Solid-state relay for self-recovery fuse and operation method

Abstract

The invention relates to the technical field of fuses, in particular to a solid-state relay for self-recovery insurance and an operation method. The device is characterized in that a controlled end of a solid-state relay optocoupler switch is connected in series between a high-voltage power supply and a load, and the device also comprises a sampling circuit which is used for collecting input current of the controlled end of the solid-state relay optocoupler switch and converting the input current of the controlled end into sampling voltage; the operational amplification circuit is adaptively connected with the sampling circuit and is used for performing operational amplification on the sampling voltage and outputting a control signal; and the controlled switch is located between the control end of the solid-state relay optocoupler switch and a power supply VCC, and the controlled switch is adaptively connected with the operational amplification circuit and used for receiving the control signal. The solid-state relay does not need to be replaced and is high in automation degree.

Description

Technical Field

[0001] This invention relates to the field of self-resetting relay operation technology, and more specifically, to a solid-state relay for self-resetting fuses and its operation method. Background Technology

[0002] A traditional fuse, also known as a circuit breaker, is an overcurrent protection device. Its core function is to melt and break the circuit when an overload or short-circuit current exceeds a specified limit, thereby protecting the electronic equipment in the circuit. The melting of a traditional fuse is irreversible; once protection is lost, it needs to be replaced, making the process cumbersome and lacking in automation. Summary of the Invention

[0003] In order to overcome the above-mentioned defects of the prior art, the present invention provides a solid-state relay for self-resetting fuse and an operating method thereof, which solves the problems mentioned in the background art through the following solution.

[0004] To achieve the above objectives, the present invention provides the following technical solution: The solid-state relay for self-resetting fuses of the present invention is characterized in that the controlled terminal of the solid-state relay optocoupler switch is connected in series between the high-voltage power supply and the load, and further includes: The sampling circuit is used to acquire the input current at the controlled end of the solid-state relay optocoupler switch and convert the input current at the controlled end into a sampling voltage. An operational amplifier circuit, adapted and connected to the sampling circuit, is used to perform operational amplification on the sampled voltage and output a sampled signal; The controlled switch is located between the control terminal of the solid-state relay optocoupler switch and the power supply VCC. The controlled switch is adapted to the operational amplifier circuit and is used to receive control signals.

[0005] This also includes an MCU, which receives the sampled signal and generates the control signal.

[0006] The above-mentioned operating method for the solid-state relay used for self-resetting fuses includes the following steps: S1: The MCU obtains the first operating data in the target load circuit through real-time sampling signals and historical sampling signals. The first operating data includes at least the instantaneous current value, current change rate, voltage waveform, and temperature parameters. S2: Based on the first operating data, analyze the instantaneous overcurrent events during the circuit startup phase or operation, and identify and obtain the first operating characteristics, including the type of inrush current; S3: Input the first running feature into a preset intelligent hierarchical delay model to calculate the dynamic delay threshold corresponding to the impact current type and generate a hierarchical delay strategy. S4: During the duration of the instantaneous overcurrent event, based on the graded delay strategy, determine and enter the corresponding multi-level delay protection state; S5: Based on the multi-level delay protection state of the target load circuit, execute the corresponding progressive protection action, and record the second operating characteristic during and after the execution of the progressive protection action; S6: Based on the statistical analysis results of the second operating characteristic, continuously optimize the graded delay strategy corresponding to the impact current type, and adaptively update the intelligent graded delay model.

[0007] Preferably, the inrush current type in S2 includes at least motor starting inrush, capacitor charging inrush, and resistive load inrush.

[0008] Preferably, in step S2, the identification of the type of impact current is achieved by constructing a decision tree classifier, which contains at least three levels: The first level is based on the rate of change of current. and peak current Perform the first classification; The second level combines voltage sag characteristics. Temperature change rate and heat accumulation Perform a second classification; The third level uses waveform similarity matching. The second classification was confirmed, based on the safety tolerance time. A confidence level assessment was performed on the confirmation results; When the confidence level of the third level is lower than the preset threshold, the auxiliary verification layer is activated: secondary identification is performed using frequency domain features or time series features.

[0009] Preferably, in S3, the intelligent hierarchical delay model includes a feature mapping layer, a dynamic delay calculator, a risk level evaluator, and a strategy generator; wherein, The feature mapping layer is configured to map the first running feature and convert it into a standardized feature vector; The dynamic delay calculator is configured to calculate the dynamic delay threshold based on the inrush current type and the feature vector; The risk level assessor is configured to calculate the instantaneous risk level based on the risk assessment matrix; The strategy generator is configured to generate a tiered delay strategy by combining the dynamic delay threshold and the instantaneous risk level.

[0010] Preferably, step S3, in the dynamic delay calculator, calculates the dynamic delay threshold, specifically including: Determine the base delay threshold based on the type of inrush current. and maximum and minimum delay boundaries , ; Multiple correction factors are constructed to dynamically adjust the basic delay threshold. The correction factors include at least a peak current correction factor, a rate of change current correction factor, a voltage sag correction factor, a heat accumulation correction factor, a safety time correction factor, and a temperature change correction factor. Combining the various correction factors with the aforementioned base delay threshold Calculate the dynamic delay threshold Specifically, it is expressed as: in, Let each correction factor be represented by its weight coefficient, and satisfy the following conditions: The summation range includes all correction factors. It is represented as a correction factor.

[0011] Preferably, S3, determining the basic delay threshold, specifically includes: Establish a mapping relationship between the inrush current type and the basic delay threshold: motor starting inrush corresponds to 100-500ms, capacitor charging inrush corresponds to 10-50ms, and resistive load inrush corresponds to 5-20ms. The basic delay threshold is adjusted based on the load power level: the upper limit of the range is used for high-power loads, and the lower limit of the range is used for low-power loads.

[0012] Preferably, in step S4, the multi-level delay protection state includes at least: First protection state: When an instantaneous overcurrent is detected, short-delay monitoring is initiated, the control signal is low, the controlled switch is closed, and the load remains in normal operation during this period; Second protection state: When continuous overcurrent is detected, mid-delay monitoring is activated, the control signal is low, the controlled switch is closed, and the output power limiting strategy is executed synchronously. Third protection state: When a serious fault is determined, the control signal is high, the controlled switch is disconnected, and the load is de-energized.

[0013] Preferably, in S5, the second operating feature includes the type of inrush current, the duration of the inrush current, the cumulative time of the current exceeding the dynamic delay threshold, and the inrush frequency.

[0014] The technical effects and advantages of this invention are as follows: 1. Because the solid-state relay used in this invention for self-resetting fuses disconnects the controlled switch and de-energizes the control terminal of the solid-state relay upon confirming an overload, the controlled terminal of the solid-state relay disconnects, thus achieving circuit protection. After the overload disappears, when the sampled voltage is lower than a preset threshold, the controlled switch closes, the control terminal of the solid-state relay is energized, and the controlled terminal of the solid-state relay conducts, thus achieving self-resetting. Therefore, the overcurrent protection using this solid-state relay for self-resetting fuses is reversible and can achieve self-resetting, thus eliminating the need for multiple fuse replacements, simplifying operation, and providing a high degree of intelligence.

[0015] 2. This invention effectively distinguishes between benign instantaneous inrush current and continuous overload or short-circuit fault current during the load startup phase through multi-feature analysis and impact type identification, thus solving the defect of existing technologies that are difficult to distinguish impact types. 3. This invention avoids the false triggering problem caused by the single threshold judgment in the prior art by constructing multiple correction factors to dynamically adjust the delay threshold, thereby reducing the probability of false action; 4. By defining protection states, this invention enables differentiated handling of overcurrent events of varying severity, avoiding the shortcomings of existing technologies that frequently trip and interrupt the normal startup and operation of the load, thus improving the reliability of circuit protection. Attached Figure Description

[0016] Figure 1 This is a structural topology diagram of the solid-state relay for self-resetting fuses according to the present invention; Figure 2 This is a flowchart illustrating the steps of a solid-state relay operation method for a self-resetting fuse provided in an embodiment of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0019] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0020] like Figure 1 As shown, the solid-state relay for self-resetting fuses of the present invention is characterized in that the controlled terminal of the solid-state relay optocoupler switch is connected in series between the high-voltage power supply and the load, and also includes a sampling circuit, an operational amplifier circuit, and an MCU. The sampling circuit is used to acquire the input current at the controlled terminal of the solid-state relay optocoupler switch and convert the input current into a sampling voltage. The operational amplifier circuit is adapted and connected to the sampling circuit to perform operational amplification on the sampling voltage and output a sampling signal. In this embodiment, the high-voltage power supply is 220V AC. The sampling circuit obtains the current between the high-voltage power supply and the load through an AC shunt. The specific circuit structures of the sampling circuit and the operational amplifier circuit are existing technologies and will not be described in detail here.

[0021] like Figure 1 As shown, there is a controlled switch between the control terminal of the solid-state relay optocoupler switch and the power supply VCC. The MCU receives the sampling signal and forms the control signal, which is used to control the controlled switch to close or open.

[0022] The MCU has three protection states. The first protection state is when a transient overcurrent is detected, short-delay monitoring is initiated, the control signal is low, the controlled switch is closed, and the load is kept running normally during this period.

[0023] Second protection state: When continuous overcurrent is detected, mid-delay monitoring is activated, the control signal is low, the controlled switch is closed, and the output power limiting strategy is executed synchronously.

[0024] Third protection state: When a serious fault is determined, the control signal is high, the controlled switch is disconnected, and the load is de-energized.

[0025] The controlled switch can sample semiconductor devices such as transistors and NMOS. The specific structure is existing technology and will not be described in detail here.

[0026] As attached Figure 2 The method for operating a solid-state relay for a self-resetting fuse, as shown, identifies the type of inrush current during transient overcurrents, dynamically calculates a graded delay strategy, and determines and executes multi-level progressive protection actions. Specifically, it includes the following steps: S1: The MCU obtains the first operating data in the target load circuit through real-time sampling signals and historical sampling signals. The first operating data includes at least the instantaneous current value, current change rate, voltage waveform, and temperature parameters. S2: Based on the first operating data, analyze the instantaneous overcurrent events during the circuit startup phase or operation, and identify and obtain the first operating characteristics, including the type of inrush current; S3: Input the first running feature into a preset intelligent hierarchical delay model to calculate the dynamic delay threshold corresponding to the impact current type and generate a hierarchical delay strategy. S4: During the duration of the instantaneous overcurrent event, based on the graded delay strategy, determine and enter the corresponding multi-level delay protection state; S5: Based on the multi-level delay protection state of the target load circuit, execute the corresponding progressive protection action, and record the second operating characteristic during and after the execution of the progressive protection action; S6: Based on the statistical analysis results of the second operating characteristic, continuously optimize the graded delay strategy corresponding to the impact current type, and adaptively update the intelligent graded delay model.

[0027] Specifically, in S1, when acquiring the first running data, a precise timestamp mechanism needs to be established and the alternating trigger mode of multiple ADC channels needs to be configured to ensure that the precise acquisition time is marked for each sampled data point, thereby achieving time consistency among various parameters.

[0028] It should be noted that the timestamp mechanism includes: using a high-precision hardware timer of the microcontroller as the time reference, with the timer clock source configured as a master clock of no less than 84MHz; marking each sampled data point with a 32-bit precision timestamp, with a timestamp precision of no less than 1 microsecond; the alternating trigger mode triggers synchronous sampling of the ADC through the timer to ensure synchronous acquisition of current and voltage parameters; establishing a sampling data time alignment table, and using a hardware DMA controller to automatically associate the sampled data with the timestamp; and storing all sampled data in real time to a circular buffer, the size of which is set according to the maximum expected fault duration to ensure that no data is lost.

[0029] In one embodiment, the instantaneous value of the current With the rate of change of current The acquisition process includes: using a closed-loop Hall effect current sensor as the core sensing element, connected in series in the target load circuit for real-time detection of the circuit current; and using an analog-to-digital converter (ADC) to digitally sample the sensor output signal, with an ADC resolution of at least 16 bits and a sampling rate of at least 1 MSPS for the instantaneous current value of the target load circuit. High-frequency sampling is performed; the rate of change of current between adjacent sampling points is calculated in real time using the first-order difference method. in, and These are respectively represented as sampling points. With sampling points Instantaneous current value , This represents the sampling time interval of the current, and its value corresponds to the sampling rate. When the sampling rate is 1 MSPS, .

[0030] In another embodiment, the acquisition of the voltage waveform includes: using a high-precision resistor divider network as the core element of the voltage sensing, which is directly connected in parallel across the two ends of the target load circuit for real-time monitoring of the operating voltage; the high-precision resistor divider network consists of two resistors connected in series: a high-voltage arm resistor... and low voltage arm resistor The formula for calculating its partial pressure ratio is: in, This represents the operating voltage of the target load circuit. This is represented as the voltage signal output to the ADC channel; the resistors used for voltage division are metal film resistors with an accuracy of 0.1% and a temperature coefficient of 25ppm / °C to ensure long-term stability and temperature adaptability. The power rating of the resistors is selected based on the maximum operating voltage and current to reduce the impact of temperature drift; thus, the voltage waveform and its instantaneous value are obtained.

[0031] In another embodiment, the acquisition of the temperature parameters includes: arranging multiple digital temperature sensors on the heatsink surface of the core power semiconductor device of the solid-state relay, the number of sensors being determined according to the heatsink size and thermal field distribution characteristics; this embodiment configures 3 to 5 sensing points; the digital temperature sensors have a measurement range covering -40°C to +125°C, an accuracy of not less than ±0.5°C, and are evenly distributed in the key hot areas of the heatsink: including the central area of ​​the power device mounting surface, the central area of ​​the heatsink fins, and the edge area of ​​the heatsink; the temperature data sampling rate is set to not less than 10Hz, triggered by a timer of the microcontroller.

[0032] Specifically, in S2, the instantaneous overcurrent event during the circuit startup phase or operation is analyzed, and the first operating characteristic obtained includes at least: the rate of change of current. Peak current Voltage sag characteristics Temperature change rate Heat accumulation Safety withstand time and type of inrush current.

[0033] Furthermore, based on the instantaneous value of the current For the rate of change of the current Establish a sliding window to analyze the rate of change characteristics within the window, including: the average rate of change and the rate of change for all currents within the window. The values ​​are taken as an arithmetic mean to reflect the overall trend of current change; peak rate of change, rate of change of current within the sliding window. The maximum absolute value is used to identify sudden and drastic changes; the sliding window is updated in a first-in-first-out manner, and an overlapping window strategy is adopted to avoid window boundary effects. In this embodiment, the step size is 10% of the window size to ensure the continuity of feature calculation.

[0034] It should be noted that the size of the sliding window is determined based on the number of sampling points; in one embodiment, the window size corresponds to 100 sampling points at a sampling rate of 1 kSPS; and the step size of the sliding window is 10% of the window size, and the step size is 10 points when the window size is 100 points; the overlapping window strategy avoids boundary effects by combining the last 90% of the data from the previous window with the newly added 10% of the data.

[0035] Furthermore, the voltage waveform within each sampling window is used to monitor the RMS voltage value of transient overcurrent events. Rate of change over time: ; Calculate the depth of voltage sag : in, The normal voltage value is defined as the average effective voltage value over a preset time period. This represents the minimum voltage during the instantaneous overcurrent period.

[0036] Furthermore, the temperature change rate at each sensing point is calculated in real time based on the temperature data from S1. The calculation formula is: in, This represents the temperature value at the current sampling time. This represents the temperature value at the previous sampling time. The sampling time interval for temperature data; the rate of temperature change at each sensing point. A weighted average is performed, with the weights determined based on the thermal distance between the sensing point and the power device. Defined as the physical distance between the sensing point and the heat source of the power device, its weighting formula is: in, This is represented as the attenuation coefficient in this embodiment. thermal distance The closer the device, the higher its weight; a heat accumulation model is established based on thermodynamic principles to calculate the heat accumulation of power devices in real time. : in, Expressed as the instantaneous value of the current, It is expressed as the on-resistance of the power device, and its value is corrected in real time as temperature changes. The sampling time interval for the current is expressed as the summation range from the start of the overcurrent event to the current moment; this predicts the safe withstand time of the power device under the current operating conditions. : in, This represents the maximum permissible junction temperature of the power device as specified in the device datasheet. This is represented by the currently measured temperature value. This is expressed as the rate of temperature change calculated in real time.

[0037] In one possible implementation, the inrush current type includes at least motor starting inrush, capacitor charging inrush, and resistive load inrush; its identification logic includes: constructing a decision tree classifier based on the first operating characteristic, the decision tree classifier containing at least three levels: the first level is based on the current change rate. and peak current The first classification is performed; the second level combines voltage sag characteristics. Temperature change rate and heat accumulation A second classification is performed; the third level uses waveform similarity matching. The second classification was confirmed, based on the safety tolerance time. The confidence level of the confirmation results is evaluated; when the confidence level of the third level is lower than the preset threshold, the auxiliary verification layer is activated: secondary identification is performed using frequency domain features or time series features.

[0038] It should be noted that the threshold of the decision tree classifier is dynamically adjusted based on historical recognition results, and the threshold is updated using an exponentially weighted moving average algorithm, with the learning rate set to 0.1.

[0039] In this embodiment, the threshold setting rule for the first classification is as follows: for motor starting impact, 50A / ms and Regarding capacitor charging surges, 200A / ms and For resistive load impacts, 20A / ms and , Represented as rated current; the threshold setting rule for the second classification: for motor starting impact, , 0.5℃ / s Regarding capacitor charging surges, , ℃ / s、 For resistive load impacts, , ℃ / s、 The waveform similarity matching uses the Dynamic Time Gauge (DTW) algorithm to calculate the similarity between the real-time current waveform and the pre-stored template waveform, and sets the similarity threshold to 0.8; if The confidence level increases by 0.2; if The confidence level decreases by 0.2. When the confidence level of the third level is lower than the preset threshold of 0.7, the frequency domain feature or time series feature is activated for secondary identification. In the secondary identification, the frequency domain feature uses the harmonic content analyzed by FFT. The motor starting impact usually contains a high second harmonic. The time series feature uses the impact duration, that is, the time from when the current exceeds the threshold to when it falls back. The motor starting impact usually lasts 100-500 milliseconds, and the capacitor charging impact lasts 10-50 milliseconds.

[0040] Specifically, in S3, an intelligent hierarchical delay model is composed of a feature mapping layer, a dynamic delay calculator, a risk level evaluator, and a strategy generator. The feature mapping layer is configured to map the first operational feature into a standardized feature vector. The dynamic delay calculator is configured to calculate a dynamic delay threshold based on the impact current type and the feature vector. The risk level evaluator is configured to calculate an instantaneous risk level based on a risk assessment matrix. The strategy generator is configured to generate a hierarchical delay strategy by combining the dynamic delay threshold and the instantaneous risk level.

[0041] It should be noted that the intelligent graded delay model adopts a hybrid model based on rules and statistical learning. The rule part is based on a predefined impact type feature library, and the statistical learning part uses a neural network to optimize the delay threshold. Its initial state is obtained by training with historical fault data. By inputting the feature vector of the first operating feature, a graded delay strategy including instantaneous risk level, dynamic delay threshold and maximum allowable current is obtained.

[0042] In one possible implementation, the dynamic delay calculator calculates the dynamic delay threshold by: determining a base delay threshold based on the inrush current type. and maximum and minimum delay boundaries , Multiple correction factors are constructed to dynamically adjust the base delay threshold. These correction factors include at least a peak current correction factor, a rate of change current correction factor, a voltage sag correction factor, a heat accumulation correction factor, a safety time correction factor, and a temperature change correction factor. The results are then combined with the base delay threshold. Calculate the dynamic delay threshold Specifically, it is expressed as: in, Let each correction factor be represented by its weight coefficient, and satisfy the following conditions: The summation range includes all correction factors. It is represented as a correction factor, and its range is mapped to [-1, 1].

[0043] Furthermore, the weighting coefficients of each correction factor are dynamically configured based on the type of impact current: For motor starting shock, assign higher weights to the safety time correction factor and the heat accumulation correction factor; For capacitor charging shock, the current change rate correction factor and the safety time correction factor are given high weights. For resistive load impacts, the temperature change correction factor and the heat accumulation correction factor are given higher weights.

[0044] In one embodiment, the weighting coefficients are configured based on the characteristics of the inrush current type: motor starting inrush: , , , , , Capacitor charging shock: , , , , , resistive load impact: , , , , ; after calculation Then, boundary constraint processing is performed: ,in, and Determined based on impact type: Motor starting impact Capacitor charging shock resistive load impact .

[0045] It should be noted that the specific values ​​of each weight coefficient were obtained through calibration using a large amount of experimental data, and adaptive optimization was performed based on historical protection effects during load operation. The optimization cycle was set to every 100 protection events.

[0046] Furthermore, determining the basic delay threshold includes: establishing a mapping relationship between the inrush current type and the basic delay threshold: 100-500ms for motor starting impact, 10-50ms for capacitor charging impact, and 5-20ms for resistive load impact; adjusting the basic delay threshold based on the load power level: using the upper limit of the range for high-power loads and the lower limit of the range for low-power loads; and adjusting the basic delay threshold based on the load aging degree: using a more conservative basic delay setting for aging loads.

[0047] It should be noted that each correction factor is calculated based on the first operating characteristic obtained from S2 and plays a different role in the delay decision: peak current correction factor. The key is to adjust the delay based on the overcurrent amplitude: when the actual peak current... When the current is significantly higher than the typical value for the corresponding inrush current type, it indicates a possible abnormal situation, and the delay should be appropriately shortened; when it is close to the typical value, the reference delay should be maintained to ensure normal startup; current change rate correction factor. The key lies in identifying the severity of current changes: the higher the rate of current change, the closer it is to the characteristics of a short-circuit fault. Negative correction significantly shortens the delay, enabling rapid response to severe faults; voltage sag correction factor. Reflecting voltage stability: a larger voltage sag indicates a more severe disturbance; negative correction shortens the delay, ensuring timely protection when stability is compromised; thermal accumulation correction factor. The key is to assess the long-term thermal stress of the load: the higher the heat accumulation, the greater the risk of overheating for the load. Negative corrections are used to gradually shorten the delay, preventing damage to the load due to thermal overload; the safety time correction factor... The key is to provide protection based on the urgency of the load: a shorter safety tolerance time indicates that the load is approaching a critical state. Negative corrections significantly shorten the delay, ensuring that the protection action is completed before the safety time expires; temperature change correction factor. The purpose is to monitor the real-time temperature rise of the load: the higher the temperature change rate, the more the load is in a rapid heating process. By negative correction, the delay can be shortened in time to deal with sudden thermal crises.

[0048] In one embodiment, the current peak correction factor The influence range is limited by using the hyperbolic tangent function: ,in, This is expressed as the peak sensitivity coefficient, with a value range of [0.1, 0.3], and in this embodiment, it is set to 0.2; The values ​​are represented as the baseline coefficients for impact types: 4.5 for motor starting impact, 6.0 for capacitor charging impact, and 1.3 for resistive load impact. Represented as rated current, when When, take 1, when When the current is -1, tanh represents the hyperbolic tangent function, which can automatically limit the output to the (-1,1) interval. When the current is extremely large, its effect of shortening the delay will tend to saturate rather than increase indefinitely.

[0049] In another embodiment, the current change rate correction factor An exponential decay function is used: ,in, It is expressed as the rate of change sensitivity coefficient, with a value range of [0.2, 0.5], and is set to 0.3 in this embodiment; The reference values ​​are expressed as the rate of change: 50 A / ms for motor starting impact, 200 A / ms for capacitor charging impact, and 20 A / ms for resistive load impact; when When -1 is taken; when When the value is small, the effect is approximately linear; when... Much larger hour, It rapidly approaches -1.

[0050] In another embodiment, the voltage sag correction factor Threshold-triggered: If ,but ;if ,but ;if ,but ;in, This is represented as a slight descent threshold, which is set to 5% in this embodiment. This is represented as the severe sag threshold, which is 15% in this embodiment.

[0051] In another embodiment, the thermal accumulation correction factor The S-shaped growth curve was used to simulate the heat accumulation effect: ,in The value is the thermal accumulation sensitivity coefficient, ranging from [0.3, 0.6], and is set to 0.4 in this embodiment; As a reference standard for heat accumulation, based on the heat capacity of power devices, the starting impact of the motor is taken as 2000J, the charging impact of the capacitor is taken as 200J, and the resistive load impact is taken as 800J; when When -1 is taken; in The effect is weak when it is small, near The time effect is significantly enhanced, in Far exceeding The time affects saturation.

[0052] In another embodiment, the safety time correction factor Using the reciprocal function: , The safety time sensitivity coefficient has a value range of [0.4, 0.8], and in this embodiment, it is set to 0.6; As a safety time reference, it is set to 5 seconds; when When -1 is taken, Take 0 at that time.

[0053] In another embodiment, the temperature change correction factor Using an exponential function: ,in, This is the temperature change sensitivity coefficient, with a value range of... In this embodiment, the value is 0.3; As a reference standard for temperature change rate, the starting impact of the motor is taken as 0.3℃ / s, the capacitor charging impact as 0.1℃ / s, and the resistive load impact as 0.5℃ / s; when Take -1.

[0054] Furthermore, the risk assessment matrix construction method of the risk level assessor specifically includes: establishing a three-dimensional risk assessment matrix based on current peak level, current change rate level, and temperature change rate level; the matrix elements of the risk assessment matrix are risk scores, and the risk scores are calculated using a linear weighted model.

[0055] Specifically, in S4, the determination and entry into the corresponding multi-level delay protection state is based on a finite state machine, which defines at least the "normal operation state", i.e. the first protection state, the "power limiting state", i.e. the second protection state, and the "delayed shutdown state", i.e. the third protection state. The transition conditions between states combine the instantaneous current amplitude, the duration of the amplitude, and the current change trend.

[0056] In one possible implementation, the multi-level delay protection states include at least the following: First protection state: when a transient overcurrent is detected, short-delay monitoring is initiated, the control signal is low, the controlled switch is closed, and the load remains in normal operation during this period; Second protection state: when a continuous overcurrent is detected, medium-delay monitoring is initiated, the control signal is low, the controlled switch is closed, and an output power limiting strategy is executed synchronously; Third protection state: when a serious fault is detected, the control signal is high, the controlled switch is open, and the load is de-energized.

[0057] It should be noted that the first protection state is when the load is operating under rated parameters and there is no overcurrent event or the overcurrent event has ended; the second protection state is when the load is in a continuous overcurrent but non-fault state, and the output power is actively limited to extend the operating time; the third protection state is when the load is in a serious fault state and a shutdown operation is performed.

[0058] In one embodiment, based on the dynamic delay threshold of the S4 output, a state transition rule is set: transitioning from the normal operating state to the power-limited state: when the instantaneous current > And triggered when the duration is greater than T1, or when the rate of change of current shows a continuous upward trend; transitioning from power-limited state to delayed shutdown state: when the instantaneous current is greater than T1. And it is triggered when the duration is >T2, or the rate of change of current exceeds the emergency threshold; Reverse migration condition: The condition for returning from the power-limited state to the normal operation state is that the instantaneous current is < And the duration is >200ms; recovery from the third protection state requires waiting for a configurable recovery time after the shutdown action, and then automatically attempts to reset. If the current is detected to have returned to normal and there are no fault characteristics, it will migrate back to the first protection state. All are taken from the output of S4.

[0059] It should be noted that the state transition process is equipped with an interlocking mechanism: before transitioning from the first state to the second state, the validity of the current sampling data must be verified; before transitioning from the second state to the third state, it must be confirmed that the power limiting strategy has been correctly executed and is ineffective; an event log is generated for each state transition for post-event analysis and diagnosis.

[0060] Specifically, in S5, the second operating feature is organized in the form of a structure, which includes: the type of inrush current, the duration of the inrush current, the cumulative time of the current exceeding the dynamic delay threshold, and the frequency of the inrush. At the same time, the data is stored in a circular buffer, the size of which can store the most recent 100 events. When the buffer is full, the oldest data is overwritten.

[0061] It should be noted that the progressive protection action includes graded current limiting: when entering the second protection state, the output current is limited to below [a certain value]. However, the current level is higher than the rated current, specifically 1.5 to 2 times the rated current in this embodiment, which is achieved through PWM technology and the output power is controlled by adjusting the duty cycle. In the second protection state, a configurable observation delay is initiated. During this period, if the current drops to a safe range, it automatically returns to the normal state. If the current continues to exceed the limit, it is upgraded to the third protection state. The timing of the observation delay is independent of the state transition timer, and the accuracy is guaranteed by the hardware timer.

[0062] Specifically, in S6, based on the statistical analysis results, the parameters of the intelligent graded delay model in S3 are dynamically adjusted. In this embodiment, if the duration of the impact current shows that 95% of the cases end within 100 milliseconds, the T1 delay threshold of this type of impact can be optimized to 100 milliseconds, and the adjustment range is limited by the safety boundary.

[0063] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A solid state relay for self-restoring insurance, characterized by, The controlled end of the solid-state relay optocoupler switch is connected in series between a high-voltage power supply and a load, and further comprises: a sampling circuit configured to collect input current at the controlled end of the solid-state relay optocoupler switch and convert the input current into a sampling voltage; an operational amplifier circuit connected to the sampling circuit and configured to perform operational amplification on the sampling voltage and output a sampling signal; a controlled switch located between the control end of the solid-state relay optocoupler switch and a power supply VCC, and connected to the operational amplifier circuit and configured to receive a control signal.

2. A solid state relay for self-restoring insurance as defined in claim 1, characterized in that An MCU is further included, which receives the sampling signal and forms the control signal.

3. The method for operating a solid state relay for self-restoring insurance according to claim 2, characterized in that The method comprises the following steps: S1: obtaining first operation data in a target load circuit by the MCU based on real-time sampling signals and historical sampling signals, wherein the first operation data at least includes current instantaneous value, current change rate, voltage waveform, and temperature parameter; S2: analyzing transient overcurrent events in a start-up stage or operation of the circuit based on the first operation data, identifying and obtaining first operation characteristics including impact current type; S3: inputting the first operation characteristics into a preset intelligent hierarchical delay model to calculate a dynamic delay threshold corresponding to the impact current type and generate a hierarchical delay strategy; S4: determining and entering a corresponding multi-level delay protection state based on the hierarchical delay strategy during the duration of the transient overcurrent event; S5: performing corresponding progressive protection actions according to the multi-level delay protection state of the target load circuit, and recording second operation characteristics during and after the execution of the progressive protection actions; S6: continuously optimizing the hierarchical delay strategy corresponding to the impact current type based on the statistical analysis result of the second operation characteristics, and adaptively updating the intelligent hierarchical delay model.

4. A method for operating a solid state relay for self-restoring insurance according to claim 3, characterized in that: The S2, the impact current type at least includes motor start-up impact, capacitor charging impact, and resistive load impact.

5. A method for operating a solid state relay for self-restoring insurance according to claim 3, characterized in that: The S2, the identification of the impact current type is performed by constructing a decision tree classifier, and the decision tree classifier comprises at least three levels: The first level is based on the rate of change of current and the current peak A first classification is made; Second level combined voltage sag features , temperature change rate and heat accumulation amount performing second classification; The third level employs waveform similarity matching Confirming the second classification and basing the safety tolerance time Performing a confidence assessment on the confirmation result; When the confidence of the third level is lower than a preset threshold, an auxiliary verification layer is started: secondary identification is performed using frequency domain features or time sequence features.

6. A method for operating a solid state relay for self-restoring insurance according to claim 5, characterized in that: The S3, the intelligent hierarchical delay model comprises a feature mapping layer, a dynamic delay calculator, a risk level evaluator, and a strategy generator; wherein, the feature mapping layer is configured to map the first operation characteristics to a standardized feature vector; the dynamic delay calculator is configured to calculate a dynamic delay threshold based on the impact current type and the feature vector; the risk level evaluator is configured to calculate a transient risk level based on a risk evaluation matrix; the strategy generator is configured to generate a hierarchical delay strategy by integrating the dynamic delay threshold and the transient risk level.

7. A method for operating a solid state relay for self-restoring insurance according to claim 6, characterized in that: The S3, in the dynamic delay calculator, the dynamic delay threshold is calculated, specifically comprising: determining a base delay threshold value depending on the type of the inrush current and maximum, minimum delay boundaries , ; A plurality of correction factors are constructed to dynamically adjust the basic delay threshold, and the correction factors at least include a current peak correction factor, a current rate of change correction factor, a voltage sag correction factor, a thermal accumulation correction factor, a safety time correction factor, and a temperature change correction factor; combining the various correction factors with the base latency threshold , to calculate a dynamic latency threshold , expressed as: wherein represents the weight coefficient corresponding to each correction factor, and satisfies , the summation range includes all the correction factors, represents the correction factor.

8. A method for operating a solid state relay for self-restoring insurance according to claim 7, characterized in that: The S3, determination of the basic delay threshold, specifically comprises: A mapping relationship between the impact current type and the basic delay threshold is established: 100-500 ms for motor starting impact, 10-50 ms for capacitor charging impact, and 5-20 ms for resistive load impact; The basic delay threshold is adjusted based on the load power level: the upper limit of the range is used for high-power loads, and the lower limit of the range is used for low-power loads.

9. A method for operating a solid state relay for self-restoring insurance according to claim 3, characterized in that: The S4, the multi-stage delay protection state at least includes: The first protection state: when it is determined to be instantaneous overcurrent, short delay monitoring is started, the control signal is low, the controlled switch is closed, and the load is kept normal during the period; The second protection state: when it is determined to be continuous overcurrent, medium delay monitoring is started, the control signal is low, the controlled switch is closed, and an output power limiting strategy is synchronously executed; The third protection state: when it is determined to be a serious fault, the control signal is high, the controlled switch is opened, and the load is powered off.

10. A method for operating a solid state relay for self-restoring insurance according to claim 3, characterized in that: The S5, the second operating feature includes an impact current type, an impact current duration, a cumulative time of a current super-dynamic delay threshold, and an impact frequency.

Images