Intelligent switch cabinet anti-misoperation locking device based on adaptive discriminant algorithm

Abstract

The application discloses an intelligent switch cabinet anti-misoperation locking device based on an adaptive discrimination algorithm, belongs to the technical field of power safety protection, and comprises optical voltage induction sensors, a signal processing module, an algorithm module, a control execution unit and a mechanical key linkage mechanism which are sequentially connected. The application adopts optical sensing to collect voltage signals, determines the live state of a circuit through a residual voltage decay model and a double-layer adaptive threshold algorithm, and realizes unlocking by independent power supply of an intelligent key. In addition, the application forms an operation closed loop by cooperating with a key forced linkage mechanism, has extremely high discrimination accuracy and operation reliability, and can be widely applied to the anti-misoperation protection of switch cabinets of various voltage grades.

Description

Technical Field

[0001] This application relates to the field of power safety protection technology, and in particular to an intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm. Background Technology

[0002] Switchgear plays a crucial role in the control and protection functions of power systems, and its operational safety directly impacts grid stability and the personal safety of maintenance personnel. According to the State Grid's 2025 power operation and maintenance accident statistics report, there were 127 accidents involving personal injury and equipment damage caused by switchgear misoperation throughout the year. Of these, 82% stemmed from two types of violations: "closing a grounding switch while it's energized" and "closing a grounding switch while it's energized." The former easily leads to direct grounding of energized circuits, causing a sudden increase in short-circuit current and potentially triggering major grid accidents such as switchgear explosions and busbar undervoltage. The latter can cause the grounding circuit to suddenly become energized, directly threatening the lives of on-site maintenance personnel. The key to avoiding such accidents lies in accurately and reliably identifying the operating status of primary equipment and preventing operational actions through interlocking or blocking mechanisms when operating conditions are not met, thus ensuring compliance with operational procedures through mandatory means.

[0003] In long-term engineering practice, mechanical interlocking was the earliest applied anti-misoperation solution. Its basic principle is to achieve direct physical interlocking through mechanical structures. For example, when the grounding switch is in the closed position, a mechanical stop prevents the circuit breaker's operating handle from moving, thus avoiding "energizing with the grounding switch closed." The advantages of this method are its simple structure, strong independence, and lack of reliance on external power sources, making it widely used in early medium and low-voltage power distribution systems. However, as power systems have expanded in scale and switchgear operation has become increasingly complex, the limitations of relying solely on mechanical interlocking have become increasingly apparent: it can only achieve interlocking based on fixed operating logic, cannot adapt to complex operation and maintenance scenarios involving multiple bays and cross-cabinet interlocks, and mechanical components are prone to jamming and wear after long-term operation, leading to interlocking failure.

[0004] Electrical interlocking is currently the mainstream anti-misoperation technology. Its core principle is to determine whether circuit breaker or grounding switch operation is permitted by judging whether the circuit is energized. Forced live indicator is the most commonly used electrical interlocking device. It not only displays whether the primary circuit is energized but also sends a signal directly to the electromagnetic lock. The electromagnetic lock controls the switch cabinet door or grounding switch, achieving the control objective of "locking when energized and unlocking when de-energized." A conventional electrical interlocking system mainly consists of four parts: a voltage transformer (PT), a high-voltage capacitive sensor (CGQ), a forced live indicator, and an electromagnetic lock. The CGQ is installed at the power input terminal of the high-voltage incoming cabinet. When the high voltage is energized, it induces a low voltage through its built-in capacitor and outputs it to the forced live indicator. The forced live indicator is usually installed on the cabinet door and is powered by the PT. When it receives an energized signal, it controls the electromagnetic lock to lock, preventing personnel from opening the cable compartment door. When the system is de-energized, the electromagnetic lock should release to allow normal operation.

[0005] However, this traditional electrical interlocking scheme still has many shortcomings in actual operation:

[0006] Insufficient power supply reliability: The system power supply relies on a PT (Power Transmission Unit). Once the PT loses voltage, the entire electrical interlocking system loses power, and the electromagnetic lock remains in a normally closed state. Even if maintenance personnel confirm the circuit is de-energized through other means, they cannot unlock the lock and can only force it open with a mechanical key. At this point, the anti-misoperation function is completely degraded, equivalent to the protection level of an ordinary brass padlock. Some solutions use an uninterruptible power supply (UPS) to power the live display, but UPS has poor reliability in the high-temperature, enclosed, and strong electromagnetic interference environment inside the switch cabinet. Batteries are prone to aging, leakage, and insufficient discharge, still making it difficult to avoid unlocking failures.

[0007] Insufficient accuracy of voltage criteria: Traditional capacitive sensors are susceptible to interference from stray electric fields inside the switchgear and induction from nearby circuits. After the primary circuit is de-energized, voltage misjudgments can easily occur during the decay of residual charge. This can lead to either a misjudgment of "energized," preventing the interlock from being released, or a misjudgment of "de-energized," posing a safety risk of opening the door while it is energized. Furthermore, traditional solutions mostly use fixed voltage thresholds for discrimination, which cannot adapt to the residual voltage decay characteristics under different voltage levels, environmental temperatures and humidity, and varying degrees of equipment aging.

[0008] Insufficient operational closed-loop: The existing electromagnetic lock and the operating key do not have a forced linkage mechanism. After unlocking, the operator can pull out the key at will, which can easily lead to problems such as the cabinet door not being relocked and the key being lost, leaving hidden dangers for operation and maintenance.

[0009] To address the aforementioned issues, various improved intelligent anti-misoperation interlocking solutions have been introduced in the industry, but they still have many limitations: some solutions introduce algorithmic discrimination logic, but only use simple threshold comparisons, without considering the dynamic attenuation characteristics of residual voltage, resulting in weak adaptive capabilities; some solutions optimize the power supply method, but still rely on the fixed power supply inside the cabinet, failing to fundamentally solve the problem of power failure interlocking. Therefore, there is an urgent need for a new type of intelligent switchgear anti-misoperation interlocking device that can simultaneously solve the three core problems of power supply reliability, discrimination accuracy, and operational closed-loop, meeting the power operation and maintenance requirements of high reliability, low maintenance, and adaptability. Summary of the Invention

[0010] This application provides an intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm. By establishing a residual voltage attenuation model and a two-layer parameter optimization mechanism, it realizes dynamic identification of voltage status to solve the problems of misjudgment and misoperation interlocking.

[0011] To achieve the above objectives, this application provides an intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm, comprising an optical voltage sensing sensor, a signal processing module, an algorithm module, a control execution unit, and a mechanical key linkage mechanism connected in sequence.

[0012] The optical voltage sensing sensor is used to acquire the voltage signal of the three-phase conductors in the primary circuit and convert the sensed signal into a low-voltage photoelectric signal output; the signal processing module performs bandpass filtering, sampling and normalization processing on the low-voltage photoelectric signal to obtain the characteristic sequence of the three-phase voltage.

[0013] The algorithm module establishes a residual voltage attenuation model and generates an adaptive discrimination threshold. It dynamically adjusts the discrimination parameters in conjunction with a two-layer parameter optimization mechanism to determine the energized state of the primary circuit. If the voltage of any one of the three-phase voltage channels is higher than the discrimination threshold, a blocking signal is output. If all three-phase voltages are lower than the discrimination threshold and the duration exceeds the set delay, an unlocking signal is output.

[0014] The control execution unit is used to receive signals from the algorithm module. When an unlock signal is received and the smart key is inserted, the smart key's built-in battery compartment provides instantaneous power to drive the electromagnetic bolt to retract, releasing the cabinet door latch to unlock. When a locking signal is received, the electromagnetic bolt remains extended to maintain locking. The mechanical key linkage mechanism forcibly locks the key in the lock body during the bolt retraction until the bolt retracts again, forming an operation closed loop.

[0015] In one embodiment, the optical voltage sensing sensor is based on the Pockels effect principle of electro-optic crystals, and its structure includes an electro-optic crystal, a polarizer, an analyzer, a quarter-wave plate, and a light source and photoelectric conversion module; the optical voltage sensing sensor is arranged near the high-voltage live parts inside the equipment and meets the safety insulation distance requirements.

[0016] In one embodiment, the algorithm module performs discrete sampling on the characteristic sequence of the acquired three-phase voltage and constructs a residual voltage attenuation model using the least squares method:

[0017] ; ①

[0018] In formula ① Let be the residual voltage at time t. Let τ be the initial voltage and τ be the voltage natural decay constant; the algorithm dynamically updates τ to obtain the predicted curve. Furthermore, a noise disturbance term is introduced to form an adaptive criterion threshold:

[0019] ;②

[0020] In formula ② β is the standard deviation of the voltage signal within the sampling window, and β is the weighting parameter; when the three-phase voltage Any channel exceeding the discrimination threshold When the primary circuit is determined to be energized, the lockout is maintained; when the three-phase voltage... All are below the discrimination threshold And the duration is not less than the delay threshold. When the time is right, it is determined to be in a safe, pressure-free state and the unlocking process is triggered.

[0021] In one embodiment, the algorithm module further introduces a two-layer parameter adaptive optimization mechanism based on the adaptive criterion threshold:

[0022] The first layer of fast adaptive correction: Within each sampling period, based on the real-time deviation between the actual false alarm rate F and the average unlocking delay D, the weight parameter β and the delay threshold are dynamically updated using a linear gain rule. ;

[0023] The second layer of slow global optimization: At multiple sampling period time scales, the weight parameter β and the delay threshold are adjusted. The historical correction sequence is used to perform a moving average to construct a long-term performance evaluation function J. The parameter combination is then optimized iteratively through gradient descent so that the actual false alarm rate F and the average unlocking delay D approach a preset equilibrium point during long-term operation.

[0024] In one embodiment, the control execution unit includes an electromagnetic latch, a reset spring, and a drive circuit. The electromagnetic latch is located at the front end of the lock body and extends to lock the switch cabinet door latch mechanism to achieve physical locking. After the unlocking conditions are met, the drive circuit is energized to cause the electromagnetic latch to retract instantaneously and release the door latch. The reset spring ensures that the electromagnetic latch automatically resets after power failure, ensuring safety in case of disability.

[0025] In one embodiment, the smart key and the lock body employ a unique insertion and removal structure to ensure that unauthorized smart keys cannot be inserted or operated; during the unlocking process, when the electromagnetic bolt retracts, the smart key is forcibly locked in the lock body by a mechanical limiting mechanism until the locking is completed and the bolt extends again before it can be pulled out.

[0026] In one embodiment, the smart key has a built-in independent battery compartment, which is powered by three AAA batteries connected in series. The smart key is also equipped with a low battery warning light for the battery compartment.

[0027] In one embodiment, the algorithm module can calibrate the criterion threshold in real time based on ambient temperature, noise level, and residual voltage attenuation rate. .

[0028] In one embodiment, the signal processing module has a multi-channel input interface, which can simultaneously receive three-phase voltage signals and achieve fault-tolerant operation based on redundancy criteria.

[0029] In one embodiment, when the unlocking conditions are met, if the mechanical key linkage mechanism detects an abnormal voltage fluctuation exceeding a set limit, it immediately stops the unlocking process and restores the locked state.

[0030] Compared with the prior art, the beneficial effects of this application are:

[0031] This invention addresses three core problems in existing switchgear anti-misoperation interlocking devices: low power supply reliability, poor discrimination accuracy, and insufficient operational closed-loop. It achieves highly reliable, adaptive, and robust anti-misoperation interlocking capabilities through multi-dimensional collaborative optimization of sensing, algorithms, power supply, and mechanical structure. Specific beneficial effects are as follows:

[0032] 1. Addressing power supply reliability issues at their root and eliminating the risk of power failure interlocking failure: The smart key uses a built-in independent battery compartment to power the unlocking action, completely eliminating reliance on the built-in power supply systems of switch cabinet PTs and UPSs. This avoids the problem of being unable to unlock due to PT voltage loss and also avoids the defects of short lifespan and poor reliability of UPS in high temperature and strong electromagnetic environments. At the same time, the smart key's low battery alarm mechanism can remind maintenance personnel to replace the battery in advance, further ensuring power supply reliability.

[0033] 2. Significantly improves the accuracy of live state identification and adapts to complex operating scenarios. It adopts an optical voltage sensing sensor to replace the traditional capacitive sensor, effectively suppressing stray electric field interference inside the switchgear. At the same time, it constructs a dynamic model of residual voltage attenuation, combined with an adaptive discrimination threshold and a two-layer parameter optimization mechanism. The criterion parameters can be calibrated in real time according to ambient temperature, noise level, and residual voltage attenuation rate. This avoids the risk of false locking caused by residual voltage interference and reduces the problem of unlocking failure caused by false live state identification. It is suitable for the operation and maintenance needs of switchgear of various voltage levels from 10kV to 110kV.

[0034] 3. Construct a closed-loop operation process to completely avoid human error. The mechanical key linkage mechanism is designed so that the smart key is forcibly locked in the lock body when the cabinet door is unlocked. The key can only be removed after the cabinet door is relocked and the bolt is extended. This physically forces the operator to complete the entire closed-loop process of "unlocking-operation-locking", avoiding safety hazards such as the cabinet door not being locked or the key being lost. At the same time, the instant locking mechanism for abnormal voltage fluctuations can immediately stop the operation in case of sudden power surge during the unlocking process, achieving secondary safety protection. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of this application 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 A schematic diagram of the unlocking and discrimination process of an intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm provided in this application;

[0037] Figure 2 A schematic diagram of the system structure of an intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm provided in this application;

[0038] Figure 3 A schematic diagram of the optical voltage sensing sensor of an intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm provided in this application;

[0039] Figure 4 A schematic diagram of the appearance of an intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm provided in this application;

[0040] Figure 5 This application provides a schematic diagram of the adaptive discrimination algorithm for an intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application are described clearly and completely below. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are also within the scope of protection of this application.

[0042] See Figures 1 to 5As shown, this application provides an intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm, which includes an optical voltage sensing sensor, a signal processing module, an algorithm module, a control execution unit, and a mechanical key linkage mechanism connected in sequence.

[0043] The optical voltage sensing sensor is used to acquire the voltage signal of the three-phase conductors in the primary circuit and convert the sensed signal into a low-voltage photoelectric signal output; the signal processing module performs bandpass filtering, sampling and normalization processing on the low-voltage photoelectric signal to obtain the characteristic sequence of the three-phase voltage.

[0044] The algorithm module establishes a residual voltage attenuation model and generates an adaptive discrimination threshold. It dynamically adjusts the discrimination parameters in conjunction with a two-layer parameter optimization mechanism to determine the energized state of the primary circuit. If the voltage of any one of the three-phase voltage channels is higher than the discrimination threshold, a blocking signal is output. If all three-phase voltages are lower than the discrimination threshold and the duration exceeds the set delay, an unlocking signal is output.

[0045] The control execution unit is used to receive signals from the algorithm module. When an unlock signal is received and the smart key is inserted, the smart key's built-in battery compartment provides instantaneous power to drive the electromagnetic bolt to retract, releasing the cabinet door latch to unlock. When a locking signal is received, the electromagnetic bolt remains extended to maintain locking. The mechanical key linkage mechanism forcibly locks the key in the lock body during the bolt retraction until the bolt retracts again, forming an operation closed loop.

[0046] Optionally, the optical voltage sensing sensor is based on the Pockels effect principle of electro-optic crystals, and its structure includes an electro-optic crystal, a polarizer, an analyzer, a quarter-wave plate, and a light source and photoelectric conversion module; the optical voltage sensing sensor is arranged near the high-voltage live parts inside the equipment and meets the safety insulation distance requirements.

[0047] In this embodiment, an optical voltage sensing sensor can be pre-embedded in enclosed equipment such as combined electrical appliances during the manufacturing stage. When a high-voltage charged body is under the influence of an electric field, the electro-optic crystal undergoes linear deformation. The light beam emitted by the LED light source is polarized by a polarizer and then decomposed into two polarized beams by the electro-optic crystal. After being processed by an analyzer and a quarter-wave plate, the phase difference is converted into an electrical signal by a photoelectric conversion module. By analyzing the phase difference between the two beams, the system can deduce the charging state of the primary circuit. Compared with traditional capacitive sensing sensors, this optical voltage sensing method has the advantages of strong directionality, good frequency response characteristics, and high bandwidth. It can also effectively suppress the interference of the complex electromagnetic environment inside high-voltage electrical equipment on voltage signal acquisition, thereby significantly improving the accuracy and reliability of electrical state determination.

[0048] Optionally, the algorithm module performs discrete sampling on the characteristic sequence of the acquired three-phase voltage and constructs a residual voltage attenuation model using the least squares method:

[0049] ; ①

[0050] In formula ① Let be the residual voltage at time t. Let τ be the initial voltage and τ be the voltage decay constant. The algorithm dynamically updates τ during execution to obtain the predicted curve. Furthermore, a noise disturbance term is introduced to form an adaptive criterion threshold:

[0051] ;②

[0052] In formula ② β is the standard deviation of the voltage signal within the sampling window, and β is the threshold weighting parameter; when the three-phase voltage Any channel exceeding the discrimination threshold When the primary circuit is determined to be energized, the lockout is maintained; when the three-phase voltage... All are below the discrimination threshold And the duration is not less than the delay threshold. When the time is right, it is determined to be in a safe, pressure-free state and the unlocking process is triggered.

[0053] Optionally, the algorithm module further introduces a two-layer parameter adaptive optimization mechanism based on the adaptive criterion threshold, which is used to optimize the threshold weight parameter β and the delay threshold. This balances locking security with unlocking sensitivity.

[0054] The first layer of fast adaptive correction: Within each sampling period, based on the real-time deviation between the actual false alarm rate F and the average unlocking delay D, the weight parameter β and the delay threshold are dynamically updated using a linear gain rule. The calculation method is as follows:

[0055] ;③

[0056] In formula ③, F ref This refers to the maximum allowable false alarm rate threshold of the system, used to measure the accuracy of live state detection. The actual false alarm rate F needs to be compared with this reference value to calculate the deviation. Dref refers to the system's expected unlocking response time benchmark. The average unlocking delay D needs to be compared with this reference value to ensure a balance between operational efficiency and safety. Both serve as the benchmark for the first layer of fast adaptive correction, calculated through... and The real-time deviation is dynamically adjusted by adjusting the weight parameter β and the delay threshold. This makes the system performance approach the preset target.

[0057] Based on the magnitude of the error, parameters are adjusted in real time using a linear gain correction rule to ensure that the system does not malfunction or fail under transient disturbances; the correction algorithm is shown below:

[0058] ④

[0059] In formula ④, , For learning rate, , This is the weighting factor.

[0060] The second layer of slow global optimization: At multiple sampling period time scales, the weight parameter β and the delay threshold are adjusted. The historical correction sequence is used to perform a moving average to construct a long-term performance evaluation function J. The parameter combination is then optimized iteratively through gradient descent so that the actual false alarm rate F and the average unlocking delay D approach a preset equilibrium point during long-term operation.

[0061] The long-term performance evaluation function J is constructed using the following algorithm:

[0062] ⑤

[0063] In formula ⑤, This is expressed as the average false alarm rate over a period of time. This represents the average unlocking delay over a period of time. , These are the weighting coefficients. Further gradient descent iterations are then employed, as shown in the algorithm below:

[0064] ⑥

[0065] In formula ⑥, , The global learning rate is used. This two-layer mechanism works synergistically, allowing the system to gradually approach the optimal parameter combination over long-term operation. This ensures both robustness to instantaneous disturbances and stability of the threshold under the influence of long-term factors such as environmental temperature changes and component aging.

[0066] It should be noted that when the operator inserts the key and presses the unlock button, the optical voltage sensing sensor quickly samples the three-phase voltage signal, and the signal processing module extracts the feature sequence of the three-phase voltage. The algorithm module then starts to judge the threshold. If any phase voltage exceeds the judgment threshold... If the voltage of any one phase exceeds the threshold, the system remains locked with the bolt extended. If all three phases are below the threshold, a delay period begins, typically 1–5 seconds, adjusted based on the response characteristics of the live indicator and the residual voltage decay pattern. During the delay, if any phase voltage exceeds the threshold again, the system immediately returns to the locked state. Once the delay is complete and the battery is fully charged, the smart key powers the lock mechanism, retracting the bolt to unlock.

[0067] After the operation is completed, when the smart key shuts off the power, the latch automatically pops out under the action of the return spring, and the cabinet door locks again. The entire process strictly follows the triple logical closed loop of "no power criterion first, manual confirmation required, and short-term power supply limited". It not only restores the inherent safety requirement of "power verification unlocking", but also prevents unexpected unlocking caused by battery depletion, poor contact or electromagnetic pulse through power self-check and power supply duration constraints. It realizes active anti-misoperation locking with human-machine collaboration and the integration of electrical, mechanical and information domains.

[0068] Optionally, the control execution unit includes an electromagnetic latch, a reset spring, and a drive circuit. The electromagnetic latch is located at the front end of the lock body and extends to lock the switch cabinet door latch mechanism to achieve physical locking. After the unlocking conditions are met, the drive circuit is energized to cause the electromagnetic latch to retract instantaneously and release the door latch. The reset spring ensures that the electromagnetic latch automatically resets after power failure, ensuring safety in case of disability.

[0069] Optionally, the smart key and the lock body adopt a unique plug-in structure to ensure that unauthorized smart keys cannot be inserted or operated.

[0070] Furthermore, to ensure the integrity and security of the operation process, the device is designed with a forced linkage between the smart key and the lock body: when the latch is retracted and the door is unlocked, the inserted smart key is mechanically locked inside the lock body, preventing the operator from removing it. The key can only be removed after the latch re-extends and re-locks the cabinet door. This mechanism ensures a closed-loop operation, effectively avoiding the risk of the cabinet door being in an "unlocked but not locked" state due to premature key removal, and preventing personnel from bypassing steps or leaving the site without authorization, thereby further enhancing the robustness of the anti-misoperation function.

[0071] Optionally, the smart key has a built-in independent battery compartment, powered by three AAA batteries connected in series. The smart key also features a low battery warning light. The three AAA batteries provide short-term energy to the electromagnetic bolt during unlocking, ensuring reliable operation. Compared to traditional solutions relying on voltage transformers for power, this design avoids the risk of the bolt locking permanently due to transformer voltage loss after a power outage. Furthermore, compared to UPS power supply, its external battery compartment layout is more suitable for the high-temperature, enclosed, and strong electromagnetic environment inside high-voltage cabinets, significantly extending the lifespan and reliability of the power supply system.

[0072] Optionally, the algorithm module can calibrate the criterion threshold in real time based on ambient temperature, noise level, and residual voltage attenuation rate. This ensures reliability under different operating conditions.

[0073] Optionally, the signal processing module has a multi-channel input interface, which can simultaneously receive three-phase voltage signals and achieve fault-tolerant operation based on redundancy criteria.

[0074] Optionally, when the unlocking conditions are met, if an abnormal voltage fluctuation is detected exceeding a set limit, the mechanical key linkage mechanism will immediately stop the unlocking process and restore the locked state, thereby achieving secondary safety protection.

[0075] It should be noted that the complete operation process is as follows:

[0076] S1. An optical voltage sensing sensor is used to acquire the voltage signal of the three-phase conductors in the primary circuit and convert the induced signal into a low-voltage photoelectric signal that can be collected;

[0077] S2. The signal processing module performs filtering, sampling, and normalization on the photoelectric signal to obtain the characteristic sequence of the three-phase voltage;

[0078] S3. The signal processing module performs filtering, sampling, and normalization on the photoelectric signal to obtain the characteristic sequence of the three-phase voltage;

[0079] S4. When the actual voltage is higher than the threshold, the primary circuit is determined to be energized and the lockout is maintained; when the actual voltage is lower than the threshold and continues to exceed the set delay, it is determined to be de-energized and the unlocking logic is triggered.

[0080] S5. A lightweight optimization method is introduced to optimize parameters online, with "minimizing the false alarm rate" as the objective function, to achieve adaptive adjustment of the threshold.

[0081] S6. When the unlocking conditions are met, the control execution module uses instantaneous electrical energy provided by the smart key to drive the electromagnetic bolt to retract, releasing the cabinet door latch and unlocking the cabinet; when any condition is not met, the bolt remains extended to maintain the lock.

[0082] S7. The mechanical key linkage mechanism forcibly locks the key in the lock body during the retraction of the bolt until the operation is completed and the bolt extends again, thus ensuring a closed-loop operation.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. An intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm, characterized in that: It includes an optical voltage sensing sensor, a signal processing module, an algorithm module, a control execution unit, and a mechanical key linkage mechanism connected in sequence; The optical voltage sensing sensor is used to acquire the voltage signal of the three-phase conductors in the primary circuit and convert the sensed signal into a low-voltage photoelectric signal output; the signal processing module performs bandpass filtering, sampling and normalization processing on the low-voltage photoelectric signal to obtain the characteristic sequence of the three-phase voltage. The algorithm module establishes a residual voltage attenuation model and generates an adaptive discrimination threshold. It dynamically adjusts the discrimination parameters in conjunction with a two-layer parameter optimization mechanism to determine the energized state of the primary circuit. If the voltage of any one of the three-phase voltage channels is higher than the discrimination threshold, a blocking signal is output. If all three-phase voltages are lower than the discrimination threshold and the duration exceeds the set delay, an unlocking signal is output. The control execution unit is used to receive signals from the algorithm module. When an unlock signal is received and the smart key is inserted, the smart key's built-in battery compartment provides instantaneous power to drive the electromagnetic bolt to retract, releasing the cabinet door latch to unlock. When a locking signal is received, the electromagnetic bolt remains extended to maintain locking. The mechanical key linkage mechanism forcibly locks the key in the lock body during the retraction of the bolt, and the key can only be removed after the bolt extends again, forming an operation closed loop.

2. The intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm according to claim 1, characterized in that: The optical voltage sensing sensor is based on the Pockels effect principle of electro-optic crystals. Its structure includes an electro-optic crystal, a polarizer, an analyzer, a quarter-wave plate, and a light source and photoelectric conversion module. The optical voltage sensing sensor is arranged near the high-voltage charged body inside the equipment and meets the safety insulation distance requirements.

3. The intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm according to claim 1, characterized in that: The algorithm module performs discrete sampling on the characteristic sequence of the acquired three-phase voltage and constructs a residual voltage attenuation model using the least squares method: ；① In formula ① Let be the residual voltage at time t. Let τ be the initial voltage and τ be the voltage natural decay constant; the algorithm dynamically updates τ to obtain the predicted curve. Furthermore, a noise disturbance term is introduced to form an adaptive criterion threshold: ；② In formula ② β is the standard deviation of the voltage signal within the sampling window, and β is the weighting parameter; when the three-phase voltage If any channel is higher than the specified value, the primary circuit is considered energized and the circuit is kept blocked; when the three-phase voltage is higher... All are below the discrimination threshold And the duration is not less than the delay threshold. When the time is right, it is determined to be in a safe, pressure-free state and the unlocking process is triggered.

4. The intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm according to claim 3, characterized in that: Based on the adaptive criterion threshold, the algorithm module further introduces a two-layer parameter adaptive optimization mechanism: The first layer of fast adaptive correction: Within each sampling period, based on the real-time deviation between the actual false alarm rate F and the average unlocking delay D, the weight parameter β and the delay threshold are dynamically updated using a linear gain rule. ; The second layer of slow global optimization: At multiple sampling period time scales, the weight parameter β and the delay threshold are adjusted. The historical correction sequence is used to perform a moving average to construct a long-term performance evaluation function J. The parameter combination is then optimized iteratively through gradient descent so that the actual false alarm rate F and the average unlocking delay D approach a preset equilibrium point during long-term operation.

5. The intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm according to claim 1, characterized in that: The control execution unit includes an electromagnetic latch, a reset spring, and a drive circuit. The electromagnetic latch is located at the front end of the lock body. When it extends, it locks the switch cabinet door latch mechanism to achieve physical locking. After the unlocking conditions are met, the drive circuit is energized to make the electromagnetic latch retract instantly and release the door latch. The reset spring ensures that the electromagnetic latch automatically resets after power failure, ensuring safety in the event of a power outage.

6. The intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm according to claim 1, characterized in that: The smart key and the lock body adopt a unique plug-in structure to ensure that unauthorized smart keys cannot be inserted or operated; during the unlocking process, when the electromagnetic bolt retracts, the smart key is forcibly locked in the lock body by the mechanical limiting mechanism until the locking is completed and the bolt extends again before it can be pulled out.

7. The intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm according to claim 1, characterized in that: The smart key has a built-in independent battery compartment, which is powered by three AAA batteries connected in series. The smart key is also equipped with a low battery warning light for the battery compartment.

8. The intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm according to claim 1, characterized in that: The algorithm module can calibrate the criterion threshold in real time based on ambient temperature, noise level, and residual voltage attenuation rate. .

9. The intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm according to claim 1, characterized in that: The signal processing module has a multi-channel input interface, which can simultaneously receive three-phase voltage signals and achieve fault-tolerant operation based on redundancy criteria.

10. The intelligent switchgear anti-misoperation interlocking device based on an adaptive discrimination algorithm according to claim 1, characterized in that: When the unlocking conditions are met, if the mechanical key linkage mechanism detects an abnormal voltage fluctuation exceeding a set limit, it will immediately stop the unlocking process and restore the locked state.

Images