Intelligent and orderly control algorithm and device for ac transfer joint with real-time state perception

Abstract

The application discloses an alternating current adapter intelligent orderly control algorithm and device with real-time state sensing, belongs to the technical field of adapter control, and comprises the following steps: collecting multi-source heterogeneous data of an alternating current adapter during operation through electrical sensors, mechanical vibration sensors and environmental sensors synchronously; adopting a sparrow algorithm to optimize hyperparameters of a Gaussian process regression model, constructing a real-time state evaluation model, inputting the multi-source heterogeneous data, and outputting adapter health scores; identifying the type of an access load based on a device fingerprint library, combining the health scores, real-time electricity price signals and load battery states, and generating a dynamic power supply priority sequence; distributing output power according to the priority sequence, and sending scheduling instructions to an execution unit through a 60GHz millimeter wave link. Through synchronous collection of electrical, mechanical and environmental sensors and optimization of the model by the sparrow algorithm, combined with dynamic scheduling of the device fingerprint and millimeter wave instructions, the single dimension misjudgment problem is solved, and the overload risk is reduced.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of adapter control, in particular to an alternating current adapter intelligent orderly control algorithm and device with real-time state sensing. BACKGROUND

[0002] According to the search, an application patent with the publication number CN117807487A discloses an intelligent sensing method and device for cable joint state, which comprises the following steps: collecting multi-source heterogeneous data of the running state of the cable joint through multiple sensors; after the multi-source heterogeneous data is preliminarily screened and the missing data and noise data are processed, the multi-source heterogeneous data is stored in a data warehouse; the multi-source heterogeneous data in the data warehouse is standardized and whitened to convert into feature vector data; a Gaussian process regression cable joint state evaluation model is constructed, the feature vector data is input into the Gaussian process regression cable joint state evaluation model, and the target state of the cable joint is obtained. The application fuses and cooperatively senses the multi-source heterogeneous data of the cable joint state through the Gaussian process regression cable joint state evaluation model, accurately predicts the running state of the cable joint, and avoids misjudgment of single data information.

[0003] In the above-mentioned disclosed technology, the running state of the cable joint is only judged by monitoring the temperature and humidity, and the mechanical condition and electrical parameters are not associated, so that it is impossible to distinguish whether the temperature rise is caused by electrical problems such as overload or mechanical problems such as plug-in wear and tear; and the multi-device dynamic scheduling capability is lacking, and it is impossible to respond to the peak-valley electricity price signal. SUMMARY

[0004] In order to overcome the above-mentioned shortcomings of the prior art, the application provides an alternating current adapter intelligent orderly control algorithm and device with real-time state sensing, which aims to solve the technical problem in the prior art that the running state of the cable joint is only judged by monitoring the temperature and humidity, and the mechanical condition and electrical parameters are not associated, so that it is impossible to distinguish whether the temperature rise is caused by electrical problems or mechanical problems.

[0005] The technical solution adopted by the application to solve the technical problem is: an alternating current adapter intelligent orderly control algorithm with real-time state sensing, comprising:

[0006] Synchronously collecting multi-source heterogeneous data of the running state of the alternating current adapter by electrical sensors, mechanical vibration sensors and environmental sensors;

[0007] Optimizing the hyperparameters of the Gaussian process regression model by using the sparrow algorithm, constructing a real-time state evaluation model, inputting the multi-source heterogeneous data, and outputting the adapter health score;

[0008] Based on the device fingerprint library, the type of the access load is identified, the health score, the real-time electricity price signal and the load battery state are combined, and a dynamic power supply priority sequence is generated.

[0009] The output power is allocated in a priority sequence, and a scheduling instruction is sent to the execution unit.

[0010] As a further improvement of the present application: the output power is allocated in a priority sequence, and a scheduling instruction is sent to the execution unit, including:

[0011] The output power is allocated in a priority sequence, and a scheduling instruction is sent to the execution unit through a 60GHz millimeter wave link.

[0012] As a further improvement of the present application: the 60GHz millimeter wave link uses HMC6300 transmitter and HMC6301 receiver chipsets, with a transmission delay of less than 200 microseconds.

[0013] As a further improvement of the present application: the electrical sensor includes a Hall current sensor and an impedance spectrum analysis circuit. The impedance spectrum analysis circuit detects 0.1 μm oxidation.

[0014] As a further improvement of the present application: the mechanical vibration sensor is a three-dimensional micro-electromechanical vibration sensor. The three-dimensional micro-electromechanical vibration sensor captures 200Hz loose signals.

[0015] As a further improvement of the present application: the environmental sensor includes a nano-zinc oxide temperature sensing array and a humidity sensor. The nano-zinc oxide temperature sensing array achieves 0.1℃ resolution, achieving micron-level failure warning.

[0016] As a further improvement of the present application: the covariance function of the Gaussian process regression model is defined as a time-varying form:

[0017] ;

[0018] wherein, is a device aging factor, with a value range of 0.01-0.05; is a signal variance, is a length scale, , respectively represent two different times. The device aging factor increases with the increase of the cumulative plug-in times of the adapter, and the device aging factor increases with the increase of the plug-in times (0.01 to 0.05) and introduces a time-varying term, greatly reducing the life prediction error.

[0019] As a further improvement of the present application: the generation formula of the dynamic power supply priority sequence is:

[0020] ;

[0021] wherein, for dynamic power supply priority, for device type identification, for device battery level, time period identification, for battery state weight coefficient, for device type weight coefficient, for tariff period weight coefficient, medical device , industrial device , consumer electronics ; peak power period , valley power period .

[0022] As a further improvement of the present application: the calculation method of the device aging factor is:

[0023] ;

[0024] wherein is the cumulative plug-in frequency of the adapter, , The unit plug-in coefficient realizes the linear quantization of the device aging factor, and the accurate replacement alarm is triggered by ten thousand plug-ins.

[0025] The application also provides a device for applying an alternating current adapter intelligent orderly control algorithm with real-time state sensing, which comprises a stacked hardware module and a self-powered circuit.

[0026] The stacked hardware module is integrated in the interior of the adapter shell and comprises a sensor array, an edge computing unit and a millimeter wave communication module.

[0027] The self-powered circuit is powered by a CT coil from an alternating magnetic field and is connected to a solid-state lithium capacitor backup power supply.

[0028] Compared with the prior art, the application has the following beneficial effects:

[0029] 1. In the application, three types of data, i.e., electrical parameters, mechanical vibration parameters and environmental parameters, are synchronously collected, the sparrow algorithm is used to optimize the hyperparameters of the Gaussian process regression model, a comprehensive state evaluation model is constructed, the adapter health score is output, the overload current is monitored by the Hall current sensor, the contact surface oxidation is detected by the impedance spectrum analysis circuit, the zero-to-five-kilohertz vibration frequency spectrum is collected by the three-dimensional micro-electromechanical vibration sensor, the plug-in wear and mechanical looseness are identified, the environmental corrosion risk is evaluated by the nanometer zinc oxide temperature sensing array and the humidity sensor, and the sparrow algorithm dynamically adjusts the hyperparameters of the Gaussian process regression model, thereby improving the model accuracy.

[0030] 2. In the application, the load type is automatically identified based on the device fingerprint library, the dynamic power supply priority sequence is generated in combination with the adapter health score, the load battery level and the real-time peak-valley power price signal.

[0031] 3、The application adopts a sixty gigahertz millimeter wave communication link to transmit control instructions, uses a HMC6300 transmitter and a HMC6301 receiver chip set, the transmission delay is less than two hundred microseconds, the non-contact transmission avoids signal attenuation caused by mechanical wear and tear, and has the anti-yaw ability, and supports stable transmission under a positive or negative fifteen degree mechanical deviation. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1 The flow chart is for the embodiment one of the application.

[0033] Figure 2 The flow chart is for step one of the embodiment one of the application.

[0034] Figure 3 The flow chart is for step two of the embodiment one of the application.

[0035] Figure 4 The flow chart is for step three of the embodiment one of the application.

[0036] Figure 5 The flow chart is for step four of the embodiment one of the application.

[0037] Figure 6 The structural block diagram is for the embodiment two in the application. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical scheme and advantages of the application clearer, the technical scheme of the application will be described clearly and completely below by combining the specific embodiments of the application with the corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application. In addition, the terms “include” and “have” and any variations thereof are intended to cover the non-exclusive inclusion, for example, the process, method, system, product or equipment including a series of steps or units does not have to be limited to the clearly listed steps or units, but can include other steps or units that are not clearly listed or inherent to the process, method, product or equipment.

[0039] Please refer to Figures 1-6 , the alternating current adapter intelligent orderly control algorithm with real-time state perception, comprising:

[0040] The multi-source heterogeneous data during the alternating current adapter operation is synchronously collected through the electrical sensor, the mechanical vibration sensor and the environmental sensor;

[0041] The sparrow algorithm is used to optimize the hyperparameters of the Gaussian process regression model, a real-time state evaluation model is constructed, multi-source heterogeneous data is input, and a health score of the adapter is output.

[0042] Based on the device fingerprint library, the type of the access load is identified, the health score, the real-time electricity price signal and the load battery state are combined, and a dynamic power supply priority sequence is generated.

[0043] The output power is allocated according to the priority sequence, and the scheduling instruction is sent to the execution unit through the 60GHz millimeter wave link.

[0044] Through the synchronous collection of electrical, mechanical and environmental sensors and the sparrow algorithm optimization model, combined with the dynamic scheduling of the device fingerprint and the millimeter wave instruction, the single dimension misjudgment problem is solved, and the overload risk is reduced.

[0045] In the present application, based on the device fingerprint library, the type of the load is automatically identified, combined with the health score of the adapter, the load battery capacity and the real-time peak-valley electricity price signal, a dynamic power supply priority sequence is generated, which avoids the problem of unordered power supply and the risk of line overload caused by the lack of multi-device power supply scheduling function in the prior art (such as the invention patent with publication number CN117807487A).

[0046] In some embodiments, the electrical sensor includes a Hall current sensor and an impedance spectrum analysis circuit, the mechanical vibration sensor is a three-dimensional micro-electromechanical vibration sensor, the environmental sensor includes a nanometer zinc oxide temperature sensing array and a humidity sensor, the impedance spectrum analysis circuit detects 0.1 μm oxidation, the three-dimensional micro-electromechanical vibration sensor captures a 200 Hz loose signal, and the nanometer zinc oxide temperature sensing array realizes 0.1 ℃ resolution, achieving micron-level fault early warning.

[0047] In some embodiments, the covariance function of the Gaussian process regression model is defined as a time-varying form:

[0048] ;

[0049] wherein is a device aging factor, the value range is 0.01-0.05, and the value increases with the increase of the cumulative plug-in times of the adapter, the device aging factor increases with the increase of the plug-in times (0.01 to 0.05) and introduces a time-varying term, so that the life prediction error is greatly reduced; is a signal variance, is a length scale, , respectively represent two different times.

[0050] In some embodiments, the generation formula of the dynamic power supply priority sequence is:

[0051] ;

[0052] wherein, is a dynamic power supply priority, is a device type identifier, is a device battery level, is a time period identifier, is a battery state weight coefficient, is a device type weight coefficient, is a power price period weight coefficient, medical device , industrial device , consumer electronics ; peak power period , valley power period .

[0053] medical weight ensure that the probability of power failure is small, peak power cut load, realize key load power supply and power grid response.

[0054] In some embodiments, the 60GHz millimeter wave link uses HMC6300 transmitter and HMC6301 receiver chipsets, the transmission delay is less than 200 microseconds, and the HMC6300 and HMC6301 chipsets support ±15° deflection. The delay of less than 200 microseconds makes the protection action speed up 50 times.

[0055] In the present application, the control instruction is transmitted by a sixty gigahertz millimeter wave communication link, using HMC6300 transmitter and HMC6301 receiver chipsets, with a transmission delay of less than two hundred microseconds. By non-contact transmission, signal attenuation caused by mechanical wear is avoided, and the anti-deflection capability is supported, stable transmission under ±15 mechanical deviation is supported, and the problem that in the prior art (such as the invention patent with publication number CN117807487A) depends on wired slip ring transmission, and once the slip ring is worn, the bit error rate exceeds one millionth, and the control delay is greater than ten milliseconds, resulting in the problem of unable to respond to faults in real time.

[0056] In some embodiments, the calculation method of the device aging factor is:

[0057] ;

[0058] wherein is the cumulative number of plug-in times of the adapter, , The unit plug-in coefficient realizes linear quantization of the device aging factor, and accurately triggers the replacement alarm after ten thousand plug-in times.

[0059] In some embodiments, the device applied to the intelligent orderly control method of the AC adapter with real-time state perception includes at least a stacked hardware module and a self-powered circuit; the stacked hardware module is integrated in the interior of the adapter shell and contains a sensor array, an edge computing unit and a millimeter wave communication module; the self-powered circuit is powered by the CT coil from the AC magnetic field and is connected to the solid-state lithium capacitor backup power supply.

[0060] The stacked module diameter is ≤25mm, greatly reducing the volume, the CT coil power is 3-5W, and the solid-state lithium capacitor maintains 10ms power failure endurance.

[0061] Case 1:

[0062] Please refer to Figures 1-5 The embodiment case provides the following technical solutions: an AC adapter intelligent orderly control algorithm with real-time state perception, comprising:

[0063] Step 1: Synchronously collect multi-source heterogeneous data of the AC adapter during operation through electrical sensors, mechanical vibration sensors and environmental sensors;

[0064] Step 2: Optimize the hyperparameters of the Gaussian process regression model using the Sparrow algorithm, build a real-time state evaluation model, input multi-source heterogeneous data, and output adapter health scores;

[0065] Step 3: Identify the type of access load based on the device fingerprint library, combine the health score, real-time electricity price signal and load battery state to generate a dynamic power supply priority sequence;

[0066] Step 4: Distribute the output power according to the priority sequence and send scheduling instructions to the execution unit through the 60GHz millimeter wave link.

[0067] Synchronously collect through electrical, mechanical and environmental sensors and optimize the model using the Sparrow algorithm, combine dynamic scheduling with device fingerprints and millimeter wave instructions to solve single-dimensional misjudgment problems and reduce the risk of overload.

[0068] The electrical sensor includes a Hall current sensor and an impedance spectrum analysis circuit, the mechanical vibration sensor is a three-dimensional micro-electromechanical vibration sensor, the environmental sensor includes a nanometer zinc oxide temperature sensing array and a humidity sensor, the impedance spectrum analysis circuit detects 0.1μm oxidation, the three-dimensional micro-electromechanical vibration sensor captures 200Hz loose signal, the nanometer zinc oxide temperature sensing array realizes 0.1℃ resolution, and micron-level fault early warning is achieved.

[0069] The covariance function of the Gaussian process regression model is defined as a time-varying form:

[0070] ;

[0071] Wherein The device aging factor is 0.01-0.05, and the value increases with the increase of the cumulative plug-in number of the adapter, and the device aging factor is With the increase of the plug-in number (0.01 to 0.05) and the introduction of the time-varying term, the life prediction error is greatly reduced. The signal variance is The length scale is 、 Respectively represent two different times.

[0072] The generation formula of the dynamic power supply priority sequence is:

[0073] ;

[0074] Wherein, The dynamic power supply priority is The device type identifier is The device battery capacity is The time period identifier is The battery state weight coefficient is The device type weight coefficient is The electricity price period weight coefficient is Medical equipment Industrial equipment Consumer electronics Peak electricity period Valley electricity period

[0075] Medical weight Ensure that the power outage probability is small, and the peak electricity Reduce the load, realize the power supply of key load and the response of power grid.

[0076] The 60GHz millimeter wave link adopts HMC6300 transmitter and HMC6301 receiver chip set, and the transmission delay is less than 200 microseconds. The HMC6300 and HMC6301 chip set supports ± 15° deflection, and the delay is less than 200 microseconds, which makes the protection action speed up 50 times.

[0077] The calculation method of the device aging factor is:

[0078] ;

[0079] Wherein The cumulative plug-in number of the adapter is , The unit plug-in coefficient realizes the linear quantization of the device aging factor, and the accurate replacement alarm is triggered for 10,000 times of plug-in.

[0080] Implementation case two:

[0081] Please refer to Figure 6, AC adapter intelligent orderly control device with real-time state perception, at least including stacked hardware module and self-powered circuit; stacked hardware module, integrated in the adapter shell, contains sensor array, edge computing unit and millimeter wave communication module; self-powered circuit, from AC magnetic field through CT coil induction power, and connects solid-state lithium capacitor backup power supply.

[0082] The stacked module diameter ≤25mm greatly reduces the volume, the CT coil takes power 3-5W, and the solid-state lithium capacitor maintains 10ms power failure endurance.

[0083] Case three:

[0084] In this implementation case, the staff integrates the AC adapter of PDU-SMART / 48 in the intelligent power distribution system of A3 cabinet in the IDC machine room of Suzhou according to the method disclosed in the application, the Hall current sensor (Allegro ACS72981, ±50A range) of which monitors the current of each port in real time, the impedance spectrum analysis circuit (AD5933 chip) scans the contact impedance at a frequency of 10 kHz; the three-dimensional MEMS vibration sensor (ST LIS2DW12) collects the 0-5 kHz spectrum, the nanometer zinc oxide temperature sensing array (precision ±0.1℃) and the Sensirion SHT40 humidity sensor work synchronously. When the No. 1 port accesses the medical image server (device fingerprint ID: MED_CT_004), the system detects an abnormality: the current suddenly increases to 42A (threshold 40A), and at the same time, the vibration spectrum appears a peak value (screw loosening feature) at 375Hz, and the temperature sensing array shows that the contact temperature rises to 8.2℃.

[0085] The sparrow algorithm (population size 50, iteration 100 times) dynamically optimizes the hyperparameters of the Gaussian process regression model, and outputs the health score after inputting the multi-source data: the current overload weight accounts for 60%, the mechanical loosening accounts for 30%, and the environmental temperature rise accounts for 10%, and the comprehensive score is reduced to 62 points (threshold 70 points to trigger alarm). At this time, the No. 3 port industrial AI training server (fingerprint ID: IND_GPU_12) battery power is reduced to 18%, and it is in the positive valley power period (23:00-07:00, μ=1). The dynamic power supply priority calculation is:

[0086] Medical equipment: Priority=3×3 (device class) +0.8×100 (power) +1×1 (valley power) =109.8;

[0087] Industrial equipment: Priority=2×2+0.18×100+1×1=23.8;

[0088] The system sends instructions through a 60GHz millimeter wave link (HMC6300 / HMC6301 chipset, delay 183us): limit the power of industrial equipment to 65%, and maintain full power operation of medical equipment.

[0089] The self-powered circuit takes 3.5W from the 230V bus through the CT coil, and the solid-state lithium capacitor maintains power supply during 10ms power grid flicker. The cumulative plug-in frequency of the adapter is N_cycle=12,387 times, the aging factor is gamma=0.01+2e-5x12387=0.2577, and the time-varying covariance function reduces the life prediction error from ±15% to ±6.2%. After 3.2 seconds of failure, the system completes overload port isolation and pushes the alarm to the operation and maintenance platform, which is 11 times faster than the traditional solution.

[0090] The working principle of the application: through the synchronous acquisition of multi-source heterogeneous data of the AC adapter during operation by electrical sensors, mechanical vibration sensors and environmental sensors, the Hall current sensor and impedance spectrum analysis circuit monitor electrical overload and contact surface oxidation, the three-dimensional micro-electromechanical vibration sensor captures mechanical loosening signals in the frequency range of zero to five kilohertz, and the nanometer zinc oxide temperature sensing array and humidity sensor evaluates environmental risks; then the sparrow algorithm is used to dynamically optimize the hyperparameters of the Gaussian process regression model, and a real-time state evaluation model is constructed, and through the introduction of device aging factor and time-varying covariance function increasing with the number of plug-in times, a high-precision health score is output; based on the device fingerprint library to identify three types of load types of medical devices, industrial devices and consumer electronics, combined with the health score, real-time peak-valley electricity price signal and load battery state, a dynamic power supply priority formula is generated according to the medical weight value three, industrial weight value two, consumer weight value one, and peak electricity negative one, valley electricity positive one, and a scheduling sequence is generated; finally, through the 60GHz millimeter wave link, the HMC6300 transmitter and HMC6301 receiver chipsets are used to send instructions to the execution unit within a transmission delay of less than 200us, and the stacked hardware module integrates all components with a diameter of 25mm, and the current transformer coil is inductively powered to maintain 3-5W power supply, and the solid-state lithium capacitor provides 10ms backup power, realizing the whole process closed-loop control from multi-modal perception to intelligent scheduling.

[0091] The main functions of the application are:

[0092] In the application, three types of data, electrical parameters, mechanical vibration parameters and environmental parameters, are synchronously collected, the sparrow algorithm is used to optimize the hyperparameters of the Gaussian process regression model, a comprehensive state evaluation model is constructed, the health score of the adapter is output, the overload current is monitored by the Hall current sensor, the contact surface oxidation is detected by the impedance spectrum analysis circuit, the three-dimensional micro-electromechanical vibration sensor collects the vibration frequency spectrum from zero to five kilohertz, identifies the plug-in wear and mechanical looseness, the nanometer zinc oxide temperature sensing array and the humidity sensor evaluate the environmental corrosion risk, the sparrow algorithm dynamically adjusts the hyperparameters of the Gaussian process regression model, and the model precision is improved. Based on the device fingerprint library, the load type is automatically identified, the health score of the adapter, the battery power of the load, and the real-time peak-valley electricity price signal are combined to generate a dynamic power supply priority sequence. Sixty gigahertz millimeter wave communication links are used to transmit control instructions, HMC6300 transmitter and HMC6301 receiver chipsets are used, the transmission delay is less than 200 microseconds, non-contact transmission is used to avoid signal attenuation caused by mechanical wear, and has anti-yaw capability, supports stable transmission under positive and negative fifteen degree mechanical deviation.

[0093] The functions described herein can be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions can be stored on or transferred over as one or more instructions or code on a computer-readable medium. Other examples and implementations are within the scope and spirit of the disclosure and appended claims. For example, due to the nature of software, functions described above can be implemented using software executed by a processor, hardware, firmware, hardwiring, or combinations of any of these. Also, each of the functions can be implemented as a separate process or combined as a single process. Further, the functions can be implemented at least in part outside an operating system routine.

[0094] In several embodiments provided in the present application, it should be understood that the disclosed technical content can be implemented by other ways. Among them, the above-mentioned device embodiments are only schematic, for example, the division of the units can be a logical function division, and actual implementation can have another division way, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or modules shown or discussed can be indirect coupling or communication connection through some interfaces, units or modules, which can be electrical or other forms.

[0095] The units described as separate components can or can not be physically separated, and the components of the control device can or can not be physical units, that is, they can be located in one place or distributed on multiple units. Part or all of the units can be selected to achieve the purpose of the embodiment scheme according to actual needs.

[0096] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the entire or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0097] The above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and the specification of the present application.

Claims

1. An AC tap changer intelligent sequencing control algorithm with real-time condition awareness, characterized by: The application relates to a power supply device for an alternating current adapter, and belongs to the field of power supply devices. Synchronous acquisition of multi-source heterogeneous data during alternating current adapter operation through an electrical sensor, a mechanical vibration sensor and an environmental sensor; A sparrow algorithm is used to optimize the hyperparameters of a Gaussian process regression model, a real-time state evaluation model is constructed, multi-source heterogeneous data is input, and an adapter health score is output; Based on a device fingerprint library, the type of an access load is identified, a dynamic power supply priority sequence is generated by combining the health score, a real-time electricity price signal and a load battery state, and the like; Output power is distributed according to the priority sequence, and scheduling instructions are sent to an execution unit; The covariance function of the Gaussian process regression model is defined in a time-varying form: where γ is the device aging factor, with a value range of 0.01-0.05; σ 2 is the signal variance, l is the length scale, and t, t' represent two different time instants; The generation formula of the dynamic power supply priority sequence is: Priority = alpha * DeviceClass + beta * BatteyLevel + mu * TimeSlot Wherein, Priority is the dynamic power supply priority, DeviceClass is the device type identifier, BatteyLevel is the device battery capacity, TimeSlot is the time period identifier, beta is the battery state weight coefficient, alpha is the device type weight coefficient, and mu is the electricity price time period weight coefficient; the medical device alpha = 3, the industrial device alpha = 2, and the consumer electronics alpha = 1; the peak electricity period mu = -1, and the valley electricity period mu = 1; The calculation method of the device aging factor is: Y = k0+ k1- N cycle where N cycle is the cumulative number of plug-in and plug-out times of the adapter, k0=0.01, k1=2x10 -5 .

2. The AC tap changer intelligent sequencing control algorithm with real-time state awareness of claim 1, wherein: The output power is distributed according to the priority sequence, and the scheduling instructions are sent to the execution unit, and the method comprises the following steps: The output power is distributed according to the priority sequence, and the scheduling instructions are sent to the execution unit through a 60GHz millimeter wave link.

3. The AC tap changer intelligent sequencing control algorithm with real-time state awareness of claim 2, wherein: The 60GHz millimeter wave link adopts an HMC6300 transmitter and an HMC6301 receiver chip set.

4. The AC tap changer intelligent sequencing control algorithm with real-time state awareness of claim 1, wherein: The electrical sensor comprises a Hall current sensor and an impedance spectrum analysis circuit.

5. The AC tap changer intelligent sequencing control algorithm with real-time state awareness of claim 4, wherein: The mechanical vibration sensor is a three-dimensional micro-electromechanical vibration sensor.

6. The AC tap changer intelligent sequencing control algorithm with real-time state awareness of claim 5, wherein: The environmental sensor comprises a nano zinc oxide temperature sensing array and a humidity sensor.

7. The apparatus for applying the method of intelligent, sequenced control of AC transfer switches with real-time condition awareness of any of claims 1-6, characterized in that: A stacked hardware module and a self-powered circuit are included. The stacked hardware module is integrated in the adapter shell and comprises a sensor array, an edge computing unit and a millimeter wave communication module. The self-powered circuit is powered by a CT coil through an alternating current magnetic field and is connected to a solid-state lithium capacitor backup power supply.

Images