Program-controlled circuit breaking execution method and device triggered by heartbeat monitoring

Abstract

The present application relates to the technical field of monitoring control, and discloses a heartbeat monitoring triggered program control circuit execution method and device, wherein heartbeat signals of N test channels of a heartbeat circuit breaker box are received, heartbeat interval characteristic sequences of the channels are constructed, a dynamic timeout threshold is calculated, a channel disconnection confirmation flag is generated through multi-stage anti-jitter decision, three-level circuit breaker decisions of channel-level DC, system-level AC coordination and channel-level alarm are executed based on a state control matrix, a circuit execution sequence is generated, a channel to be restored is identified, a restoration priority queue is constructed, staggered electrical restoration is executed, and stable restoration and conflict avoidance are realized by combining gradual voltage soft start and communication time slot allocation; the present application takes the channel as the minimum control granularity, realizes accurate fault isolation and orderly restoration, avoids local channel disconnection triggering global power failure, and improves the continuity and reliability of a multi-station test system.

Description

Technical Field

[0001] This invention relates to the field of monitoring and control technology, and more specifically, to a method and apparatus for triggering program control circuit breaker execution in heartbeat monitoring. Background Technology

[0002] In multi-station testing scenarios such as aging and durability testing of on-board chargers, a multi-channel parallel testing mode is often adopted to improve testing efficiency. Each test channel needs to maintain stable communication and electrical connection to ensure the validity of test data and the continuity of the testing process. Among these, heartbeat monitoring is the core means of judging the communication status of the channel, while circuit breaker control is the key link to ensure system safety and prevent the spread of faults when a channel communication abnormality occurs. The synergistic effect of the two directly determines the operational stability and testing efficiency of the multi-station testing system.

[0003] In the prior art, Chinese patent CN119862059B discloses an intelligent fault handling method, device, computer-readable storage medium, and electronic device for distributed systems. This method generates heartbeat packets by starting an application service, sends these packets to a heartbeat queue at set time intervals, and after receiving and saving the heartbeat queue data, a monitoring service continuously collects data from each node for preprocessing and fault detection feature extraction. It uses an isolated forest model for training and an adaptive threshold adjustment method to analyze the heartbeat packet data. When an abnormal heartbeat is detected in a node, the fault location is determined based on the feature code, and a compensation message is generated for the application service to execute the compensation task. This method can adapt to complex distributed environments and improve system stability and reliability. Chinese patent application CN109480810A discloses a pulse signal error correction method based on a circular queue. This method initializes a timer and general-purpose input / output ports, enables interrupts to capture pulse signals, continuously measures pulse signals three times or more, and calculates the pulse rate by averaging the values. If the signal acquisition is abnormal, it re-acquires the signal. This method can eliminate abnormal signals caused by inaccurate sensor positioning and muscle tremors, ensuring the accuracy of pulse signal measurement.

[0004] However, the aforementioned existing technologies still have significant technical defects in the multi-station testing scenario of on-board charger aging and durability testing, and it is difficult to solve the coupling accidental damage problem of "partial disconnection - global current interruption". Specifically, while existing technology CN119862059B can achieve fault detection and task compensation in distributed systems, it focuses on fault recovery at the data task level and does not address the circuit-breaking execution logic of "hierarchical control of electrical circuits" in multi-station testing. Its compensation mechanism cannot adapt to the independent control requirements of DC and AC circuits required for on-board charger testing. If applied to multi-channel testing, when a single channel experiences a heartbeat abnormality (partial disconnection), it lacks precise circuit-breaking means at the channel level, which can easily trigger a system-level global electrical cutoff, causing other normal channel tests to be interrupted. Existing technology CN109480810A focuses on error correction and accurate measurement of a single pulse signal, only solving the accuracy problem of a single path signal. It lacks an independent data processing mechanism for multi-channel parallel monitoring and cannot distinguish the communication rhythm differences of different test channels. In multi-station scenarios, if the fixed measurement logic of this technology is used, it will either lead to misjudgment due to its inability to adapt to the communication fluctuation characteristics of each channel (treating instantaneous interference as a partial disconnection) or fail to detect real disconnection due to the lack of dynamic judgment criteria. Neither of these technologies involves electrical execution design for "partial fault isolation". In the aging and durability testing of on-board chargers, multi-station systems typically contain N (N≥2) independent test channels. The communication status of each channel varies due to factors such as electromagnetic interference and changes in sample load. Existing technologies suffer from coarse control granularity (e.g., global coupling control) and a lack of channel-specific judgment criteria. When a channel experiences a "partial disconnection" (e.g., a momentary communication interruption or a true loss of connection), either the DC power supply circuit cannot be cut off for that channel only, triggering a "global current interruption" of the total AC power supply, forcing the termination of long-cycle tests on other normal channels, resulting in invalid test data and extended test cycles; or the momentary interference is misjudged as a true loss of connection, leading to frequent unnecessary circuit breaking actions, further exacerbating the risk of test interruption, ultimately resulting in a coupled accidental damage of "partial disconnection - global current interruption," which severely restricts the resource utilization and test continuity of the multi-station test system. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of existing technologies, this invention provides a method and apparatus for program-controlled circuit breaker execution triggered by heartbeat monitoring. By constructing a channel-level heartbeat interval feature sequence and a dynamic timeout threshold, personalized judgment of communication status is achieved. Based on a state control matrix, a three-level circuit breaker decision is executed, performing partial power cuts only on faulty channels to avoid global power outages. Through a recovery priority queue and a staggered voltage soft-start strategy, the safe and orderly restart of faulty channels is ensured. This method significantly improves the tolerance of multi-station testing systems to local faults, preventing "partial disconnection" from inducing "global power outages," and effectively ensuring testing continuity and equipment safety.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The program control circuit breaker execution method triggered by heartbeat monitoring includes:

[0008] Receive heartbeat signals from N test channels of the heartbeat circuit breaker box and construct the heartbeat interval feature sequence of each channel. Calculate the dynamic timeout threshold of each channel based on the heartbeat interval feature sequence of each channel. Based on the dynamic timeout threshold of each channel, generate a channel disconnection confirmation flag through multi-level anti-jitter decision.

[0009] Initialize the status control matrix, update the status control matrix according to the channel disconnection confirmation flag to obtain the real-time status control matrix; based on the real-time status control matrix, execute three-level circuit breaker decisions and generate the circuit breaker execution sequence for each channel; the first level of the three-level circuit breaker decision is the channel-level DC circuit breaker decision, the second level is the system-level AC coordinated circuit breaker decision, and the third level is the channel-level independent alarm decision.

[0010] The system reads the real-time status control matrix to identify the channels to be restored, calculates the restoration priority value of each channel, and constructs a restoration priority queue. It then performs staggered electrical restoration according to the restoration priority queue, and achieves stable electrical restoration and communication conflict avoidance by interacting with the detection heartbeat frame through gradual voltage soft start and communication time slot allocation.

[0011] The method for constructing the heartbeat interval feature sequence of each channel includes:

[0012] The arrival time of the heartbeat signal for each channel is recorded. By calculating the time interval between adjacent heartbeat signals, the time interval between adjacent heartbeat signals is defined as the heartbeat interval, and a heartbeat interval feature sequence specific to each channel is constructed.

[0013] The method for calculating the dynamic timeout threshold of each channel based on the heartbeat interval feature sequence of each channel includes:

[0014] Based on the heartbeat interval characteristic sequence of each channel, calculate the average value μ of the heartbeat interval for each channel i. i and standard deviation σ i According to the average value μ i and standard deviation σ i Construct a stable interval for channel i; set an initial timeout baseline threshold for each channel;

[0015] Based on the stable range, the communication stability status of each channel is determined; according to the communication stability status of each channel, the initial timeout reference threshold of each channel is adaptively adjusted, and the dynamic timeout threshold of each channel is output.

[0016] The method for determining the communication stability status of each channel includes:

[0017] Real-time monitoring of new heartbeat interval values ​​for each channel. When k1 consecutive new heartbeat interval values ​​for channel i fall within the stable range, channel i is determined to be in a stable communication state. When k2 consecutive new heartbeat interval values ​​for channel i exceed the stable range, channel i is determined to be in a fluctuating communication state. Here, k1 is the number of consecutive stable determinations and k2 is the number of consecutive fluctuating determinations.

[0018] The method for adaptively adjusting the initial timeout reference threshold for each channel includes:

[0019] For channels in a stable communication state, the initial timeout reference threshold of the channel is tightened using a stable state adjustment coefficient α. For channels in a fluctuating communication state, the initial timeout reference threshold of the channel is relaxed using a fault-tolerant state adjustment coefficient β. The dynamic timeout threshold of each channel is then output after adaptive adjustment.

[0020] The method for generating a channel disconnection confirmation flag through multi-level anti-jitter decision includes:

[0021] The system monitors the reception status of heartbeat signals in each channel in real time, identifies channels where heartbeat signals are lost, and records the duration of signal loss in channels where heartbeat signals are lost.

[0022] When the signal loss duration of a channel with a lost heartbeat signal first exceeds the dynamic timeout threshold of the channel, the channel is placed in a "suspected disconnection" state and enters an observation period, the duration of which is set.

[0023] During the observation period, the heartbeat signal of the channel in the "suspected disconnection" state is continuously monitored; if no heartbeat signal is detected at the end of the observation period, the channel in the "suspected disconnection" state is placed in the "actual disconnection" state, and a channel disconnection confirmation flag is generated for the channel in the "actual disconnection" state.

[0024] The status control matrix is ​​N rows and 3 columns. The row index corresponds to N test channels, and the column index corresponds to three types of core control objects: DC output control, AC input control, and audible and visual alarm control. The three columns of the status control matrix are DC output control column, AC input control column, and audible and visual alarm control column, respectively.

[0025] When initializing the state control matrix, all elements in the DC output control column are set to 1, all elements in the AC input control column are set to 1 / N, and all elements in the audible and visual alarm control column are set to 1.

[0026] The execution method of the channel-level DC circuit breaker decision is as follows: monitor the changes in the elements of the DC output control column of the real-time state control matrix, and generate a channel-level DC circuit breaker action request for the corresponding channel when any element is detected to change from 1 to 0.

[0027] The execution method for the system-level collaborative circuit breaker decision is as follows:

[0028] Calculate the sum Sac of all elements in the AC input control column of the real-time state control matrix, compare Sac with the AC collaborative circuit breaker threshold γ, and generate a system-level AC circuit breaker action request when Sac is less than γ.

[0029] A heartbeat monitoring-triggered program control circuit breaker execution device, used to implement the above-described heartbeat monitoring-triggered program control circuit breaker execution method, the device comprising:

[0030] Disconnection confirmation module: used to receive heartbeat signals from N test channels of the heartbeat circuit breaker box and construct the heartbeat interval feature sequence of each channel. Based on the heartbeat interval feature sequence of each channel, it calculates the dynamic timeout threshold of each channel. Based on the dynamic timeout threshold of each channel, it generates a channel disconnection confirmation flag through multi-level anti-jitter decision.

[0031] Circuit breaker decision module: used to initialize the status control matrix, update the status control matrix according to the channel disconnection confirmation flag, and obtain the real-time status control matrix; based on the real-time status control matrix, execute three-level circuit breaker decisions and generate the circuit breaker execution sequence for each channel; the first level of the three-level circuit breaker decision is the channel-level DC circuit breaker decision, the second level is the system-level AC coordinated circuit breaker decision, and the third level is the channel-level independent alarm decision.

[0032] Stable recovery module: It is used to read the real-time status control matrix to identify the channels to be restored, calculate the recovery priority value of each channel to be restored, and build a recovery priority queue; it performs staggered electrical restoration according to the recovery priority queue, and achieves stable electrical restoration and communication conflict avoidance by interacting with the detection heartbeat frame through gradual voltage soft start and communication time slot allocation.

[0033] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0034] This invention utilizes personalized calculations of heartbeat interval characteristic sequences and dynamic timeout thresholds to adapt the disconnection judgment criteria for each channel to its communication rhythm, avoiding misjudgments or omissions caused by fixed thresholds. The state control matrix quantifies channel states into calculable control weights, enabling independent channel-level control and system-level collaborative protection for DC output, AC input, and alarm control. This ensures that a single channel fault only triggers a local circuit breaker, without affecting the power supply to other channels. The three-level circuit breaker decision-making uses the channel as the smallest action unit; DC circuit breakers only act on the faulty channel, AC circuit breakers only activate when most channels are faulty, and alarm actions are independently located, achieving a progressive response of "local isolation for local faults and global protection for system faults." During the recovery phase, recovery priorities are calculated based on power outage duration, test progress, and sample type. Combined with staggered recovery, progressive voltage soft-start, and communication time slot allocation, this avoids current surges and bus conflicts caused by simultaneous power-on of multiple channels, ensuring a smooth and orderly recovery process. This transforms the traditional "partial disconnection equals global outage" coupling-related damage into a controlled process of "precise isolation of faulty channels, continuous operation of normal channels, and orderly restart of recovery channels," significantly improving the operational continuity, fault tolerance, and resource utilization of multi-station aging test systems. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 A flowchart of a method for triggering program control circuit breaker execution based on heartbeat monitoring, provided in an embodiment of the present invention;

[0037] Figure 2 A flowchart illustrating the principle of generating a channel disconnection confirmation flag through multi-level anti-jitter decision-making, as provided in an embodiment of the present invention;

[0038] Figure 3 A schematic diagram illustrating the principle of a three-level circuit breaker decision-making method provided in an embodiment of the present invention;

[0039] Figure 4 This is a functional block diagram of a heartbeat monitoring-triggered program-controlled circuit breaker execution device provided in an embodiment of the present invention. Detailed Implementation

[0040] 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.

[0041] Example 1

[0042] Please see Figure 1 As shown, this embodiment provides a method for triggering program control circuit breaker execution based on heartbeat monitoring, including:

[0043] Step S10: Receive heartbeat signals from N test channels of the heartbeat circuit breaker box and construct the heartbeat interval feature sequence of each channel. Calculate the dynamic timeout threshold of each channel based on the heartbeat interval feature sequence of each channel. Based on the dynamic timeout threshold of each channel, generate a channel disconnection confirmation flag through multi-level anti-jitter decision.

[0044] Further, step S10 includes:

[0045] Step S11: Receive heartbeat signals from N test channels of the heartbeat circuit breaker box, record the arrival time of the heartbeat signal of each channel, and define the time interval between adjacent heartbeat signals as the heartbeat interval by calculating the time interval between adjacent heartbeat signals, and construct a heartbeat interval feature sequence for each channel.

[0046] The heartbeat circuit breaker box is a core device in the aging and durability testing of on-board chargers, used to monitor the communication status of each test channel and trigger circuit breaker protection in case of abnormalities. For the N independent test channels of the heartbeat circuit breaker box, where N is a positive integer and N≥2, representing the total number of configurable test channels, a dedicated register area needs to be allocated in the PLC memory for each channel. This area needs to be divided into a timing storage area and an interval storage area. The former stores the arrival time of the heartbeat signal, and the latter stores the calculated heartbeat interval value, thereby achieving physical isolation of the data from each channel and avoiding mutual interference. The PLC (Programmable Logic Controller) acts as the system control center, undertaking data storage, calculation, and control instruction output functions. When the host computer sends a heartbeat frame carrying the channel identifier to the serial port server via the TCP / IP protocol, the serial port server converts the heartbeat frame into a signal conforming to the RS485 interface communication standard. The PLC receives this signal through its RS485 communication module and, based on the channel identifier in the heartbeat frame, records the arrival time of the corresponding channel's heartbeat signal to the designated address in the timing storage area of ​​that channel, marking it as the timing value T. i(n), where i is the channel number, ranging from 1 to N; n is the heartbeat sequence number, a positive integer representing the nth heartbeat signal received by that channel. Subsequently, the PLC calls its built-in subroutine to read the timing value T of two adjacent heartbeats from the timing storage area of ​​the same channel. i (n) and T i (n-1), the time interval ΔT between adjacent heartbeats is obtained through difference calculation. i =T i (n)-T i (n-1), and ΔT i Write the interval data to the end of the channel's interval storage area. To ensure that the heartbeat interval feature sequence can reflect the current communication rhythm of the channel in real time, a sliding window mechanism is used to maintain the interval storage area data: the preset sliding window length is m, where m is a positive integer, and the value needs to be determined in conjunction with the test cycle length and the heartbeat transmission frequency to ensure that the sequence contains enough samples to characterize normal communication patterns. When the number of interval values ​​stored in the interval storage area reaches m, the PLC automatically deletes the oldest written interval value and then writes the newly calculated ΔT. i Write to the end to ensure that the interval storage area always contains m latest interval values, which together constitute the heartbeat interval feature sequence of this channel.

[0047] Traditional technologies often fail to provide independent data storage and processing units for each channel, resulting in the mixing of heartbeat signals from all channels. This makes it impossible to distinguish the differences in normal communication rhythms between different channels, and transient interference in a single channel during multi-channel operation can easily be misjudged as a global fault. Step S11 achieves isolated storage and independent processing of heartbeat data for each channel through independent register area division, ensuring that the communication status data of each channel is not interfered with by other channels, fundamentally solving the problem of confused judgment benchmarks caused by multi-channel data confusion. The introduction of the sliding window mechanism ensures that the heartbeat interval feature sequence always focuses on the latest m samples, avoiding interference from early outdated data on the judgment of the current communication status, and ensuring that the sequence can dynamically adapt to changes in communication status, solving the defect that static data sequences cannot reflect real-time communication rhythms. In addition, the independent register configuration and the dynamic maintenance logic of the sliding window allow the system to flexibly adjust the number of channels N and the window length m according to test requirements without reconstructing the data processing logic due to changes in N, providing good scalability and solving the problem that traditional fixed configuration systems cannot be compatible with different numbers of test channels. The heartbeat interval feature sequence constructed in step S11 is the only data source for calculating the dynamic timeout threshold in step S12. Without this step, step S12 will be unable to adaptively adjust the threshold due to the lack of quantitative data support, resulting in a lack of accurate basis for subsequent heartbeat loss judgment.

[0048] Step S12: Calculate the dynamic timeout threshold for each channel based on the heartbeat interval feature sequence of each channel;

[0049] Further, step S12 includes:

[0050] Step S121: Calculate the average value μ of the heartbeat interval for each channel i based on the heartbeat interval characteristic sequence of each channel. i and standard deviation σ i According to the average value μ i and standard deviation σ i Construct a stable interval for channel i; set an initial timeout baseline threshold for each channel;

[0051] Step S122: Based on the stable interval, determine the communication stability status of each channel: monitor the new heartbeat interval value of each channel in real time. When the new heartbeat interval values ​​of channel i fall within the stable interval for k1 consecutive times, channel i is determined to be in a stable communication state. When the new heartbeat interval values ​​of channel i exceed the stable interval for k2 consecutive times, channel i is determined to be in a fluctuating communication state. Here, k1 is the number of consecutive stable determinations and k2 is the number of consecutive fluctuating determinations.

[0052] Step S123: Based on the communication stability status of each channel, adaptively adjust the initial timeout reference threshold of each channel and output the dynamic timeout threshold of each channel: tighten the initial timeout reference threshold of the channel in the stable communication state using the stable state adjustment coefficient α, and relax the initial timeout reference threshold of the channel in the fluctuating communication state using the fault tolerance state adjustment coefficient β, and output the adaptively adjusted dynamic timeout threshold of each channel.

[0053] Specifically, in step S121, the PLC calls the data statistics subroutine to read the m interval values ​​of the heartbeat interval characteristic sequence of each channel, and obtains the average value μ by arithmetic mean calculation. i The standard deviation σ is obtained through standard deviation calculation. i Based on the statistical law of normal distribution, normal data mostly fall within the range of "mean ± k times standard deviation", where k is a positive integer. The value of k is determined according to the system's tolerance for fluctuations to ensure that normal fluctuations are not misjudged. A common range is 1.5-2.5. Therefore, the stable interval of channel i is set to [μ...]. i -kσ i ,μ i +kσ i At the same time, an initial timeout reference threshold T0 is set for each channel. The method for determining the value of T0 is as follows: first, obtain the preset transmission period Ts of the heartbeat signal, and then, considering the slight delay in normal communication, set T0 to 1.5 to 2 times Ts to ensure that the heartbeat is not misjudged as lost due to slight delay when there is no interference.

[0054] In step S122, the PLC monitors the relationship between the newly added heartbeat interval values ​​and the stable intervals of each channel in real time: A continuous stability judgment count k1 is set, where k1 is a positive integer. The value must exclude the influence of single, accidental fluctuations on the state judgment to ensure stable state judgment. When k1 consecutive newly added interval values ​​of channel i fall within the stable interval, the channel is determined to be in a stable communication state. A continuous fluctuation judgment count k2 is set, where k2 is a positive integer and its value is less than k1 to ensure timely detection of communication fluctuations and avoid delayed adjustments. When k2 consecutive newly added interval values ​​of channel i exceed the stable interval, the channel is determined to be in a communication fluctuation state. Common value ranges: k1 is 3~5, k2 is 1~2, for example, k1=3, k2=1.

[0055] In step S123, the PLC calls the threshold adjustment subroutine based on the communication stability status to adaptively adjust T0: For channels with stable communication, due to their stable communication rhythm and small fluctuations, the threshold needs to be tightened to improve fault response speed. The adjustment formula is T... dyn,i =T0×α. The value of α is related to σ. i Related, σ i The smaller the value of α (the more stable the communication), the smaller the value of α (the more obvious the threshold tightening). The value range of α is 0.8-1.0, T dyn,i This represents the dynamic timeout threshold for channel i. For channels experiencing communication fluctuations, due to their unstable communication rhythms and large fluctuations, the threshold needs to be relaxed to accommodate normal fluctuations. The adjustment formula is T. dyn,i =T0×β. The value of β ranges from 1.2 to 1.5; the more drastic the fluctuation, the larger the value of β, ensuring that the current fluctuation is contained. After adjustment, T... dyn,i It is stored in the threshold storage area of ​​the corresponding channel as the basis for determining heartbeat loss in step S13.

[0056] Traditional techniques use fixed timeout thresholds, which cannot adapt to the differences in communication status across different channels. When communication is stable, a threshold that is too wide leads to slow response to actual faults, while a threshold that is too strict leads to frequent false alarms when communication fluctuates. Step S12 addresses this issue through a multi-dimensional approach: In step S121, based on the μ of each channel... i With σ i A personalized stable range is constructed to avoid the shortcomings of traditional fixed ranges that cannot reflect individual channel differences. At the same time, a reasonable value for T0 provides a reliable benchmark for subsequent adjustments. In step S122, the communication status is determined by a series of consecutive checks to eliminate misjudgments caused by single fluctuations, ensuring the accuracy of status identification. In step S123, α and σ... i The correlation makes the threshold tighter for channels with more stable communication, improving fault response efficiency. β is adjusted according to the intensity of fluctuations, making the threshold wider for channels with more severe fluctuations, accommodating more normal fluctuations. The two work together to achieve precise threshold adaptation under different conditions.

[0057] The dynamic timeout threshold in step S12 provides a core judgment standard for the multi-level jitter prevention decision in step S13. Without this step, step S13 would rely solely on a fixed threshold, failing to distinguish between transient interference and genuine faults, leading to misjudgments or missed judgments. The dynamic threshold calculation is only related to the heartbeat interval feature sequence of a single channel and is independent of the total number of channels N. The calculation logic does not need to be modified when N changes. It works in synergy with the independent channel data processing mechanism in step S11. Step S11 provides personalized data input for step S12, and step S12 outputs personalized thresholds based on this data. Both ensure the independence and accuracy of the judgment standards for each channel in a multi-channel environment, avoiding the confusion in judgment logic caused by an increase in the number of channels. Furthermore, the introduction of the dynamic threshold indirectly improves the accuracy of the subsequent circuit breaker decision in step S20. Only based on accurate disconnection judgment can the circuit breaker decision avoid the coupling error of "local disconnection" leading to "global disconnection," providing key technical support for solving the problem of coupling error in multi-station operation.

[0058] Step S13: Determine whether there is a loss of heartbeat signal for each channel. If there is a loss of heartbeat signal, calculate the signal loss duration for the channel with the loss of heartbeat signal. Based on the signal loss duration and dynamic timeout threshold of the channel with the loss of heartbeat signal, generate a channel disconnection confirmation flag through multi-level anti-jitter decision.

[0059] Please see Figure 2 As shown, step S13 further includes:

[0060] Step S131: Monitor the reception status of heartbeat signals of each channel in real time, identify channels where heartbeat signals are lost, and record the signal loss duration of channels where heartbeat signals are lost.

[0061] Step S132: When the signal loss duration of a channel with a lost heartbeat signal exceeds the dynamic timeout threshold of the channel for the first time, the channel is placed in a "suspected disconnection" state and enters an observation period, and the observation period duration is set.

[0062] Step S133: During the observation period, continuously monitor the heartbeat signal of the channel in the "suspected disconnection" state; if no heartbeat signal is detected at the end of the observation period, then place the channel in the "suspected disconnection" state into the "real disconnection" state and generate a channel disconnection confirmation flag for the channel in the "real disconnection" state.

[0063] Specifically, heartbeat signal loss refers to the PLC failing to receive a new heartbeat frame from a certain channel within a preset monitoring period. Heartbeat signal loss must be determined based on channel-specific timing data, rather than a globally unified standard. In step S131, the PLC monitors the heartbeat signal reception status of each channel in real time, using its own scan cycle as the unit. For each channel i, the PLC reads the latest recorded heartbeat time T from the channel's timing storage area within each scan cycle. i (n), and call the system clock module to obtain the current time T. current The signal loss duration T is obtained through interpolation. lost,i =T current -T i (n). If the timing storage area of ​​channel i is not updated with a new T within two consecutive scan cycles. i If (n+1), it is determined that the channel has lost a heartbeat signal, and the real-time calculated T is... lost,i The lost time is stored in the channel's memory area to ensure real-time data updates; if a new heartbeat frame is detected, the T is immediately... lost,i Reset and update T i (n) represents the new heartbeat moment, maintaining the normal communication status of the channel.

[0064] In step S132, the PLC sets the T value of channel i in each scan cycle. lost,i The dynamic timeout threshold T in the channel threshold storage area dyn,i Comparison. When T is first detected. lost,i >T dyn,i When this happens, the PLC does not directly determine that the connection is lost. Instead, it calls the status control subroutine to set the status flag of the channel stored in the PLC's internal status register to "suspected connection lost" and simultaneously starts the channel's dedicated observation period timer T. obs,i The "suspected disconnection" state is an intermediate state between "normal communication" and "actual disconnection," used to buffer the impact of transient interference on circuit breaker decisions. The observation period is a time window set to verify whether the "suspected disconnection" is indeed due to transient interference; the duration of the observation period is set based on the current communication characteristics of the channel, taking the value T. dyn,i One-third, the value selection logic is based on: if T dyn,i The current communication fluctuations in the channel have been adapted. Its 1 / 3 duration provides recovery time for transient interference (such as brief communication interruptions caused by electromagnetic pulses) while avoiding delays in handling actual faults due to excessively long windows, ensuring a balance between false positive rate and response speed. The transient interference is such as a brief communication interruption caused by an electromagnetic pulse. After the observation period timer starts, the PLC updates T every scan cycle. obs,i The remaining duration, until the observation period ends or the heartbeat signal is restored.

[0065] In step S133, during the observation period, the PLC continuously monitors the heartbeat signal of channel i and T. obs,i Remaining time. If at T obs,i A new heartbeat frame was detected before the countdown ended. The PLC determined that this "suspected loss of connection" was a transient interference and immediately executed a reset operation: clearing T. obs,i For the remaining time, restore the channel status flag to "normal communication," and simultaneously mark an interference event in the metadata area of ​​the heartbeat interval characteristic sequence used to store sequence-related auxiliary information, recording the time and duration of the interference; if T obs,i If no new heartbeat frame is detected when the countdown ends, the PLC determines that the channel is "truly disconnected". The channel that was in the "suspected disconnected" state is moved to the "truly disconnected" state. The flag setting subroutine is called to set the channel disconnection confirmation flag of the "truly disconnected" channel to a valid state. The flag remains valid until the channel heartbeat is restored and the subsequent recovery process is completed.

[0066] In traditional technology, heartbeat loss judgment lacks an intermediate buffer state. Once the signal loss duration exceeds a fixed threshold, it is judged as disconnected and triggers a circuit breaker. This fails to distinguish between transient interference and real faults, leading to frequent false triggers during long-term endurance testing and interrupting normal testing. Step S131's real-time scanning monitoring and dedicated loss duration calculation ensure that the loss status of each channel is captured instantly, and the data is independent and unmixed, solving the judgment delay problem caused by the lag in traditional timed monitoring or the mixing of multi-channel data. Step S132's "suspected disconnection" intermediate state and associated duration observation period provide a verification window for the system, avoiding direct triggering of actions due to a single threshold exceeding the limit, thus solving the rigidity problem of fixed threshold judgment. Step S133's dynamic processing and interference marking during the observation period not only achieves automatic reset of transient interference but also provides data support for subsequent communication status analysis through metadata marking. The cumulative record of interference events can be used to optimize the dynamic timeout threshold in step S12. If interference is frequent in a certain channel, β is appropriately increased, forming a closed loop of "monitoring-judgment-optimization". The channel disconnection confirmation flag generated in step S13 is the sole trigger signal for updating the state control matrix in step S20. Without this step, step S20 will lack accurate faulty channel information, potentially leading to false disconnections of normal channels or missed disconnections of faulty channels. Simultaneously, the dynamic timeout threshold T... dyn,i The personalized adaptation makes the "suspected disconnection" judgment criteria more in line with the channel characteristics. Compared with the fixed threshold, the false judgment rate is significantly reduced, which provides a guarantee for the continuity of long-term testing.

[0067] Traditional heartbeat monitoring circuit breaker methods have flaws: multi-channel data ambiguity leads to inconsistent judgment criteria, fixed thresholds cannot adapt to differences in communication states, and the lack of buffered judgment results in frequent false triggers, ultimately causing "local disconnections" to be misjudged as "global faults," interrupting long-term testing. Step S10 forms a systematic solution through the collaboration of sub-steps: S11's independent sequence construction solves the data ambiguity problem, giving each channel its own "communication rhythm profile"; S12's dynamic threshold calculation solves the adaptability problem, allowing the judgment criteria to be dynamically adjusted according to the channel's communication state; S13's buffered confirmation solves the false trigger problem, making fault judgment more accurate. The three work together to achieve "personalized judgment" of multi-channel communication states. The judgment basis (sequence), standard (threshold), and result (flag) for each channel are independent and adapted to their own characteristics, completely breaking the traditional "one-size-fits-all" global judgment mode and fundamentally avoiding global disconnections caused by local interference. The precise disconnection flag output in step S10 enables step S20 to accurately locate the faulty channel, allowing for circuit breaking only on the faulty channel while continuous testing of normal channels. Compared to traditional global circuit breaking, this significantly improves testing efficiency. Without step S10, subsequent steps S20 and S30 would lack accurate fault information and a basis for judgment, potentially leading to chaotic circuit breaking decisions (false or missed breaks) or disordered recovery processes, such as inrush current or communication conflicts. The entire multi-station control scheme would lose its core judgment support. Simultaneously, step S10 transforms the abstract communication state into a concrete "sequence-threshold-flag," upgrading the system from qualitative fault perception to quantitative fault confirmation, providing a data foundation for subsequent fault analysis and preventative maintenance, such as the signal σ for a certain channel. i A continuous increase in size may indicate that the hardware communication module is aging and needs to be replaced in advance.

[0068] Step S20: Initialize the state control matrix, update the state control matrix according to the channel disconnection confirmation flag to obtain the real-time state control matrix; based on the real-time state control matrix, execute the three-level circuit breaker decision and generate the circuit breaker execution sequence for each channel;

[0069] Further, step S20 includes:

[0070] Step S21: Initialize an N-row, 3-column state control matrix. The row indices of the state control matrix correspond to the N test channels, and the column indices correspond to the three types of core control objects: DC output control, AC input control, and audible and visual alarm control. Update the state control matrix according to the channel disconnection confirmation flag to obtain the real-time state control matrix.

[0071] The status control matrix is ​​a two-dimensional data structure within the PLC used to quantify the permissions of each channel to different controlled objects. Its core function is to transform the abstract logic of "channel status - control permission" into calculable numerical weights, avoiding the ambiguity of logical judgments in traditional control. Within the PLC's internal data area, a dedicated storage area is designated using memory partitioning configuration instructions to create an N-row, 3-column two-dimensional matrix structure, i.e., the status control matrix. N is the total number of test channels, taking a positive integer value and N≥2. The matrix row index i (values ​​1 to N) corresponds one-to-one with the N test channels; row index i=1 corresponds to channel 1, row index i=2 corresponds to channel 2, and so on, ensuring that each channel has a unique data row in the matrix. As shown in Table 1, column indices j (values ​​1 to 3) correspond to three types of core controlled objects: column index j=1 for DC output control, j=2 for AC input control, and j=3 for audible and visual alarm control, ensuring that the classification of controlled objects is free of redundancy and omissions.

[0072] Table 1 Initialization Weights of the State Control Matrix

[0073]

[0074] During initialization, the state control matrix sets the control weights for DC output, AC input, and audible / visual alarms for each channel. For the DC output control column, each channel needs to independently control its own DC circuit to prevent other channel failures from affecting its own power supply. Therefore, all elements M[i,1] in this column are set to 1, indicating that channel i has 100% independent control over its own DC contactor. For the AC input control column, the total AC power supply needs to be controlled collaboratively by all channels, and a single channel has no right to cut off the total power supply alone. Therefore, all elements M[i,2] in this column are set to 1 / N, so that the sum of the AC control weights of all channels is always 1, ensuring that the control weights are evenly distributed among the channels and that the sum of the weights reflects the number of normal channels. For the audible / visual alarm control column, each channel needs to independently trigger its own alarm for fault location. Therefore, all elements M[i,3] in this column are set to 1, indicating that channel i has 100% independent control over its own audible / visual alarm.

[0075] The method for updating the status control matrix based on the channel disconnection confirmation flag is as follows: When a channel disconnection confirmation flag is received, the control weights of the corresponding channel's DC output, AC input, and audible / visual alarm are cleared to zero, and the corresponding channel's status is marked as "open circuit". Specifically, when the PLC detects that the channel disconnection confirmation flag for channel i is set, it immediately calls the matrix update subroutine, locates the i-th row of the matrix, and clears the control weights of all columns in that row to zero, i.e., M[i,1]=0, M[i,2]=0, M[i,3]=0. At the same time, in the PLC's internal status register, the status flag of channel i is updated from "normal operation" to "open circuit". The updated matrix is ​​the real-time status control matrix, which needs to be compared with the initialization matrix in each scan cycle of the PLC to ensure that the weight changes reflect the channel status changes in real time, without delay or data inconsistency.

[0076] Traditional technology employs globally coupled control logic without subdividing channel control rights. A single channel fault signal directly triggers the actions of all controlled objects, causing the normal channel test to be interrupted. Step S21 solves this problem through matrix-based and quantized control rights design: the independent weight of the DC output control column (M[i,1]=1) ensures that the faulty channel only loses its own DC control rights, while the normal channel retains 100% DC control rights, thus solving the defect of global linkage in traditional DC loops; the cooperative weight of the AC input control column (M[i,2]=1 / N) ensures that a single channel fault only reduces the AC control rights by 1 / N, while the total control rights remain at (N-1) / N, avoiding the problem of traditional single faults cutting off the total AC; the independent weight of the audible and visual alarm control column (M[i,3]=1) ensures that the faulty channel only triggers its own alarm, while the normal channel has no alarm action, thus solving the defect of traditional global alarms that cannot locate the faulty channel. Steps S21 and S13 form a collaborative link of "fault confirmation - permission update": the precise disconnection flag output by step S13 provides a unique trigger condition for matrix update, avoiding erroneous updates; the weight changes after matrix update provide a quantitative basis for the circuit breaker decision in step S22. Without step S21, step S22 would lack a clear standard for judging control rights and would have to revert to traditional global control. In addition, the scalable design of the matrix (row indexes automatically adapt to changes in N) allows the system to adjust the number of channels according to test requirements without reconstructing the matrix logic. For example, when N increases from 4 to 6, the AC control column elements automatically adjust from 1 / 4 to 1 / 6, and the total control weight always remains 1, solving the problem that traditional fixed control logic cannot adapt to changes in the number of channels.

[0077] Step S22, please refer to Figure 3As shown, based on the real-time state control matrix, a three-level circuit breaker decision is executed, generating a hierarchical circuit breaker decision result containing multiple circuit breaker actions to be executed; the first level of the three-level circuit breaker decision is the channel-level DC circuit breaker decision, the second level is the system-level AC coordinated circuit breaker decision, and the third level is the channel-level independent alarm decision.

[0078] Further, step S22 includes:

[0079] Step S221: Monitor the changes in the DC output control column elements of the real-time status control matrix. When any element is detected to change from 1 to 0, generate a channel-level DC circuit breaker action request for the corresponding channel.

[0080] Step S222: Calculate the sum Sac of all elements in the AC input control column of the real-time state control matrix, compare Sac with the AC collaborative circuit breaker threshold γ, and generate a system-level AC circuit breaker action request when Sac is less than γ.

[0081] Step S223: Monitor the changes in the elements of the audible and visual alarm control column of the real-time status control matrix. When any element changes from 1 to 0, generate a channel-level audible and visual alarm action request for the corresponding channel. Summarize the channel-level DC circuit breaker action request, the system-level AC circuit breaker action request, and the channel-level audible and visual alarm action request to form a hierarchical circuit breaker decision result.

[0082] Specifically, within each scan cycle, the PLC executes decision logic in three layers based on the weight changes of the real-time state control matrix, as shown in Table 2.

[0083] Table 2 Comparison of Three-Level Circuit Breaking Decisions

[0084]

[0085] The first layer is channel-level DC circuit breaker decision-making: This decision-making only targets the DC circuit breaker action of the faulty channel, without affecting the DC power supply of other channels. The control granularity is single-channel. The PLC calls the column monitoring subroutine to scan the DC output control column (j=1) of the real-time status control matrix. This subroutine identifies the row index i where the element changes from 1 to 0 by comparing the matrix element values ​​of the current scan cycle with those of the previous scan cycle. When such an element change is detected, a channel-level DC circuit breaker action request for the corresponding channel i is immediately generated. This request includes the channel number i and the action type "DC circuit breaker" and is sent to the PLC's action request buffer. Simultaneously, the PLC triggers the output relay bound to channel i, which directly controls the high-voltage DC contactor coil of channel i to de-energize, causing the contactor contacts to open and cutting off the DC circuit of channel i. The DC output control column element of the normal channel remains 1, and its DC contactor remains engaged, so the power supply is unaffected.

[0086] The second layer is system-level AC collaborative circuit breaker decision-making: System-level AC collaborative circuit breaker decision-making is based on the total AC circuit breaker action determined by the sum of the AC control rights of all channels, with the control granularity being the system level. The PLC calls the summation subroutine to calculate the sum Sac of all elements in the AC input control column (j=2) of the real-time status control matrix in real time. Subsequently, the threshold comparison subroutine is called to compare Sac with the preset AC collaborative circuit breaker threshold γ. The setting of γ is determined based on the system redundancy requirements: if the system allows less than half of the channels to continue operating despite failure, γ is set to 0.5, meaning the number of normal channels must be ≥ N / 2; if the system has a lower fault tolerance, γ can be set to 0.6, meaning the number of normal channels must be ≥ 0.6N. For example, when N=5, γ=0.5 corresponds to Sac<0.5, meaning that system-level protection is activated when there are ≤ 2 normal channels. When Sac < γ is detected, a majority of channels are deemed faulty, and a system-level AC circuit breaker action request is generated. This request includes the action type "AC circuit breaker" and is sent to the action request buffer. The PLC triggers an intermediate relay, which energizes the main AC circuit breaker coil, causing the circuit breaker contacts to open and disconnecting the system's main AC power supply. If Sac ≥ γ, the number of normal channels is determined to meet operational requirements, no AC circuit breaker request is generated, and the main AC power supply remains operational.

[0087] The third layer is channel-level independent alarm decision-making: Channel-level independent alarm decision-making only applies to the audible and visual alarm actions of the faulty channel itself, with a control granularity of single channel. The PLC calls the column monitoring subroutine to scan the audible and visual alarm control column (j=3) of the real-time status control matrix in real time, identifying the row index i where the element changes from 1 to 0. When such an element change is detected, a channel-level audible and visual alarm action request for the corresponding channel i is immediately generated. This request includes the channel number i and the action type "audible and visual alarm" and is sent to the action request buffer. At the same time, the PLC triggers the alarm output relay bound to channel i, which controls the audible and visual alarm of channel i to be energized, and the alarm emits an audible and visual alert. The audible and visual alarm control column elements of normal channels remain at 1, and their alarms do not activate, avoiding irrelevant alarms from interfering with operators. After the three-level decision is completed, the PLC calls the request summary subroutine to organize all requests (DC circuit breaker, AC circuit breaker, audible and visual alarm) in the action request buffer according to the format of "action type-channel number" to form a hierarchical circuit breaker decision result. This result must contain complete information on all actions to be executed, such as "DC circuit breaker-channel 2; audible and visual alarm-channel 2; AC circuit breaker-system", to ensure that the subsequent step S23 can clearly define the action object and sequence.

[0088] Traditional fault handling logic only offers two modes: "total disconnection" or "continuous operation." It cannot adjust control granularity based on the fault range, leading to either a single channel fault interrupting global testing or multiple channel faults leaving the system unprotected. Step S22 addresses this issue through three levels of decision-making, focusing on both control granularity and protection timing: Channel-level DC circuit breaker decision-making refines DC control granularity from "system" to "single channel," ensuring that a single channel fault only affects itself, while normal channels continue testing. This solves the test interruption problem caused by traditional global DC linkage, directly improving the continuity of long-cycle endurance testing. The DC circuit is a dedicated power supply circuit for each channel; independent circuit breaking will not affect other channels. Without this level of decision-making, local fault isolation cannot be achieved, reverting to traditional global control. System-level AC collaborative decision-making, through quantitative comparison of Sac and γ, accurately determines whether to activate system-level protection, avoiding "premature protection" (power outage due to a single fault) and "late protection" (no protection for multiple faults), overcoming the deficiency of traditional AC control lacking quantitative judgment standards. Channel-level independent alarm decision-making refines the alarm granularity from "system" to "single channel." Operators can directly locate the fault location through the alarm channel number, solving the problem of traditional global alarms being unable to distinguish faulty channels and reducing troubleshooting time. Its beneficial effects also include not interfering with normal testing: normal channels have no alarms, preventing operators from misjudging the status of normal channels and improving maintenance efficiency. Step S22, together with steps S21 and S13, forms a complete "fault confirmation - authority update - action decision" chain: S13 confirms the fault → S21 updates control authority → S22 makes an action decision based on control authority. These three steps work together to ensure the accuracy and timeliness of fault handling. The independence of the three-layer decision-making makes fault handling more flexible: for example, when a channel fails, only DC circuit breaking and an independent alarm are executed, while the total AC power supply remains to the normal channels, allowing normal channels to continue testing; if more channels subsequently fail, causing Sac < γ, then AC circuit breaking is executed, achieving a progressive response of "local protection → system protection," ensuring both testing continuity and system safety. If step S22 is missing, the control rights updated in step S21 cannot be translated into specific protection actions, the matrix design loses its practical meaning, and the system still cannot solve the problem of global coupling and accidental damage.

[0089] Step S23: When multiple channels in the hierarchical circuit breaker decision result need to perform circuit breaker actions simultaneously, the circuit breaker actions to be executed for the multiple channels are sorted according to a preset priority to form an execution queue, a safety delay is inserted between adjacent circuit breaker actions, and a time-sequential circuit breaker execution sequence is output.

[0090] The implementation of step S23 needs to be divided into two scenarios: "multi-channel concurrent circuit breaking" and "single-channel individual circuit breaking". For the multi-channel concurrent circuit breaking scenario: First, a preset priority list is configured. In the PLC's internal parameter configuration area, a channel priority parameter table is created, and a unique priority level is assigned to each channel. The priority allocation is determined according to the test task requirements: if a channel's test sample is a key verification object, such as a new type of vehicle charger prototype, it is given a higher priority; if it is a regular sample, it is given a lower priority. For example, the priorities of channels 1 to N can be set as "channel 1 > channel 2 > ... > channel N", or priority levels can be divided according to sample type, such as "core sample channel > regular sample channel > spare channel". After the priority list is configured, it is stored in the PLC's power-off retention area to ensure that it does not need to be reconfigured after the system restarts. When the hierarchical circuit breaking decision result output in step S22 contains circuit breaking action requests for multiple channels, the PLC calls the arbitration subroutine, reads all channel numbers in the decision result, compares them with the preset priority list, and sorts the circuit breaking action requests of each channel from high to low priority to form an execution queue. For example, if the decision result includes DC circuit breaker requests for channels 3, 1, and 2, and the priority list is "Channel 1 > Channel 2 > Channel 3", then the execution queue order is "Channel 1 DC circuit breaker → Channel 2 DC circuit breaker → Channel 3 DC circuit breaker". After the execution queue is generated, the PLC calls the timing control subroutine to insert a safety delay between two adjacent circuit breaker actions. The safety delay is set based on the characteristics of the electrical components: it needs to cover the mechanical action time of the contactor and circuit breaker (the time from coil energization to complete contact disconnection) to avoid current superposition caused by the subsequent component starting before the previous component has fully acted. The setting method is to obtain the maximum mechanical action time t of the contactor and circuit breaker used through actual measurement. mech Set the safety delay to t mech The timing control subroutine operates at 1.2 to 1.5 times the normal speed to ensure complete component operation. Following the logic of "action execution → delay wait → next action execution," the execution queue is transformed into a timed circuit breaker execution sequence containing action time, action object, and action type. For example, "t0 time: Channel 1 DC circuit breaker; t0+Δt time: Channel 2 DC circuit breaker; t0+2Δt time: Channel 3 DC circuit breaker," where Δt is a safety delay.

[0091] For single-channel isolated circuit breaking scenarios: When the hierarchical circuit breaking decision result contains only a circuit breaking action request for one channel, the PLC does not need to activate the priority arbitration logic and directly calls the simplified timing subroutine. This subroutine first verifies whether there are other potential concurrent requests in the decision result by comparing the action request buffer of the current scan cycle with that of the previous scan cycle. After confirming that there are no other concurrent requests, it generates a timing-based execution sequence containing only the circuit breaking action for that channel. The sequence retains a basic safety check step: Before the action is executed, the PLC checks the current state of the electrical components corresponding to that channel, such as whether the contactor is in the engaged state. After confirming that the state is normal, the circuit breaking action is initiated at the current moment without inserting a safety delay, because there are no other concurrent actions. However, the execution time of the action needs to be recorded in the PLC event log for subsequent fault tracing. For example, if the decision result only contains a DC circuit breaking request for channel 5 and confirms that there are no other concurrent requests, the timing-based execution sequence is "t1 time: DC circuit breaking for channel 5". Regardless of the scenario, after the timed circuit breaker execution sequence is generated, the PLC triggers the corresponding relay action according to the sequence instruction through the output module, and at the same time stores the sequence information in the PLC historical data area for operators to query the action execution record.

[0092] In traditional technologies, the lack of priority distinction and delay control when multiple circuit-breaking actions are executed concurrently leads to the simultaneous disconnection of high-power circuits, causing instantaneous current superposition, impacting the power grid and interfering with other equipment. Furthermore, the lack of standardized execution logic for single-channel circuit breaking makes the action prone to failure due to missing status detection. Step S23 addresses this issue through multi-dimensional technical means: the introduction of a preset priority list ensures that more important channels in the test task complete circuit breaking first, reducing the risk of damage to critical samples due to delayed fault handling, thus solving the problem of delayed critical channel processing caused by the lack of priority control in traditional methods; serialized management of the execution queue transforms concurrent requests into ordered execution, avoiding overlapping actions of electrical components and resolving the mechanical conflict problem of parallel execution; the insertion of a safety delay covers the mechanical action time of components, eliminating the hidden danger of instantaneous current superposition and solving the electrical impact problem caused by the lack of delay control in traditional methods; and simplified logic for single-channel scenarios reduces unnecessary delays while ensuring safety, avoiding reduced test efficiency and solving the resource waste problem caused by unified timing logic. Step S23 and step S22 form a collaborative "decision-execution" link: the hierarchical circuit breaker decision output by step S22 provides action input for step S23. Step S23 ensures the safe implementation of the decision result through time-sequential processing. If step S23 is missing, the decision of step S22 will lead to electrical faults due to disordered execution, and the goal of "precise circuit breaker" cannot be achieved.

[0093] Step S20 forms a complete circuit breaker control link of "quantified authority - hierarchical decision-making - sequential execution". The state control matrix is ​​the core data carrier of the link, transforming channel states into calculable control weights. The three-level circuit breaker decision is the logical core of the link, realizing fault handling at different granularities based on weights. The sequential circuit breaker execution sequence is the execution core of the link, ensuring the safe implementation of decision results. The three are progressively integrated to achieve the goal of "local fault isolation - system-level protection - execution safety". Traditional technology adopts global coupled control, without quantified control weight design, hierarchical decision logic, and sequential execution mechanism, resulting in a single channel failure triggering a global circuit breaker, and circuit breaker execution posing electrical and mechanical risks. Step S20 fundamentally solves this problem through the collaboration of three sub-steps: the state control matrix in step S21 transforms abstract control logic into concrete weights, solving the problem of ambiguity in the mapping between channel states and control weights in traditional control. The independence of DC control rights ensures that faulty channels only lose their own DC authority, while normal channels continue to be powered, achieving local isolation of local faults. The synergy of AC control rights means that system-level protection only triggers when most channels fail, avoiding premature protection interruptions to normal testing. The independence of alarm control rights allows faulty channels to alarm individually, facilitating rapid location. These three elements together construct a control foundation of "quantifiable authority and isolated faults." Step S22's three-level decision-making refines fault handling granularity from "system-level" to "channel-level + system-level," solving the problem of traditional single-granularity control. Channel-level DC circuit breaking and alarms ensure minimal fault impact, while system-level AC circuit breaking ensures system safety in extreme situations, forming a "progressive protection" logic: a single channel fault only affects that channel, and the main power is cut off only when most channels fail, balancing test continuity and system safety. Step S23's sequential execution addresses the safety risks of traditional parallel execution. Synergistic with the decision results of step S22, it ensures both "correct decision-making" and "safe execution," avoiding grid impact and equipment damage caused by concurrent actions, and providing a guarantee for the stability of long-term testing. If step S20 is missing, traditional global coupling control will continue to cause "local disconnection → global power outage", and the entire multi-station control scheme will lose its core fault handling capability.

[0094] Step S30: Read the real-time status control matrix to identify the channels to be restored, calculate the restoration priority value of each channel to be restored, and construct a restoration priority queue; perform staggered electrical restoration according to the restoration priority queue, and achieve stable electrical restoration and communication conflict avoidance by interacting with the detection heartbeat frame through gradual voltage soft start and communication time slot allocation.

[0095] Further, step S30 includes:

[0096] Step S31: Extract the channels in the "disconnected" state from the real-time status control matrix. When the recovery heartbeat signal of the "disconnected" channel is detected, the "disconnected" channel is taken as the channel to be restored. Calculate the recovery priority value of each channel to be restored and construct a recovery priority queue.

[0097] Specifically, the PLC calls the status scanning subroutine to read the status flag bits of each channel in the real-time status control matrix and extracts the channels marked as "disconnected," i.e., the channels with a weight of 0 in the i-th row of the matrix. For these channels, the PLC continuously monitors their heartbeat signal reception status, determining that the conditions for heartbeat signal recovery must meet two verifications: first, two consecutive heartbeat frames must be received to exclude false recovery from a single, accidental signal; second, the time interval between these two heartbeat frames must fall within the stable range of the channel, ensuring that the communication rhythm has returned to normal. When a "disconnected" channel simultaneously meets both conditions, the PLC marks it as a "channel to be restored" and records the moment when the first matching heartbeat frame is detected.

[0098] Recovery priority value of the channel to be recovered , where: T off The power outage duration refers to the time difference from the moment channel i is marked as "disconnected" to the moment the first heartbeat signal is detected. The PLC obtains T by reading the timestamp of the "disconnected" mark in its internal status register and the timestamp of the heartbeat detection moment, and then performing a difference calculation. off Its physical meaning is the duration of the channel interruption test due to fault, T off The longer the delay, the more severe the test schedule will be, and recovery should be prioritized to reduce the cost of repeated testing. off The weighting of power outage duration is related to T. off Positive correlation, used to quantify the impact of power outage duration on recovery priority. W off The settings are based on the preset total test duration T of the test task. total Based on the statistical results of the data validity loss rate, T total These are known parameters issued by the host computer, such as the T value for a 48-hour durability test. total =48h. The specific setting logic is: by statistically analyzing different power outage durations T... off T total The relationship between the proportion and the test data validity loss rate, T off The higher the percentage, the greater the probability of data loss due to interruption, requiring a corresponding increase in W. off Prioritize recovery. Accordingly, W will be... off Divided into three levels: When T off T total When the proportion is less than 10%, the data validity loss rate is low. off Take the smaller value w1, such as 0.2; when Toff T total When the proportion is between 10% and 30%, the data validity loss rate is moderate. off Take a moderate value for w2, such as 0.5; when T off T total When the proportion is greater than 30%, the data validity loss rate increases significantly. off Take the larger value w3, such as 0.8, where w3 > w2 > w1.

[0099] The testing schedule has been completed. done This refers to the duration of testing completed by channel i before the fault, read from the PLC's historical test log, and the ratio. This ratio reflects the test completion rate. The higher the ratio, the closer the channel is to test completion. Prioritizing recovery can avoid restarting tests that are close to completion due to interruption, thus reducing time costs. prog The weighting of test progress is related to... Positive correlation, used to prioritize high-completion channels. W prog The setting is based on testing resource efficiency: if a certain channel When the value is close to 1, such as 0.9, retesting after an interruption will consume a significant amount of time, therefore W prog The value is relatively large; if it is close to 0, such as 0.1, the cost of retesting is low. prog The value is relatively small. For example, W prog The value ranges from 0.3 to 0.7. For every 0.2 increase, W prog Increase by 0.1 to ensure the logic of prioritizing progress is implemented. type Weights are assigned based on sample type, with fixed weights assigned to test samples of different importance to prioritize the recovery of core sample channels. Sample importance is determined by the testing task requirements: if the channel tests a prototype of a novel on-board charger (core sample), W... type Take the higher value (e.g., 0.8); for regular mass production samples (ordinary samples), W type Take the lower value (e.g., 0.3); if it is a spare sample, W type Take the lowest value (e.g., 0.1). W type The parameters are stored in the PLC sample parameter configuration area and loaded when the test task starts. After the PLC calculates the recovery priority value of each channel to be recovered, it calls the sorting subroutine to arrange the channels in descending order of recovery priority value, forming a recovery priority queue. The head of the queue contains the highest priority channels, and the tail contains the lowest priority channels, ensuring that resources are allocated to the channels that most need recovery.

[0100] The design goal of the recovery priority calculation formula is to comprehensively consider multi-dimensional test requirements and achieve "prioritizing the recovery of the channels that most need to be recovered," avoiding resource waste or test losses caused by single-dimensional prioritization. This lays a reasonable sequential foundation for subsequent steps S32's staggered recovery, S33's gradual voltage soft-start, and communication conflict avoidance. Its design logic is first reflected in the dimension selection, specifically addressing several core contradictions during channel recovery, choosing "power outage duration T." off "and the weighting of power outage duration W" off This is because the longer the power outage lasts, the more severe the channel testing delays become. off With T off The positive correlation setting and the product of the two can amplify the priority ratio of channels with long power outages, reducing the increase in test costs caused by delays; select "Test Progress". "and the weight W of the test progress" prog The reason is that the closer the schedule is to completion, the more time is wasted on rework after an interruption. The product of these two factors can strengthen the priority of high-schedule channels and minimize rework losses. Selecting "Sample Type Influence Weight W" is also beneficial. type The reason for this is that there are differences between core samples (such as a prototype of a new on-board charger) and regular samples during testing. By assigning a fixed high weight to these core samples and including them as an additional factor, it ensures that core samples recover first even if they are slightly inferior in other dimensions, guaranteeing that key R&D tasks are not delayed. In terms of calculation, a "product + addition" approach is used instead of simple accumulation or multiplication to avoid problems such as inconsistent dimensional units, masked effects, or mutual exclusion: the "product" adjusts W... off W prog The numerical values ​​unify the results of the three dimensions to a similar range, making the impact quantifiable and quantifiable. The "addition" allows the impact of each dimension to be added independently, avoiding the lowering of the overall priority due to a small value in one dimension, and ensuring that the coordination of multi-dimensional needs does not conflict. For test resources, this formula prioritizes the recovery of channels with "long power outages, near progress, and core samples" by prioritizing them, minimizing the three types of losses: increased delays, rework waste, and delays in core tasks, thus significantly improving resource utilization. For the recovery process, the priority queue it generates avoids the problem of "concentrated recovery of high-demand channels" caused by random recovery order. Combined with the peak-shifting interval of S32 and the soft start and communication time slot allocation of S33, it reduces the risk of inrush current superposition and communication conflict from the source, providing a key sequence basis for "smooth, orderly, and fault-free" recovery, and directly serving the overall solution to the problem of "partial disconnection - global current interruption" coupling damage in multi-stations.

[0101] Step S32: Starting from the head of the recovery priority queue, perform staggered electrical recovery control on each channel to be restored.

[0102] The core of step S32 is to solve the problem of instantaneous inrush current caused by simultaneous power-up of multiple channels through the progressive logic of "restoration queue scheduling - pre-charge circuit control - main circuit switching". Its implementation requires association with the restoration priority queue in step S31 and relies on the inherent series structure of the pre-charge circuit to ensure controllable current. The pre-charge circuit is a series circuit composed of a pre-charge relay and a current-limiting resistor. It is an essential structure for DC restoration of the channel, not an optional configuration. Its core function is to achieve flexible charging of the capacitor through resistor current limiting. Main circuit switching refers to the action of switching from power supply from the pre-charge circuit to power supply from the main circuit contactor after pre-charging is completed, ensuring no additional resistance loss during normal operation.

[0103] The implementation process of step S32 is as follows: The PLC reads the recovery priority queue and schedules the recovery of each channel in sequence according to the queue order. For the first channel in the queue, the PLC calls the relay control subroutine to send a closing command to the pre-charge relay. This pre-charge relay is connected in series with the current-limiting resistor. The current must pass through the current-limiting resistor before flowing to the internal capacitor of the sample, such as the DC side filter capacitor of the on-board charger. This series structure ensures that the charging current is always limited by the resistance. The resistance value calculation of the current-limiting resistor must follow the dual constraints of "sample tolerance limit - Ohm's law": First, obtain the maximum allowable charging current I of the capacitor specified in the sample component datasheet. max The first step is to read the maximum surge current that the capacitor can withstand, such as the surge current parameter of an electrolytic capacitor. This value directly determines the upper limit of the charging current; exceeding this value will cause the internal electrodes of the capacitor to break down. The second step is to read the system's preset DC bus rated voltage U, such as the high-voltage DC voltage commonly used in on-board charger testing. This is a fixed system parameter stored in the PLC power parameter configuration area. The third step is to apply Ohm's law R=U / I. max Calculate the resistance value. When the initial voltage of the capacitor is 0, the peak value of the initial charging current is determined by both U and R, which can be expressed as R = U / I. max This ensures that the initial peak current is exactly equal to I. max To avoid exceeding the sample's tolerance limit.

[0104] The monitoring and judgment logic for the pre-charging process is as follows: The PLC collects the DC bus voltage in real time through the voltage sampling module. When the voltage reaches a preset percentage of the rated value, the pre-charging is determined to be complete. At this time, the PLC executes the main circuit switching action: first, it sends a closing command to the main circuit DC contactor; after the contactor contacts are fully engaged, it sends a disconnect command to the pre-charging relay to avoid instantaneous overlap of the two power supplies. The preset percentage is determined based on the capacitor charging characteristics. It is necessary to ensure that the capacitor has accumulated enough charge to reduce the impact when the main circuit closes. It is usually taken as 30%-40% of the rated value. If the percentage is too low, there will still be a large current impact when the main circuit closes; if the percentage is too high, it will prolong the pre-charging time and affect the recovery efficiency.

[0105] For the nth channel (n≥2) in the queue, its recovery start time is calculated using the formula T'(n)=T'(n-1)+Tr×(1+δn). The goal of this formula is to provide a precise time reference for multi-channel peak-shifting recovery, ensuring "orderly advancement + prevention of electrical resonance," thus completely resolving the shortcomings of traditional multi-channel recovery processes, such as "simultaneous power-on causing inrush current" and "fixed interval leading to resonance." Here, T'(n-1) refers to the recovery start time of the (n-1)th channel (i.e., the previous channel) in the recovery priority queue. Its function is to force the nth channel to start after the (n-1)th channel, eliminating parallel recovery of multiple channels in time sequence and avoiding current superposition. Tr is the basic recovery interval, which is the time difference between the start-up recovery processes of adjacent channels. Tr needs to cover the pre-charging and main circuit closing time in the recovery process of a single channel to avoid current superposition caused by the start of a subsequent channel before the previous channel has completed its recovery. The setting method is: the actual measured time t from recovery start-up of a single channel to stable power supply to the main circuit. rec Set Tr to t rec The current is 1.1 to 1.3 times that of the previous channel to ensure that the current of the previous channel stabilizes before starting the next channel, avoiding current superposition. δn is a random delay factor, generated by the PLC's internal random number generator, with a value ranging from 0 to 0.3. Its function is to break the electrical resonance that may be caused by a fixed recovery cycle, such as resonance caused by the superposition of current fluctuation frequencies when multiple channels recover at fixed intervals. The randomness of δn ensures that there are slight differences in the recovery intervals of each channel, avoiding the risk of resonance. T'(n-1), Tr, and δn form a complete logic chain of "sequential control + safety interval + random anti-resonance". The sequential constraints of T'(n-1) and the safety interval of Tr prevent the instantaneous inrush current caused by the simultaneous power-up of multiple channels, while the randomness of δn eliminates the risk of electrical resonance, ensuring the safety of equipment and power grid. In terms of recovery efficiency and sequence, it can strictly cooperate with the recovery priority queue generated in step S31 to ensure that the high-priority core channels are restored first and the process is not affected by subsequent channels. At the same time, the 1.1-1.3 times setting of Tr avoids excessively prolonging the recovery time on the basis of safety, balancing safety and efficiency. The acquisition of T'(n-1) depends on the PLC's time recording mechanism. When the (n-1)th channel starts recovery according to the recovery priority queue, the PLC will automatically record its start time to the internal historical data area or event log and associate it with the channel number and priority information. When calculating the start time of the nth channel, the PLC can directly read the record from the historical data area or event log through the channel number and queue order to obtain T'(n-1).

[0106] The shortcomings of traditional technologies lie in the lack of peak-shaving scheduling, leading to simultaneous power-on of multiple channels. The initial capacitor voltage of 0 triggers a momentary short-circuit current, easily causing the upstream circuit breaker to trip. The absence of a pre-charge circuit or reliance solely on relay power supply results in unrestricted current, potentially damaging the sample capacitor. The lack of random delay introduces the risk of electrical resonance. In step S32, the series structure of the pre-charge circuit physically limits the current peak, preventing capacitor surges. Peak-shaving scheduling, via Tr, enables sequential channel recovery, eliminating the superposition of currents from simultaneous power-on. The random delay factor δn mitigates resonance risks; these three elements work together to ensure electrical safety during the recovery process. Without a pre-charge circuit, directly closing the main circuit contactor would damage the sample capacitor due to a large current surge, rendering the recovery process meaningless. Without peak-shaving scheduling, the surge current from simultaneous power-on of multiple channels would trigger system-level protection, causing all channels to fail to recover. Without random delay, resonance could cause grid voltage instability, affecting the normal test operation of the channels. Meanwhile, steps S32 and S31 work together: the recovery priority queue ensures that critical channels are restored first, and the off-peak scheduling ensures that the restoration process of priority channels is not disturbed by subsequent channels. The combination of the two achieves the dual goal of "resource priority allocation - safe execution". For example, when the core sample channel is restored first, Tr and δn ensure that its restoration process is stable and avoids being affected by the restoration actions of subsequent ordinary channels.

[0107] Step S33: Perform progressive voltage soft start and heartbeat detection frame interaction on each channel to be restored, and combine communication time slot allocation to achieve electrical stability restoration and communication conflict avoidance.

[0108] The core of step S33 is to solve the stress impact of sudden voltage rise on the sample and the communication conflict problem of multiple channels through the collaborative logic of "phased voltage ramp-current change rate monitoring-communication time slot allocation". Progressive voltage soft start refers to controlling the adjustable power supply output voltage in stages to achieve a smooth ramp from 0 to the rated value, avoiding stress impact on the internal components of the sample caused by sudden voltage rise. The implementation process of progressive voltage soft start is as follows: the PLC sends a voltage control signal to the adjustable power supply through the analog output module, dividing the voltage ramp into three stages:

[0109] Slow start-up phase: The voltage rises from 0 to 30% of the rated value (i.e., rated voltage Uq), with the ramp-up rate set to a low rate, such as 0.5Uq / min, meaning the voltage increases by 0.5 times the rated voltage per minute. During this phase, the small-capacity capacitors inside the sample begin to charge. The low rate avoids excessive capacitor voltage stress caused by a sudden voltage rise, ensuring the safety of small-capacity components.

[0110] Rapid ramp-up phase: The voltage rises from 30% to 80%, with the ramp-up rate set to a medium rate, such as 2Uq / min, meaning the voltage increases by twice the rated voltage per minute. During this phase, the small-capacity capacitor is fully charged, and the large-capacity capacitor enters a stable charging phase. The medium rate can shorten the recovery time while ensuring safety, avoiding excessively long soft starts that could affect test efficiency.

[0111] Stable approach phase: The voltage is increased from 80% to 100% at a low ramp rate, such as 0.3 Uq / min, which means the voltage increases by 0.3 times the rated voltage per minute. During this phase, the voltage approaches the rated value, and the low ramp rate avoids voltage overshoot, ensuring that the final voltage stabilizes at the rated value and preventing damage to the sample due to overvoltage.

[0112] The communication conflict avoidance process is as follows: When the channel voltage reaches 80% of its rated value, the sample's electrical state is nearly stable and ready for communication reconstruction. The host computer then sends a "detection heartbeat frame" to that channel. To avoid conflicts caused by multiple recovery channels simultaneously sending data on the RS485 bus, communication time slots are allocated according to the recovery priority queue: the start time of the communication time slot for the m'th channel in the queue is set to "the moment when the channel voltage reaches 80% + (m'-1)×t". slot ”, where t slot The communication time slot duration for a single channel is determined based on the heartbeat detection frame length and RS485 communication rate, such as 100ms. Each channel can only send and receive data within its dedicated time slot to ensure no overlap in bus data transmission. When the host computer receives a "ready signal" from the sample within its dedicated time slot, it reintegrates that channel into the regular heartbeat monitoring cycle of step S11. The ready signal contains the sample's internal status code, such as "capacitor charging complete" or "module ready." If no signal is received, the communication time slot is adjusted at interval t. retry Then resend the heartbeat detection frame to avoid recovery interruption caused by a single communication failure; t retry Based on communication reliability settings, such as t retry For t slot The traditional technique suffers from drawbacks: instantaneous voltage loading can damage sample components due to voltage stress; simultaneous communication across multiple channels can cause RS485 bus data collisions, leading to communication reconstruction failure. Step S33 avoids voltage stress surges through phased voltage ramp-up and eliminates bus conflicts through communication time slot allocation, ensuring dual stability for both electrical recovery and communication reconstruction. Without gradual voltage soft-start, a sudden voltage surge can damage internal components of the sample, preventing the recovered sample from participating in testing normally; without communication time slot allocation, RS485 bus conflicts can lead to communication reconstruction failure, preventing the channel from being reintegrated into regular monitoring and rendering the recovery meaningless.

[0113] Step S20 aims to precisely isolate and protect the out-of-connection channels confirmed in Step S13, preventing the spread of local channel failures from affecting overall testing, while ensuring that normal channels that are not out of contact can continue to conduct on-board charger aging and durability testing. Step S30 aims to identify channels that meet the recovery conditions from the disconnected channels, enabling them to safely and orderly return to the testing process and reducing test interruption losses caused by channel disconnection. The connection between the two is that the real-time status control matrix updated in Step S20 is the core basis for identifying channels to be restored in Step S30, and together they form a complete fault handling closed loop of "precise disconnection of faulty channels - safe restoration of channels that meet the conditions," both revolving around the goal of "ensuring the continuity and safety of multi-channel testing," ensuring that the testing process is not interrupted by local failures and that testing can be efficiently restarted after the channel is restored. Step S20, through precise fault isolation logic, only disconnects the faulty channel experiencing a "partial disconnection," while maintaining power supply and testing for normal channels. This fundamentally avoids a "global power outage" caused by a "partial disconnection." By employing layered decision-making, it cuts off the DC circuit of the faulty channel only and the system's AC power supply only when most channels fail, completely overcoming the shortcomings of traditional global coupling control and directly resolving the core contradiction of "partial disconnection equals global power outage." Step S30, through orderly channel recovery logic, avoids potential electrical shocks or communication conflicts during channel recovery. These problems, if left uncontrolled, could indirectly cause system fluctuations or even a global power outage, further consolidating the solution to the "coupling accidental damage problem." It ensures that the recovery process is carried out orderly only on a single channel to be recovered, without affecting the testing of other normal channels, avoiding the risk of a new "global power outage" due to improper recovery operations.

[0114] Example 2

[0115] This embodiment, based on embodiment 1, provides a program-controlled circuit breaker execution device triggered by heartbeat monitoring, such as... Figure 4 As shown, it includes:

[0116] Disconnection confirmation module: used to receive heartbeat signals from N test channels of the heartbeat circuit breaker box and construct the heartbeat interval feature sequence of each channel. Based on the heartbeat interval feature sequence of each channel, it calculates the dynamic timeout threshold of each channel. Based on the dynamic timeout threshold of each channel, it generates a channel disconnection confirmation flag through multi-level anti-jitter decision.

[0117] Circuit breaker decision module: used to initialize the status control matrix, update the status control matrix according to the channel disconnection confirmation flag, and obtain the real-time status control matrix; based on the real-time status control matrix, execute three-level circuit breaker decisions and generate the circuit breaker execution sequence for each channel; the first level of the three-level circuit breaker decision is the channel-level DC circuit breaker decision, the second level is the system-level AC coordinated circuit breaker decision, and the third level is the channel-level independent alarm decision.

[0118] Stable recovery module: It is used to read the real-time status control matrix to identify the channels to be restored, calculate the recovery priority value of each channel to be restored, and build a recovery priority queue; it performs staggered electrical restoration according to the recovery priority queue, and achieves stable electrical restoration and communication conflict avoidance by interacting with the detection heartbeat frame through gradual voltage soft start and communication time slot allocation.

[0119] Furthermore, in the loss of connection confirmation module, the method for generating a channel loss of connection confirmation flag through multi-level jitter stabilization decision includes:

[0120] Step S131: Monitor the reception status of heartbeat signals of each channel in real time, identify channels where heartbeat signals are lost, and record the signal loss duration of channels where heartbeat signals are lost.

[0121] Step S132: When the signal loss duration of a channel with a lost heartbeat signal exceeds the dynamic timeout threshold of the channel for the first time, the channel is placed in a "suspected disconnection" state and enters an observation period, and the observation period duration is set.

[0122] Step S133: During the observation period, continuously monitor the heartbeat signal of the channel in the "suspected disconnection" state; if no heartbeat signal is detected at the end of the observation period, then place the channel in the "suspected disconnection" state into the "real disconnection" state and generate a channel disconnection confirmation flag for the channel in the "real disconnection" state.

[0123] Furthermore, in the circuit breaker decision module, the method for executing the three-level circuit breaker decision includes:

[0124] Step S221: Monitor the changes in the DC output control column elements of the real-time status control matrix. When any element is detected to change from 1 to 0, generate a channel-level DC circuit breaker action request for the corresponding channel.

[0125] Step S222: Calculate the sum Sac of all elements in the AC input control column of the real-time state control matrix, compare Sac with the AC collaborative circuit breaker threshold γ, and generate a system-level AC circuit breaker action request when Sac is less than γ.

[0126] Step S223: Monitor the changes in the elements of the audible and visual alarm control column of the real-time status control matrix. When any element changes from 1 to 0, generate a channel-level audible and visual alarm action request for the corresponding channel. Summarize the channel-level DC circuit breaker action request, the system-level AC circuit breaker action request, and the channel-level audible and visual alarm action request to form a hierarchical circuit breaker decision result.

[0127] The methods and apparatus of this application may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the method is for illustrative purposes only, and the steps of the method of this application are not limited to the order specifically described above, unless otherwise specifically stated.

[0128] In addition, the parts of the technical solutions provided in the embodiments of this application that are consistent with the implementation principles of the corresponding technical solutions in the prior art have not been described in detail, so as to avoid excessive elaboration.

[0129] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the invention. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for triggering program control circuit breaker execution based on heartbeat monitoring, characterized in that, The method includes: Receive heartbeat signals from N test channels of the heartbeat circuit breaker box and construct the heartbeat interval feature sequence of each channel. Calculate the dynamic timeout threshold of each channel based on the heartbeat interval feature sequence of each channel. Based on the dynamic timeout threshold of each channel, generate a channel disconnection confirmation flag through multi-level anti-jitter decision. The status control matrix is ​​initialized and updated according to the channel disconnection confirmation flag to obtain the real-time status control matrix. Based on the real-time status control matrix, a three-level circuit breaker decision is executed, and a circuit breaker execution sequence for each channel is generated. The first level of the three-level circuit breaker decision is the channel-level DC circuit breaker decision, the second level is the system-level AC coordinated circuit breaker decision, and the third level is the channel-level independent alarm decision. The status control matrix is ​​N rows and 3 columns, and the three columns of the status control matrix are the DC output control column, the AC input control column, and the audible and visual alarm control column, respectively. The execution method of the system-level AC collaborative circuit breaker decision is as follows: calculate the sum Sac of all elements in the AC input control column of the real-time state control matrix, compare Sac with the AC collaborative circuit breaker threshold γ, and generate a system-level AC circuit breaker action request when Sac is less than γ. The channels in the "disconnected" state are extracted from the real-time status control matrix. When a recovery heartbeat signal of a channel in the "disconnected" state is detected, the channel in the "disconnected" state is taken as the channel to be restored. The recovery priority value of each channel to be restored is calculated, and a recovery priority queue is constructed. Off-peak electrical restoration is performed according to the recovery priority queue. Through gradual voltage soft start and interaction with the detection heartbeat frame, combined with communication time slot allocation, electrical stability restoration and communication conflict avoidance are achieved.

2. The method for triggering program control circuit breaking execution based on heartbeat monitoring according to claim 1, characterized in that, The method for constructing the heartbeat interval feature sequence of each channel includes: The arrival time of the heartbeat signal for each channel is recorded. By calculating the time interval between adjacent heartbeat signals, the time interval between adjacent heartbeat signals is defined as the heartbeat interval, and a heartbeat interval feature sequence specific to each channel is constructed.

3. The method for triggering program control circuit breaking execution based on heartbeat monitoring according to claim 2, characterized in that, The method for calculating the dynamic timeout threshold of each channel based on the heartbeat interval feature sequence of each channel includes: Based on the heartbeat interval characteristic sequence of each channel, calculate the average value μ of the heartbeat interval for each channel i. i and standard deviation σ i According to the average value μ i and standard deviation σ i Construct a stable interval for channel i; set an initial timeout baseline threshold for each channel; Based on the stable range, the communication stability status of each channel is determined; according to the communication stability status of each channel, the initial timeout reference threshold of each channel is adaptively adjusted, and the dynamic timeout threshold of each channel is output.

4. The method for triggering program control circuit breaking execution based on heartbeat monitoring according to claim 3, characterized in that, The method for determining the communication stability status of each channel includes: Real-time monitoring of new heartbeat interval values ​​for each channel. When k1 consecutive new heartbeat interval values ​​for channel i fall within the stable range, channel i is determined to be in a stable communication state. When k2 consecutive new heartbeat interval values ​​for channel i exceed the stable range, channel i is determined to be in a fluctuating communication state. Here, k1 is the number of consecutive stable determinations and k2 is the number of consecutive fluctuating determinations.

5. The method for triggering program control circuit breaking execution based on heartbeat monitoring according to claim 4, characterized in that, The method for adaptively adjusting the initial timeout reference threshold for each channel includes: For channels in a stable communication state, the initial timeout reference threshold of the channel is tightened using a stable state adjustment coefficient α. For channels in a fluctuating communication state, the initial timeout reference threshold of the channel is relaxed using a fault-tolerant state adjustment coefficient β. The dynamic timeout threshold of each channel is then output after adaptive adjustment.

6. The method for triggering program control circuit breaking execution based on heartbeat monitoring according to claim 5, characterized in that, The method for generating a channel disconnection confirmation flag through multi-level anti-jitter decision includes: The system monitors the reception status of heartbeat signals in each channel in real time, identifies channels where heartbeat signals are lost, and records the duration of signal loss in channels where heartbeat signals are lost. When the signal loss duration of a channel with a lost heartbeat signal first exceeds the dynamic timeout threshold of the channel, the channel is placed in a "suspected disconnection" state and enters an observation period, the duration of which is set. During the observation period, the heartbeat signal of the channel in the "suspected disconnection" state is continuously monitored; if no heartbeat signal is detected at the end of the observation period, the channel in the "suspected disconnection" state is placed in the "actual disconnection" state, and a channel disconnection confirmation flag is generated for the channel in the "actual disconnection" state.

7. The method for triggering program control circuit breaker execution based on heartbeat monitoring according to claim 6, characterized in that, The row index of the state control matrix corresponds to N test channels, and the column index corresponds to three types of core control objects: DC output control, AC input control, and audible and visual alarm control. When initializing the state control matrix, all elements in the DC output control column are set to 1, all elements in the AC input control column are set to 1 / N, and all elements in the audible and visual alarm control column are set to 1.

8. The method for triggering program control circuit breaker execution based on heartbeat monitoring according to claim 7, characterized in that, The execution method of the channel-level DC circuit breaker decision is as follows: monitor the changes in the elements of the DC output control column of the real-time state control matrix, and generate a channel-level DC circuit breaker action request for the corresponding channel when any element is detected to change from 1 to 0.

9. A heartbeat monitoring-triggered program control circuit breaker execution device, used to implement the heartbeat monitoring-triggered program control circuit breaker execution method according to any one of claims 1-8, characterized in that, The device includes: Disconnection confirmation module: used to receive heartbeat signals from N test channels of the heartbeat circuit breaker box and construct the heartbeat interval feature sequence of each channel. Based on the heartbeat interval feature sequence of each channel, it calculates the dynamic timeout threshold of each channel. Based on the dynamic timeout threshold of each channel, it generates a channel disconnection confirmation flag through multi-level anti-jitter decision. Circuit breaker decision module: used to initialize the status control matrix, update the status control matrix according to the channel disconnection confirmation flag, and obtain the real-time status control matrix; based on the real-time status control matrix, execute three-level circuit breaker decisions and generate the circuit breaker execution sequence for each channel; the first level of the three-level circuit breaker decision is the channel-level DC circuit breaker decision, the second level is the system-level AC coordinated circuit breaker decision, and the third level is the channel-level independent alarm decision. Stable recovery module: It is used to read the real-time status control matrix to identify the channels to be restored, calculate the recovery priority value of each channel to be restored, and build a recovery priority queue; it performs staggered electrical restoration according to the recovery priority queue, and achieves stable electrical restoration and communication conflict avoidance by interacting with the detection heartbeat frame through gradual voltage soft start and communication time slot allocation.

Images