing equipment, which is used to perform specific power-off operations. By generating accurate power-off protection instructions, the response speed and execution accuracy of the power-off operation can be effectively improved, ensuring that the power supply can be quickly cut off in the event of a serious fault.

[0075] Furthermore, the execution steps of the device also include:

[0076] Set the test step length of the input variation and obtain the sensitivity evaluation coefficient of each test step length.

[0077] Clustering is performed according to the discreteness of the sensitivity evaluation coefficients to obtain sensitivity level partitions.

[0078] Monitoring compensation information is established according to the sensitivity evaluation coefficient of the sensitivity level partition, wherein when the sensitivity evaluation coefficient reaches a low sensitivity threshold, the monitoring compensation information is to increase the acquisition frequency.

[0079] Specifically, the range of the variation is set according to the input signal type of the device under test (such as current, voltage, temperature, etc.), and the test step length is determined. For example, different step lengths such as 0.1A, 0.2A, and 0.5A can be selected for step-by-step testing. Then, using the preset input variation step length, gradually increase the input signal, record the output variation of the device under each step length, and calculate the corresponding sensitivity evaluation coefficient.

[0080] Specifically, the sensitivity evaluation coefficients obtained under different test step lengths are usually discrete and volatile. By performing cluster analysis on the discreteness of these coefficients, different sensitivity level partitions can be divided. Exemplarily, data with similar sensitivity characteristics are divided into the same partition through cluster analysis, and high sensitivity areas, medium sensitivity areas, and low sensitivity areas are identified based on the partition results. Among them, the discreteness is used to describe the volatility between the sensitivity evaluation coefficients under each test step length, and can be calculated using indicators such as standard deviation and coefficient of variation.

[0081] Furthermore, the response capabilities of the monitoring equipment vary under different sensitivity partitions. Therefore, monitoring compensation information is established to automatically adjust the acquisition frequency or other monitoring parameters in low-sensitivity partitions to compensate for the problem of insufficient sensitivity. The establishment of compensation information is mainly based on the partition results of the sensitivity evaluation coefficient and the specific threshold setting. Exemplarily, for low-sensitivity partitions with low response capabilities to input signals, the lack of sensitivity is compensated by increasing the acquisition frequency or improving the sampling accuracy. For high-sensitivity partitions with strong response capabilities, no additional compensation is required. By establishing monitoring compensation information, frequency compensation or accuracy improvement can be automatically performed under low-sensitivity conditions to prevent monitoring blind spots caused by insufficient sensitivity.

[0082] In some implementations, clustering is performed according to the discreteness of the sensitivity evaluation coefficients to obtain sensitivity level partitions, and the execution steps include:

[0083] According to the dispersion of the sensitivity evaluation coefficient, the cluster number k value is determined by calculating the intra-cluster sum of squares under different k values, wherein the cluster number k value is the value of the number of decreases in the intra-cluster sum of squares after increasing the k value.

[0084] The sensitivity evaluation coefficients are clustered according to the cluster quantity k value to obtain k clusters, and the sensitivity levels are configured according to the size order of the sensitivity evaluation coefficients of the k clusters to obtain the sensitivity level partitions.

[0085] Specifically, first, multiple clustering number k values ​​are determined, and the intra-cluster sum of squares under different k values ​​are calculated, and the k value corresponding to the inflection point of the intra-cluster sum of squares (the point where the decrease in the intra-cluster sum of squares decreases after increasing the k value) is selected as the optimal clustering number. Then, the k-means clustering algorithm is used to cluster the sensitivity evaluation coefficients to obtain k clusters, and the k clusters correspond to k-level sensitivity level partitions, where the maximum and minimum values ​​of the sensitivity evaluation coefficient in each cluster are the upper and lower limits of the interval corresponding to the sensitivity level partition.

[0086] For example, the within-cluster sum of squares is calculated as follows: ;

[0087] Among them, WCCS represents the intra-cluster sum of squares;K is the number of clusters; C k It is k clusters; x i It is i data points. k It is k The center (centroid) of each cluster.

[0088] Through the above steps, we can determine the reasonable number of clusters, cluster the sensitivity evaluation coefficients, and configure the sensitivity level according to the clustering results.

[0089] In summary, the high-sensitivity power distribution cabinet power failure protection device provided by the present invention has the following technical effects:

[0090] The monitoring parameter acquisition module collects the data collected by the high-sensitivity monitoring equipment deployed in the power distribution cabinet, and obtains the data collection unit of each monitoring equipment and its corresponding sensitivity evaluation coefficient; the rule configuration module sets and configures the adaptive feedback rules suitable for the load environment based on the data collection unit and sensitivity evaluation coefficient of the monitoring equipment, combined with the characteristics of the load environment; the identification relationship construction module establishes the decision-making identification relationship between the monitoring parameters of the monitoring equipment and the adaptive feedback rules, and builds the power-off protection identification monitoring channel based on this, and performs the relevant protection mechanism identification processing; the data import module connects the high-sensitivity monitoring equipment, imports the real-time monitoring data of the power distribution cabinet into the identification monitoring channel, analyzes and determines whether there is a power-off protection demand; the protection instruction generation module matches the corresponding power-off protection processing equipment according to the identified power-off protection demand, generates and issues the power-off protection instruction, and ensures the effective execution of the safety protection strategy of the power distribution cabinet. Thus, the technical effect of improving protection reliability and environmental adaptability and ensuring stable power distribution is achieved.

[0091] Embodiment 2

[0092] Figure 2 This is a flow chart of a high-sensitivity power distribution cabinet power failure protection method of the present invention. For example, Figure 1 The structural schematic diagram of a high-sensitivity power distribution cabinet power failure protection device in the present invention can be used to achieve the following Figure 2 The process shown.

[0093] Based on the same concept as the high-sensitivity power-off protection device for a power distribution cabinet in the above embodiment, the present invention also provides a high-sensitivity power-off protection method for a power distribution cabinet, including:

[0094] Collect data from high-sensitivity monitoring equipment in the distribution cabinet and obtain the collection data unit and sensitivity evaluation coefficient of the monitoring equipment.

[0095] According to the collected data unit and the sensitive evaluation coefficient of the monitoring device and in combination with the load environment characteristics, a load environment adaptive feedback rule is configured.

[0096] A decision identification relationship between monitoring parameters of the monitoring equipment and adaptive feedback rules is established, and a power failure protection identification monitoring channel is constructed according to the decision identification relationship. The power failure protection identification monitoring channel performs protection mechanism identification processing.

[0097] Connect the high-sensitivity monitoring device, obtain real-time cabinet monitoring data and import it into the identification monitoring channel to determine the power-off protection requirements.

[0098] A power-off protection processing device is matched according to the power-off protection requirement to generate a power-off protection instruction.

[0099] In some embodiments, obtaining the acquisition data unit of the monitoring device and its sensitivity evaluation coefficient includes:

[0100] The input change and output change of the monitoring device are obtained; based on the input change and the output change, a minimum monitoring amount is obtained, and the minimum monitoring amount is used as the data acquisition unit; based on the ratio of the output change to the input change, the sensitivity evaluation coefficient is calculated.

[0101] In some embodiments, according to the data collection unit and the sensitivity evaluation coefficient of the monitoring device, combined with the load environment characteristics, the load environment adaptive feedback rule is configured, and the execution steps include:

[0102] According to the historical sample data, a power-off identification relationship between the load characteristics, environmental characteristics, collected data units and their sensitive evaluation coefficients and abnormal protection is established; according to the power-off identification relationship of the abnormal protection, a compensation analysis of the power-off protection mechanism is performed to determine the compensation feedback strategy; the compensation feedback strategy is added to the power-off protection mechanism, and a compensation mapping relationship between the load characteristics, environmental characteristics, collected data units and their sensitive evaluation coefficients is established to obtain the load environment adaptive feedback rule.

[0103] In some embodiments, a decision recognition relationship between monitoring parameters of a monitoring device and an adaptive feedback rule is established, and a power failure protection recognition monitoring channel is constructed according to the decision recognition relationship. The execution steps include:

[0104] Establish a decision-making identification relationship between load characteristics and feedback mechanisms, and construct a primary feedback rule layer, wherein the load characteristics include load fluctuation type, load fluctuation amplitude, and load trend characteristics; establish a decision-making identification relationship between environmental characteristics and feedback mechanisms, and construct a secondary feedback rule layer, wherein the environmental characteristics include temperature / humidity characteristics; establish a decision-making identification relationship between acquisition data units and their sensitive evaluation coefficients and dynamic time windows, and construct a tertiary feedback rule layer; establish a decision-making identification relationship between dynamic time windows and feedback warning thresholds, and construct a top feedback rule layer; introduce a feedback closed-loop mechanism to perform feature identification connection from the primary feedback rule layer to the top feedback rule layer, and obtain a decision-making identification relationship for adaptive feedback rules; establish an identification decision tree based on the decision identification relationship, and construct the power-off protection identification monitoring channel based on the identification decision tree.

[0105] In some implementations, a decision recognition relationship between a dynamic time window and a feedback warning threshold is established, and a top feedback rule layer is constructed. The execution steps include:

[0106] According to the dynamic time window, a first warning threshold and a second protection threshold are set, and the first warning threshold is less than the second protection threshold; in the dynamic time window, when the first warning threshold is reached, a warning message is sent, and when the second protection threshold is reached, the power off protection requirement is obtained.

[0107] Furthermore, the method further comprises:

[0108] The test step length of the input variation is set, and the sensitivity evaluation coefficients of each test step length are obtained respectively; clustering is performed according to the discreteness of the sensitivity evaluation coefficients to obtain sensitivity level partitions; and monitoring compensation information is established according to the sensitivity evaluation coefficients of the sensitivity level partitions, wherein when the sensitivity evaluation coefficient reaches a low sensitivity threshold, the monitoring compensation information is to increase the acquisition frequency.

[0109] In some implementations, clustering is performed according to the discreteness of the sensitivity evaluation coefficients to obtain sensitivity level partitions, and the execution steps include:

[0110] According to the discreteness of the sensitivity evaluation coefficient, the cluster number k value is determined by calculating the intra-cluster sum of squares under different k values, wherein the cluster number k value is a numerical value at which the decrease amplitude of the intra-cluster sum of squares decreases after the k value is increased; the sensitivity evaluation coefficient is clustered according to the cluster number k value to obtain k clusters, and the sensitivity levels are configured in the order of the size of the sensitivity evaluation coefficients of the k clusters to obtain the sensitivity level partitions.

[0111] It should be understood that the embodiments mentioned in this specification focus on their differences from other embodiments. The specific embodiments in the aforementioned embodiment one are also applicable to a high-sensitivity distribution cabinet power-off protection method described in embodiment two. For the sake of brevity of the specification, they will not be further elaborated here.

[0112] It should be understood that the embodiments disclosed in the present invention and the above description can enable those skilled in the art to use the present invention to implement the present invention. At the same time, the present invention is not limited to the above-mentioned embodiments. It should be understood that those skilled in the art can still modify the technical solutions recorded in the above-mentioned embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A high-sensitivity power-off protection device for a power distribution cabinet, characterized in that: include: A monitoring parameter acquisition module, which is used to acquire data from a high-sensitivity monitoring device of a power distribution cabinet, and obtain acquisition data units and sensitivity evaluation coefficients of the monitoring device; A rule configuration module, the rule configuration module is used to configure a load environment adaptive feedback rule according to the acquisition data unit and the sensitivity evaluation coefficient of the monitoring device and in combination with the load environment characteristics; An identification relationship building module, the identification relationship building module is used to establish a decision identification relationship between the monitoring parameters of the monitoring equipment and the adaptive feedback rule, and to build a power failure protection identification monitoring channel according to the decision identification relationship, and the power failure protection identification monitoring channel performs protection mechanism identification processing; A data import module, the data import module is used to connect to the high-sensitivity monitoring device, obtain real-time cabinet monitoring data and import it into the identification monitoring channel to determine the power-off protection demand; A protection instruction generation module, the protection instruction generation module is used to match the power-off protection processing device according to the power-off protection requirement and generate a power-off protection instruction; The steps of obtaining the data collection unit and the sensitivity evaluation coefficient of the monitoring equipment include: Obtain the input change and output change of the monitoring equipment; According to the input variation and the output variation, a minimum monitoring amount is obtained, and the minimum monitoring amount is used as the data collection unit; Calculating the sensitivity evaluation coefficient according to the ratio of the output change to the input change; Set the test step length of the input variation and obtain the sensitivity evaluation coefficient of each test step length; Clustering is performed according to the dispersion of the sensitivity evaluation coefficient to obtain sensitivity level partitions; Establishing monitoring compensation information according to the sensitivity evaluation coefficient of the sensitivity level partition, wherein when the sensitivity evaluation coefficient reaches a low sensitivity threshold, the monitoring compensation information is to increase the acquisition frequency; The steps of clustering according to the discreteness of the sensitivity evaluation coefficient to obtain the sensitivity level partition include: According to the dispersion of the sensitivity evaluation coefficient, by calculating the intra-cluster sum of squares under different k values, the cluster number k value is determined, wherein the cluster number k value is the number value at which the decrease of the intra-cluster sum of squares decreases after increasing the k value; The sensitivity evaluation coefficients are clustered according to the cluster quantity k value to obtain k clusters, and the sensitivity levels are configured according to the size order of the sensitivity evaluation coefficients of the k clusters to obtain the sensitivity level partitions.

2. The high-sensitivity power-off protection device for a power distribution cabinet as claimed in claim 1, characterized in that: According to the data collection unit and the sensitivity evaluation coefficient of the monitoring device, combined with the load environment characteristics, the load environment adaptive feedback rule is configured, and the execution steps include: Establish the power-off identification relationship between load characteristics, environmental characteristics, data collection units and their sensitivity evaluation coefficients and abnormal protection based on historical sample data; Performing a power-off protection mechanism compensation analysis based on the power-off identification relationship of the abnormal protection and determining a compensation feedback strategy; The compensation feedback strategy is added to the power-off protection mechanism, and a compensation mapping relationship among the load characteristics, environmental characteristics, acquisition data units and their sensitive evaluation coefficients is established to obtain the load environment adaptive feedback rule.

3. The high-sensitivity power-off protection device for a power distribution cabinet as claimed in claim 2, characterized in that: The steps of establishing a decision recognition relationship between monitoring parameters of the monitoring device and an adaptive feedback rule, and constructing a power failure protection recognition monitoring channel according to the decision recognition relationship include: Establish a decision-making identification relationship between load characteristics and feedback mechanisms, and construct a primary feedback rule layer. The load characteristics include load fluctuation type, load fluctuation amplitude, and load trend characteristics; Establishing a decision recognition relationship between environmental characteristics and feedback mechanisms, and constructing a secondary feedback rule layer, wherein the environmental characteristics include temperature / humidity characteristics; Establish the decision-making identification relationship between the data collection unit and its sensitive evaluation coefficient and the dynamic time window, and construct a three-level feedback rule layer; Establish the decision recognition relationship between the dynamic time window and the feedback warning threshold, and construct the top feedback rule layer; Introducing a feedback closed-loop mechanism to perform feature recognition connection from the primary feedback rule layer to the top feedback rule layer to obtain a decision recognition relationship of an adaptive feedback rule; An identification decision tree is established according to the decision identification relationship, and the power failure protection identification monitoring channel is constructed based on the identification decision tree.

4. The high-sensitivity power-off protection device for a power distribution cabinet as claimed in claim 3, characterized in that: Establish a decision recognition relationship between the dynamic time window and the feedback warning threshold, and build the top feedback rule layer. The execution steps include: According to the dynamic time window, a first warning threshold and a second protection threshold are set, wherein the first warning threshold is less than the second protection threshold; In the dynamic time window, when the first warning threshold is reached, a warning message is sent, and when the second protection threshold is reached, the power-off protection requirement is obtained.

5. A high-sensitivity power-off protection method for a power distribution cabinet, characterized in that: The method is applied to a high-sensitivity power-off protection device for a power distribution cabinet as claimed in any one of claims 1 to 4, and the method comprises: Collect data from the high-sensitivity monitoring equipment of the power distribution cabinet, and obtain the data collection unit and its sensitivity evaluation coefficient of the monitoring equipment; According to the data collection unit and the sensitivity evaluation coefficient of the monitoring device, combined with the load environment characteristics, configure the load environment adaptive feedback rule; Establishing a decision recognition relationship between monitoring parameters of the monitoring equipment and an adaptive feedback rule, constructing a power failure protection recognition monitoring channel according to the decision recognition relationship, and performing protection mechanism recognition processing on the power failure protection recognition monitoring channel; Connect the high-sensitivity monitoring device, obtain real-time cabinet monitoring data and import it into the identification monitoring channel to determine the power-off protection requirements; A power-off protection processing device is matched according to the power-off protection requirement to generate a power-off protection instruction.

Images