Self-adaptive switching system and method for power supply and distribution

Abstract

The invention discloses a self-adaptive switching system and method for power supply and distribution, and the system comprises a data collection module, an analysis module, a modeling module, a decision module, an execution module and an early warning module, and the data collection module generates a real-time monitoring signal; the analysis module performs dynamic weighted evaluation according to the real-time monitoring signal and generates an evaluation signal; the modeling module constructs a virtual model of the power supply system and generates an analog signal; the decision signal generates a control signal containing a path selection instruction; the execution module controls the semiconductor device to switch the physical connection state of the power supply according to the control signal and generates a switch feedback signal; and the early warning module forms an equipment health assessment result according to the switch feedback signal and an abnormal arc signal when the execution module switches the connection state and transmits the result to the decision module. The self-adaptive switching system for power supply and distribution can solve the problem that a traditional power supply and distribution switching system depends on a single threshold to judge the state of a power supply and cannot comprehensively evaluate the quality of electric energy.

Description

Technical Field

[0001] The present invention relates to the field of power system automation control, and in particular to an adaptive switching system and method for power supply and distribution. Background Art

[0002] Traditional power supply and distribution switching technology mainly relies on dual-power automatic switching devices (ATS), which achieve power conversion based on a simple criterion of voltage presence or absence, but do not consider power quality issues such as harmonic distortion and frequency fluctuations. As a result, precision equipment may still be damaged due to substandard power supply quality after switching.

[0003] As the proportion of new energy sources connected to the grid increases, the demand for coordinated switching of multiple power sources in microgrid scenarios becomes prominent. Existing technologies use fixed threshold strategies, which cannot adapt to changes in grid parameters caused by fluctuations in wind, solar, and storage output, and often cause power oscillations or overload tripping. In the industrial field, the back electromotive force generated when switching large-capacity inductive loads will cause transient current shocks. Traditional mechanical contactors have limited arc extinguishing capabilities, and frequent operations can easily cause contact sticking failures. Existing monitoring systems mostly use remote Internet of Things architectures. Signal transmission delays cause switching command lags, making it difficult to meet millisecond-level real-time requirements. In terms of fault warning, conventional methods rely on threshold alarm mechanisms and can only provide passive protection after a fault occurs. They lack the ability to predict the aging status of switching devices. In addition, the multi-power supply system of commercial buildings lacks an economic optimization model and cannot dynamically select the power supply source based on time-of-use electricity prices, resulting in wasted energy costs.

[0004] While digital twin technology has been applied in power grid simulation in recent years, existing solutions lack closed-loop verification of the virtual and real systems. Model accuracy gradually decreases with equipment aging, leading to cumulative deviations between simulation results and actual conditions. The application of edge computing technology in power systems is largely limited to data collection, and a policy optimization mechanism that deeply collaborates with the cloud has yet to be established. These technical deficiencies collectively result in significant deficiencies in the safety, cost-effectiveness, and intelligence of existing power supply and distribution switching systems. Summary of the Invention

[0005] In view of the shortcomings of the above-mentioned prior art, the purpose of the present invention is to provide an adaptive switching system and method for power supply and distribution, which is used to solve the problem that the traditional power supply and distribution switching system relies on a single threshold to judge the power supply status and cannot comprehensively evaluate the power quality. The present invention collects voltage, current and load impedance signals in real time through multiple types of sensors, constructs an analysis module including a dynamic weight evaluation model, and comprehensively calculates power quality indicators; uses digital twin technology to create a virtual power supply system, simulates transient responses under different switching paths, and screens the optimal solution through a reinforcement learning algorithm; adopts a hybrid actuator with semiconductor switches and mechanical contactors in parallel, and preferentially completes fast switching through solid-state devices to reduce arc shock; establishes a closed-loop verification mechanism, compares actual switching parameters with simulation results, and dynamically corrects model accuracy; introduces a high-frequency arc feature library to identify abnormal signals, and triggers preventive switching in combination with equipment life prediction; and achieves a dynamic balance between power supply cost and safety through edge computing and cloud collaborative optimization strategies.

[0006] The present invention provides an adaptive switching system for power supply and distribution, comprising: Data acquisition module, the data acquisition module collects voltage waveform signals, current phase signals and load impedance spectrum signals, and generates real-time monitoring signals; Analysis module: The analysis module performs dynamic weighted evaluation on voltage stability, harmonic distortion rate, frequency deviation, load matching and power supply cost based on real-time monitoring signals and generates evaluation signals; Modeling module, the modeling module builds a virtual model of the power supply system and inputs the evaluation signal into the virtual model to generate a simulation signal including impact prediction and recovery time; A decision module, which processes analog signals through algorithms and generates control signals containing path selection instructions; An execution module, which controls the semiconductor device to switch the physical connection state of the power supply according to the control signal and generates a switch feedback signal at the same time; The early warning module forms an equipment health assessment result based on the switch feedback signal and the abnormal arc signal when the execution module switches the connection state and transmits it to the decision module.

[0007] In one embodiment of the present invention, the data acquisition module includes a wideband impedance detection unit, which obtains a load impedance spectrum signal by injecting a test signal of a specific frequency and works in conjunction with a load type identification unit. When a change in the load type is detected, the sampling frequency range of the sensor array is dynamically adjusted. The analysis results of the load impedance spectrum signal are synchronously transmitted to the modeling module for correcting the equivalent impedance parameters of the load end in the virtual model. A signal compensation mechanism is established between the wideband impedance detection unit and the current phase signal acquisition channel to eliminate interference caused by the test signal injection on the original current phase measurement.

[0008] In one embodiment of the present invention, the weight coefficient of the dynamic weighted evaluation in the analysis module is automatically adjusted according to the power supply scenario mode. When the system is in the economic priority mode, the weight ratio of the power supply cost indicator is increased, and in the safety priority mode, the comprehensive weight of voltage stability and harmonic distortion rate is strengthened. The weight adjustment process realizes pattern recognition through feature extraction of historical operation data, and keeps linkage with the virtual model parameter update of the modeling module to form a closed-loop optimization mechanism of the evaluation strategy.

[0009] In one embodiment of the present invention, the virtual model of the modeling module includes a dynamic reconstruction function for the power supply network topology. When a distributed power source is detected or a load node is changed, the virtual model automatically generates a new equivalent circuit model and verifies its convergence. A time domain synchronization channel is established between the virtual model and the physical system, and the time scale deviation caused by the signal transmission delay is compensated by an interpolation algorithm to ensure that the impact prediction results are aligned with the time domain characteristics of the actual transient process.

[0010] In one embodiment of the present invention, the algorithm of the decision module includes a multi-objective optimization engine, which quantifies the power supply reliability index, equipment loss cost and switching operation risk into multi-dimensional constraints, screens the non-inferior solution set through Pareto front analysis, and introduces a fuzzy comprehensive evaluation model to prioritize the candidate solutions. The output results of the multi-objective optimization engine are cross-validated with the equipment health assessment results of the early warning module to form a quantitative scoring mechanism for decision credibility.

[0011] In one embodiment of the present invention, the execution module includes a fast semiconductor switch group and a mechanical contactor group arranged in parallel, wherein the semiconductor switch group is configured to preferentially respond to high-frequency switching instructions and undertake transient current transfer, and the mechanical contactor group automatically closes after the semiconductor switch completes conduction to maintain a steady-state current carrying capacity. An action timing coordination unit is set between the semiconductor switch group and the mechanical contactor group to ensure that the opening and closing operations of the two groups of devices meet the preset phase synchronization requirements.

[0012] In one embodiment of the present invention, the early warning module includes a high-frequency signal feature library, which stores the waveform pattern characteristics of typical fault arcs. The switch feedback signal is continuously analyzed by a sliding time window. When an abnormal signal is detected whose matching degree with the feature library exceeds a set threshold, an early warning instruction containing fault location information is automatically generated. The early warning instruction triggers the modeling module to start the simulation deduction of a specific fault scenario, and compares the deduction results with the real-time monitoring data to confirm the fault type.

[0013] In one embodiment of the present invention, an edge computing node is included, which is deployed between the data acquisition module and the analysis module and is configured to perform local preprocessing on the real-time monitoring signal, including signal noise reduction, feature extraction and data compression. The edge computing node establishes a secure communication link with the cloud management platform, regularly uploads desensitized operation data and receives updated algorithm models, so as to realize remote iterative optimization of the control strategy.

[0014] In one embodiment of the present invention, a closed-loop verification channel is established between the modeling module and the execution module. When the deviation between the transient impact peak value measured during the actual switching process and the simulated prediction value exceeds the tolerance range, a model correction instruction is automatically generated and fed back to the virtual model. The correction instruction includes a parameter adjustment coefficient and a topology update mark, which triggers the virtual model to perform adaptive calibration of the structural parameters and synchronizes the calibration results to the evaluation weight calculation process of the analysis module.

[0015] The present invention also provides an adaptive switching method for power supply and distribution, comprising: S1: Collect voltage waveform signals, current phase signals and load impedance spectrum signals, and generate real-time monitoring signals; S2: Performs a dynamic weighted evaluation of voltage stability, harmonic distortion, frequency deviation, load matching, and power supply cost based on real-time monitoring signals, and generates an evaluation signal. S3: Build a virtual model of the power supply system and input the assessment signal into the virtual model to generate a simulation signal including impact prediction and recovery time; S4: sorting and processing the analog signal through the algorithm and generating a control signal including a path selection instruction; S5: Controlling the semiconductor device to switch the physical connection state of the power supply according to the control signal, and generating a switch feedback signal at the same time; S6: The equipment health assessment result is formed based on the switch feedback signal and the abnormal arc signal when the execution module switches the connection state and is transmitted to the decision module.

[0016] The present invention provides an adaptive switching system and method for power supply and distribution. The system collects voltage, current and load impedance signals in real time through multiple types of sensors, constructs an analysis module including a dynamic weight evaluation model, and comprehensively calculates power quality indicators. Digital twin technology is used to create a virtual power supply system, simulates transient responses under different switching paths, and selects the optimal solution through a reinforcement learning algorithm. A hybrid actuator with semiconductor switches and mechanical contactors in parallel is used to preferentially complete fast switching through solid-state devices to reduce arc shock. A closed-loop verification mechanism is established to compare actual switching parameters with simulation results and dynamically correct model accuracy. A high-frequency arc feature library is introduced to identify abnormal signals, and preventive switching is triggered in combination with equipment life prediction. A dynamic balance between power supply cost and safety is achieved through edge computing and cloud-based collaborative optimization strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A system architecture diagram of an adaptive switching system for power supply and distribution; Figure 2 The present invention is a flow chart of a method for adaptive switching of power supply and distribution. DETAILED DESCRIPTION

[0019] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0020] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0021] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the present invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments of the present invention.

[0022] See Figure 1-2, shown is an adaptive switching system and method for power supply and distribution of the present invention. An adaptive switching system for power supply and distribution of the present invention includes a data acquisition module, an analysis module, a modeling module, a decision module, an execution module and an early warning module. The data acquisition module collects voltage waveform signals, current phase signals and load impedance spectrum signals and generates real-time monitoring signals; the analysis module performs dynamic weighted evaluation of voltage stability, harmonic distortion rate, frequency deviation, load matching and power supply cost based on the real-time monitoring signals and generates an evaluation signal; the modeling module constructs a virtual model of the power supply system and inputs the evaluation signal into the virtual model to generate a simulation signal including impact prediction and recovery time; the decision module sorts and processes the simulation signal through an algorithm and generates a control signal including a path selection instruction; the execution module controls the semiconductor device to switch the physical connection state of the power supply according to the control signal and generates a switch feedback signal at the same time; the early warning module forms an equipment health assessment result based on the switch feedback signal and the abnormal arc signal when the execution module switches the connection state and transmits it to the decision module.

[0023] like Figure 1As shown, the present invention relates to an adaptive switching system for power supply and distribution. The core innovation of the data acquisition module lies in the design and collaborative working mechanism of its wideband impedance detection unit. This unit uses active signal injection technology to inject a test signal within a specific frequency range into the power supply circuit to acquire full-band characteristic data of the load impedance spectrum in real time. The test signal frequency range automatically adjusts based on the load type. For example, for inductive loads such as motors, the focus is on capturing impedance characteristics in the low-frequency range (50Hz-1kHz), while for nonlinear loads such as inverters, the focus is on harmonic impedance analysis in the mid- and high-frequency ranges (1kHz-100kHz). This dynamic frequency adjustment capability enables the system to accurately identify complex impedance characteristics in mixed load scenarios. To ensure that the test signal injection process does not affect normal power supply quality, a real-time signal compensation mechanism is established between the wideband impedance detection unit and the current phase signal acquisition channel. A reverse superposition cancellation algorithm is used to eliminate interference from the injected signal on the original current phase measurement, ensuring phase measurement accuracy at the milliradian level. When the load type recognition unit detects a change in load characteristics (such as the start-up of a large motor or the commissioning of variable-frequency equipment), it immediately triggers the sensor array's adaptive sampling frequency adjustment mechanism. For sudden load conditions, the sampling frequency is increased to more than three times the normal level. High-frequency noise filtering is also activated to prevent signal aliasing while ensuring data integrity. The acquired load impedance spectrum signal is not only used for real-time monitoring but also synchronously transmitted to the modeling module's virtual model parameter update interface. An impedance parameter iteration algorithm is used to correct the equivalent impedance network at the load end of the virtual model, improving the load characteristic simulation accuracy of the digital twin by over 40%. This dynamic parameter correction capability is particularly suitable for new energy microgrid scenarios, effectively addressing the impedance characteristic fluctuations caused by the integration of power electronic equipment such as photovoltaic inverters and energy storage converters. Furthermore, the module features redundant signal channels. If an anomaly in the primary acquisition channel occurs, a backup channel takes over signal acquisition within 10 milliseconds. The self-diagnostic system then sends a maintenance alert to the early warning module, ensuring continuous and reliable data acquisition.

[0024] Furthermore, the key technical aspect of the analysis module's dynamic weighted evaluation mechanism lies in the construction of an adaptive weighting system for multi-dimensional evaluation indicators. This system dynamically adjusts the weighting coefficients of various evaluation indicators based on intelligent identification of the system's operating scenarios. For example, in commercial building power supply scenarios, it automatically activates the economy-first mode, increasing the weight of the power supply cost indicator to 1.5 times that of the conventional mode. It also incorporates electricity price time factors to adjust the cost calculation model in real time. In locations with stringent power quality requirements, such as hospitals and data centers, it switches to a safety-first mode, increasing the combined weighting of voltage stability and harmonic distortion to over 70%, and implementing strict power quality standards and thresholds. The weighting adjustment algorithm, based on a deep neural network, analyzes characteristic patterns in historical operating data (such as load curve patterns, fault occurrence time distribution, and power output fluctuation characteristics) to establish a scenario classification model and automatically generate the optimal weighting combination. To continuously optimize the evaluation strategy, a bidirectional data channel is established between this module and the modeling module. Actual performance data after each switching operation is fed back to the evaluation model. By comparing simulation predictions with actual operating data, the characteristic parameters in the weighting strategy are dynamically adjusted. For example, if statistics from multiple switching operations reveal that the impact of harmonic distortion on equipment lifespan is underestimated, the system automatically adjusts the baseline weight coefficient for this metric and strengthens harmonic monitoring in subsequent evaluations. This closed-loop optimization mechanism enables the evaluation model to adaptively evolve based on long-term factors such as equipment aging and changes in grid structure, improving evaluation accuracy by over 35% compared to traditional fixed-weight models. The module also features the ability to identify abnormal weight combinations. If the weight of a particular metric consistently deviates from a reasonable range (e.g., the power supply cost weight exceeds 80% for 10 consecutive operations), a manual review mechanism is automatically triggered and a system parameter verification request is sent to operations and maintenance personnel to prevent the algorithm from becoming trapped in a local optimum. In scenarios with high renewable energy penetration, the module also incorporates a source-load matching assessment sub-model. By analyzing the spatiotemporal correlation between the output characteristics of distributed generation (DGs) and the load demand curve, it optimizes the switching timing strategy and maximizes clean energy utilization.

[0025] In one embodiment of the present invention, the innovation focus of the modeling module involved is to achieve dynamic reconstruction of the power supply network topology and synchronization of simulation time domain. The built-in virtual model of this module adopts a modular modeling method to decompose the power supply network into a standardized component library including power supply nodes, transmission lines, switching devices, load units, etc. Each component contains multi-dimensional attributes such as electrical parameters, thermodynamic parameters, and aging coefficients. When a distributed power source is connected or a load node is changed (such as a new photovoltaic grid connection point or an important load is switched on and off), the topology reconstruction engine will automatically scan the connection relationship changes of the physical system, generate a new network topology based on the graph theory algorithm, and construct a corresponding simulation model through equivalent circuit conversion technology. During the model verification stage, the system will perform convergence testing and parameter sensitivity analysis. For high impedance ratio loops that may cause simulation divergence, virtual damping elements are automatically inserted to ensure calculation stability. To achieve time synchronization between virtual and real systems, the module has developed a high-precision time domain alignment technology. By deploying a hardware timestamp marking unit on the physical system side, a microsecond time tag is added to each data packet of the collected signal. On the virtual model side, a variable step-size simulation algorithm is used, combined with cubic spline interpolation technology to compensate for the time scale deviation caused by signal transmission delay, so that the synchronization error between the simulation clock and the physical system clock is controlled within 100 microseconds. This time domain synchronization capability is crucial for transient process simulation, especially when analyzing microsecond transient events such as short-circuit faults or lightning surges. It can ensure that the phase error between the predicted impact current waveform and the actual recorded data is less than 1%. In view of the high dynamic characteristics of the wind, solar and storage combined system, the modeling module also introduces a real-time parameter identification algorithm to perform online identification of key parameters such as the equivalent impedance and inertia time constant of renewable energy power generation equipment every 5 minutes, and update them to the virtual model database. When the system detects that the deviation between the transient impact peak value during the actual switching process and the simulated predicted value exceeds the tolerance threshold, it will initiate a model parameter self-calibration process, adjusting the equivalent inductance, capacitance, and other parameters in the virtual model through a particle swarm optimization algorithm until the error between the simulation result and the actual measurement value returns to a reasonable range. This dynamic calibration mechanism enables the model to maintain high-precision simulation capabilities under long-term changes such as equipment aging and increased contact resistance. In addition, the modeling module also has a multi-scenario parallel simulation function, which can simultaneously run simulation calculations for various preset scenarios such as normal operating conditions, N-1 faults, and extreme weather, providing multi-dimensional scenario preview data support for the decision-making module.

[0026] like Figure 1As shown in Figure 2, the core innovation of the decision-making module lies in the design and comprehensive evaluation system of its multi-objective optimization engine. This engine constructs a mathematical model that incorporates multi-dimensional constraints such as power supply reliability, equipment loss costs, and operational risk levels. It then screens candidate switching solutions by incorporating Pareto optimality theory. During the computational process, a multidimensional space is first established, containing constraint boundaries such as voltage fluctuation tolerance, maximum allowable harmonic content, and minimum power supply continuity time. A genetic algorithm is then used to intelligently search the feasible solution space, generating a set of non-inferior solutions that satisfy all hard constraints. For each candidate solution, the system quantitatively evaluates its comprehensive benefit indicators, including parameters such as expected equipment life loss, power quality improvement, and operating cost change rate, to form a multidimensional feature vector. The fuzzy comprehensive evaluation model uses a trapezoidal membership function to address the fuzzy boundaries of each indicator. The expert system weighting method is then used to determine the relative importance of each evaluation dimension, ultimately calculating a comprehensive score for each solution. To improve the credibility of decisions, the output results of the optimization engine need to be cross-validated with the equipment health assessment data provided by the early warning module. When it is detected that the remaining life of the switching devices involved in the recommended solution is lower than the safety threshold, the system automatically reduces the priority of the solution and triggers an equipment replacement prompt. This verification mechanism effectively avoids the decision-making deviation that may be caused by relying solely on simulation data, especially when dealing with gradual faults such as contactor contact oxidation and line insulation aging. In the new energy microgrid application scenario, the optimization engine also integrates clean energy utilization indicators. By analyzing the matching degree between the photovoltaic output forecast curve and the load demand, it gives priority to the switching path that can maximize the absorption of renewable energy. When the system is in island operation mode, the engine automatically increases the weight coefficients of frequency stability and voltage regulation capability to ensure the autonomous operation capability of the microgrid after the switching operation.

[0027] Specifically, the hybrid switch architecture and action coordination mechanism of the execution module hinges on the coordinated control strategy of semiconductor switches and mechanical contactors. The semiconductor switch group utilizes reverse-conducting IGBTs in parallel branches, equipped with an active current-sharing circuit to ensure balanced current distribution across the parallel units. Its gate drive circuit features nanosecond-response overcurrent protection, enabling rapid latching within 3 microseconds upon detecting a short-circuit current. The mechanical contactor group utilizes a dual-breakpoint magnetic blowout structure, with the moving and static contacts plated with a silver-based composite material. The accompanying electromagnetic operating mechanism is equipped with a buffer device to reduce closing bounce. The action timing coordination unit controls the opening and closing phases of the two devices through a high-precision clock synchronization system. During the initial switching operation, the semiconductor switches are first triggered to conduct at the current zero crossing, utilizing their arc-free characteristics to complete the initial current transfer. Once the load current stabilizes, the mechanical contactors close at the voltage zero crossing to assume the steady-state current-carrying role. This step-by-step operation strategy reduces arc energy during the switching process to less than 20% of traditional methods. The timing coordination unit is also designed with a dynamic adjustment function. When high-frequency harmonic components are detected in the load current, the trigger phase of the semiconductor switch is automatically corrected to avoid device stress concentration caused by conduction during the harmonic peak period. In a fault handling scenario, if the early warning module issues an emergency trip command, the coordination unit will initiate a reverse operation sequence: first disconnect the mechanical contactor, and then the semiconductor switch completes the final isolation at the preset safe phase point. This operation sequence can effectively avoid the risk of contact welding during high current disconnection. To improve system reliability, the execution module is also equipped with a dual control circuit. The main circuit uses optical fiber to transmit PWM drive signals, and the backup circuit transmits analog control signals through isolation transformer coupling. When the main circuit fails, it can seamlessly switch to the backup channel.

[0028] Furthermore, the core technology of the early warning module lies in the construction of an intelligent diagnostic system based on a high-frequency signal feature library. This feature library, established through long-term collection of typical arc fault samples, contains fingerprint features for 12 fault types, including contactor contact wear, insulation degradation, and loose connections. Each fault type corresponds to a specific combination of time-frequency domain feature vectors. The signal analysis unit uses sliding time window technology to process switch feedback signals in real time. The length of each time window can be adaptively adjusted. During steady-state operation, a 200-millisecond window length is used for routine monitoring. When transient processes are detected, a shorter 10-millisecond window is automatically switched for high-resolution analysis. The spectrum analysis algorithm combines fast Fourier transform and wavelet packet decomposition techniques to simultaneously capture steady-state features and transient mutation information across a wide frequency band. When an abnormal signal is detected whose match with the feature library exceeds a set threshold, the localization algorithm accurately calculates the fault location by comparing the amplitude attenuation characteristics and time delay differences of multiple sensor signals, achieving branch-level location accuracy. To verify the fault type, the system triggers the modeling module to initiate a special simulation, reconstructing the physical scenario corresponding to the abnormal signal in a virtual environment. By adjusting the fault parameters, the simulated waveform is maximized in correlation with the actual measured data. The difference comparison unit uses a dynamic time warping algorithm to address the phase deviation problem of the time series data, and combines it with the correlation coefficient matrix analysis to confirm the fault type. For new unknown faults, the system automatically generates feature templates and initiates an online learning mechanism to update the feature library. The module is also designed with a fault evolution prediction function. By establishing a Markov state transition model to analyze the changing trend of fault characteristics, it generates preventive maintenance recommendations in advance when hidden dangers such as a continuous decrease in insulation resistance or a linear increase in contact resistance are detected.

[0029] like Figure 1As shown, the edge computing node has a localized intelligent processing system. Its hardware platform utilizes a multi-core heterogeneous architecture, comprising dedicated signal processing cores, AI inference acceleration cores, and communication coprocessors. Upon entering the edge node, the raw monitoring signal first passes through an adaptive filter to eliminate power frequency interference and random noise. The filter parameters are dynamically adjusted based on real-time spectrum analysis results, reducing the noise floor to less than one-fifth of the original level while maintaining signal feature integrity. The feature extraction unit utilizes a parallel processing pipeline design, simultaneously performing time-domain statistical calculations, frequency-domain feature extraction, and joint time-frequency analysis. Key feature vectors are dimensionality-reduced using a compressed sensing algorithm before being transmitted to the analysis module. The data compression algorithm utilizes an improved sparse autoencoder network, reducing the amount of transmitted data by 60% while maintaining data reconstruction accuracy above 98%. The secure communication link utilizes an end-to-end encryption scheme that complies with national security standards, with data integrity checksums added at the application layer to prevent data tampering during transmission. A digital twin training cluster is deployed on the cloud management platform, which regularly downloads desensitized operational data uploaded by the edge nodes weekly. The control strategy model is optimized using a deep reinforcement learning algorithm, and the updated model parameters are digitally signed and pushed to the edge nodes. This collaborative optimization mechanism enables the system to rapidly respond to local real-time demands while continuously absorbing global operational experience. This is particularly true when responding to changes in regional grid parameters or new load access scenarios, allowing for rapid adaptation to environmental changes through model hot updates. Edge nodes also feature an offline emergency mode. Upon detecting a network outage, they automatically activate the latest locally stored policy model and record complete operational data for subsequent transmission upon network restoration, ensuring the system's continued operational capabilities in extreme situations.

[0030] Furthermore, the core of the closed-loop verification mechanism lies in establishing a dynamic parameter calibration channel between the virtual and real systems. This mechanism uses a high-precision measurement unit to capture key parameters such as transient surge current peaks, voltage sag amplitudes, and recovery times during module switching, and compares them synchronously with the modeling module's simulated predictions at the millisecond level. When the deviation between the actual measured value and the simulated result exceeds the tolerance threshold, the parameter identification engine automatically activates. It first performs multi-dimensional feature extraction on the difference data, including waveform similarity analysis, spectral energy distribution comparison, and timing phase offset calculation, to generate a difference feature vector containing a deviation type marker. The model correction instruction generation unit selects the corresponding calibration strategy based on the feature vector category: for steady-state parameter deviations (such as equivalent impedance mismatch), a least-squares fitting algorithm is used to adjust the line resistance and inductance parameters in the virtual model; for dynamic characteristic deviations (such as transient response delay), a transfer function reconstruction technique is used to update the inertia link time constant in the simulation algorithm. The generation of topology update markers relies on network structure change detection. If the parameter adjustment amplitude of a branch exceeds the set limit three times in a row during the verification process, the system determines that the branch may have a topological connection error, triggering the topology scanning process to reconfirm the physical system wiring relationship. The parameter adjustment coefficients generated during the calibration process are fed back to the modeling module's virtual model database via an encrypted data channel, and a version control log is established to record the details of each correction. The synchronous transmission mechanism of calibration results is designed with a priority queue. For key parameters that affect the calculation of evaluation weights (such as load impedance characteristics), the analysis module's weight coefficient recalculation process is immediately triggered. Minor parameter updates are buffered and processed during system idle periods. This layered processing strategy ensures the real-time performance of core functions while avoiding excessive consumption of system resources. In the application scenario of new energy stations, the closed-loop verification mechanism has specially added a special calibration function for the equivalent model of distributed power sources. When the output of the photovoltaic array changes suddenly or the wind turbine is switched in and out, the system automatically calls the preset energy type characteristic template, compares the actual output characteristics with the model prediction curve, and dynamically corrects the MPPT algorithm parameters of the photovoltaic unit in the virtual model or the transient reactance value of the doubly fed generator to ensure that the simulation model always keeps up with the evolution of the operating status of the physical equipment.

[0031] like Figure 1As shown, the priority coverage logic involved focuses on addressing rapid response to emergency faults. Its technical implementation relies on a multi-level condition monitoring network and intelligent decision-making architecture. The equipment health assessment system integrates multi-dimensional sensor data to construct a life prediction model for key components. This includes a mechanical life curve based on the number of switching operations, a thermal aging model based on temperature rise history, and contact loss calculations based on arc energy analysis. When the assessment results indicate that the remaining life of a key component (such as the main contactor operating mechanism) falls below a safety threshold, the health status encoder immediately generates a structured warning message containing the device ID, fault level, and expected failure time. The priority determination unit initiates the corresponding response mechanism based on the fault level: for a Level 1 emergency fault (e.g., a remaining life of less than 24 hours), a red warning code is immediately triggered and a protective switching instruction is activated; for a Level 2 severe fault, an orange code is generated to initiate backup plan planning. The protective switching instruction includes a preset safe path sequence. This sequence is generated during system initialization by traversing the safety assessment of all possible power supply paths and is regularly updated based on changes in the grid topology. When the instruction is executed, the normal control signal to the decision module is forcibly interrupted, and the system switches to fault emergency mode, prioritizing the backup path with the highest electrical isolation for load transfer. The fault isolation procedure adopts a three-level in-depth protection strategy: first, disconnect the fast circuit breakers upstream and downstream of the fault section, then implement reverse current blocking through semiconductor switches, and finally switch on the grounding switch to form a visible disconnection point. During this process, the early warning module continuously monitors the execution effect of the isolation operation. If it is found that the fault current has not been completely cut off, the backup protection plan will be activated, such as forcibly cutting off the associated load or triggering the local power outage plan. When sending the maintenance notice, the maintenance request generation unit will simultaneously attach a detailed diagnostic report of the faulty equipment, including historical operation records, key parameter trend charts and a list of recommended replacement parts. To cope with complex fault scenarios, the system is also designed with a cross-regional coordination mechanism. When the local backup power capacity is insufficient, support from adjacent distribution areas is requested through cross-site communication links to form a multi-node collaborative power supply network. This priority coverage logic has shown significant advantages in important load scenarios such as data centers. It can achieve seamless switching before the equipment fails completely, reducing the scope of the fault impact to the level of a single functional module.

[0032] like Figure 2The figure shows an adaptive switching method for power supply and distribution according to the present invention. The method comprises: S1: collecting voltage waveform signals, current phase signals, and load impedance spectrum signals, and generating real-time monitoring signals; S2: dynamically weighted evaluation of voltage stability, harmonic distortion rate, frequency deviation, load matching, and power supply cost based on the real-time monitoring signals, and generating an evaluation signal; S3: constructing a virtual model of the power supply system and inputting the evaluation signal into the virtual model to generate a simulation signal including impact prediction and recovery time; S4: sorting the simulation signal through an algorithm and generating a control signal including a path selection instruction; S5: controlling the semiconductor device to switch the physical connection state of the power supply based on the control signal, and generating a switch feedback signal at the same time; and S6: generating a device health assessment result based on the switch feedback signal and the abnormal arc signal when the execution module switches the connection state, and transmitting the result to the decision module.

[0033] like Figure 2As shown, the edge computing architecture and cloud-based collaborative mechanism demonstrate technological innovation through the construction of a hierarchical and progressive intelligent processing system. Edge computing nodes utilize a heterogeneous computing platform, integrated with FPGA chips for signal preprocessing acceleration and paired with multi-core processors to run local analysis algorithms. At the data acquisition end, the nodes have a built-in adaptive noise reduction module that uses a hybrid algorithm combining wavelet threshold denoising and Kalman filtering to eliminate on-site electromagnetic interference while preserving sharp transitions in the signal waveform. The feature extraction pipeline utilizes a parallel architecture, simultaneously calculating time-domain statistics (such as crest factor and crest factor), extracting frequency-domain features (harmonic amplitude ratio and spectral centroid), and performing joint time-frequency analysis (wavelet packet energy entropy). Key feature vectors are dimensionally compressed using principal component analysis before being uploaded. The local decision engine deploys a lightweight neural network model, capable of completing over 80% of routine switching decisions without relying on the cloud, forwarding only decision requests for complex scenarios to the cloud. The cloud management platform utilizes a distributed training cluster, integrating operational data from multiple edge nodes using a federated learning framework for global model optimization while protecting data privacy. The model update mechanism incorporates a differential privacy layer, adding Gaussian noise during parameter upload to prevent the reverse engineering of sensitive information about individual sites. A multi-objective evolutionary algorithm is incorporated into the policy iteration optimization process, searching for the optimal control strategy within tens of millions of historical data samples. The effectiveness of the new strategy is verified using a digital twin simulation environment. Updated models are released incrementally, generating incremental packages by comparing parameter differences between the new and old versions. Edge nodes only need to download megabyte-sized update files to complete the model upgrade. To ensure communication reliability, the system employs a dual-channel redundant transmission design. The primary channel utilizes a 5G network for low-latency communication, while the backup channel maintains basic data transmission via power line carriers. In extreme network outage scenarios, edge nodes activate autonomous operation, maintaining basic switching functions based on contingency plans stored in their local knowledge base. They also record complete operation logs and synchronize them to the cloud for post-analysis upon network restoration. This architecture is particularly well-suited for widely distributed distribution networks, leveraging the cloud's powerful computing resources for global optimization while ensuring real-time control through edge intelligence. It demonstrates strong collaborative resilience to regional grid outages.

[0034] An adaptive switching system and method for power supply and distribution of the present invention collects voltage, current and load impedance signals in real time through multiple types of sensors, constructs an analysis module including a dynamic weight evaluation model, and comprehensively calculates power quality indicators; uses digital twin technology to create a virtual power supply system, simulates transient responses under different switching paths, and screens the optimal solution through a reinforcement learning algorithm; adopts a hybrid actuator that connects semiconductor switches and mechanical contactors in parallel, and preferentially completes fast switching through solid-state devices to reduce arc shock; establishes a closed-loop verification mechanism, compares actual switching parameters with simulation results, and dynamically corrects model accuracy; introduces a high-frequency arc feature library to identify abnormal signals, and triggers preventive switching in combination with equipment life prediction; and achieves a dynamic balance between power supply cost and safety through edge computing and cloud-based collaborative optimization strategies.

[0035] Therefore, the adaptive switching system and method for power supply and distribution of the present invention solves the problem that the traditional power supply and distribution switching system relies on a single threshold to judge the power supply status and cannot comprehensively evaluate the power quality.

[0036] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical principles disclosed herein are intended to be covered by the claims of the present invention.

Claims

1. An adaptive switching system for power supply and distribution, characterized in that: include: A data acquisition module, which collects voltage waveform signals, current phase signals, and load impedance spectrum signals and generates real-time monitoring signals; an analysis module, wherein the analysis module performs a dynamic weighted evaluation on voltage stability, harmonic distortion rate, frequency deviation, load matching degree and power supply cost according to the real-time monitoring signal, and generates an evaluation signal; a modeling module, the modeling module constructing a virtual model of the power supply system and inputting the assessment signal into the virtual model to generate a simulation signal including an impact prediction and a recovery time; A decision module, which processes the analog signals by sorting them using an algorithm and generates a control signal including a path selection instruction; an execution module, wherein the execution module controls the semiconductor device to switch the physical connection state of the power supply according to the control signal and generates a switch feedback signal; An early warning module forms an equipment health assessment result based on the switch feedback signal and the abnormal arc signal when the execution module switches the connection state, and transmits the result to the decision module.

2. The adaptive switching system for power supply and distribution according to claim 1, characterized in that: The data acquisition module includes a wideband impedance detection unit, which obtains a load impedance spectrum signal by injecting a test signal of a specific frequency and works in conjunction with a load type identification unit. When a change in the load type is detected, the sampling frequency range of the sensor array is dynamically adjusted. The analysis results of the load impedance spectrum signal are synchronously transmitted to the modeling module for correcting the equivalent impedance parameters of the load end in the virtual model. A signal compensation mechanism is established between the wideband impedance detection unit and the current phase signal acquisition channel to eliminate interference caused by the test signal injection on the original current phase measurement.

3. The adaptive switching system for power supply and distribution according to claim 1, characterized in that: The weight coefficient of the dynamic weighted evaluation in the analysis module is automatically adjusted according to the power supply scenario mode. When the system is in the economic priority mode, the weight ratio of the power supply cost indicator is increased, and in the safety priority mode, the comprehensive weight of voltage stability and harmonic distortion rate is strengthened. The weight adjustment process realizes pattern recognition through feature extraction of historical operation data, and keeps linkage with the virtual model parameter update of the modeling module, forming a closed-loop optimization mechanism of the evaluation strategy.

4. The adaptive switching system for power supply and distribution according to claim 1, characterized in that: The virtual model of the modeling module includes a dynamic reconstruction function for the power supply network topology. When a distributed power source is detected or a load node is changed, the virtual model automatically generates a new equivalent circuit model and verifies its convergence. A time domain synchronization channel is established between the virtual model and the physical system, and an interpolation algorithm is used to compensate for the time scale deviation caused by signal transmission delay, ensuring that the impact prediction results are aligned with the time domain characteristics of the actual transient process.

5. The adaptive switching system for power supply and distribution according to claim 1, characterized in that: The algorithm of the decision-making module includes a multi-objective optimization engine, which quantifies power supply reliability indicators, equipment loss costs and switching operation risks into multi-dimensional constraints, screens non-inferior solution sets through Pareto front analysis, and introduces a fuzzy comprehensive evaluation model to prioritize candidate solutions. The output results of the multi-objective optimization engine are cross-validated with the equipment health assessment results of the early warning module to form a quantitative scoring mechanism for decision credibility.

6. The adaptive switching system for power supply and distribution according to claim 1, characterized in that: The execution module includes a fast semiconductor switch group and a mechanical contactor group arranged in parallel, wherein the semiconductor switch group is configured to preferentially respond to high-frequency switching instructions and undertake transient current transfer, and the mechanical contactor group automatically closes after the semiconductor switch completes conduction to maintain steady-state current carrying. An action timing coordination unit is set between the semiconductor switch group and the mechanical contactor group to ensure that the opening and closing operations of the two groups of devices meet the preset phase synchronization requirements.

7. The adaptive switching system for power supply and distribution according to claim 1, characterized in that: The early warning module includes a high-frequency signal feature library, which stores the waveform pattern characteristics of typical fault arcs. Continuous spectrum analysis is performed on the switch feedback signal through a sliding time window. When an abnormal signal is detected whose matching degree with the feature library exceeds a set threshold, an early warning instruction containing fault location information is automatically generated. The early warning instruction triggers the modeling module to start simulation deduction of a specific fault scenario, and compares the deduction results with real-time monitoring data to confirm the fault type.

8. The adaptive switching system for power supply and distribution according to claim 1, characterized in that: It includes an edge computing node, which is deployed between the data acquisition module and the analysis module and is configured to perform local preprocessing on the real-time monitoring signal, including signal noise reduction, feature extraction and data compression. The edge computing node establishes a secure communication link with the cloud management platform, regularly uploads desensitized operation data and receives updated algorithm models to achieve remote iterative optimization of the control strategy.

9. The adaptive switching system for power supply and distribution according to claim 1, characterized in that: A closed-loop verification channel is established between the modeling module and the execution module. When the deviation between the transient impact peak value measured during the actual switching process and the simulated prediction value exceeds the tolerance range, a model correction instruction is automatically generated and fed back to the virtual model. The correction instruction contains a parameter adjustment coefficient and a topology update mark, triggering the virtual model to perform adaptive calibration of the structural parameters, and at the same time synchronizing the calibration results to the evaluation weight calculation process of the analysis module.

10. An adaptive switching method for an adaptive switching system for power supply and distribution using any one of claims 1 to 9, characterized in that: include: S1: Collect voltage waveform signals, current phase signals and load impedance spectrum signals, and generate real-time monitoring signals; S2: Performing a dynamic weighted evaluation of voltage stability, harmonic distortion rate, frequency deviation, load matching, and power supply cost based on the real-time monitoring signal, and generating an evaluation signal; S3: constructing a virtual model of the power supply system and inputting the evaluation signal into the virtual model to generate a simulation signal including an impact prediction and a recovery time; S4: sorting the analog signals through an algorithm and generating a control signal including a path selection instruction; S5: Controlling the semiconductor device to switch the physical connection state of the power supply according to the control signal, and generating a switch feedback signal at the same time; S6: Forming a device health assessment result based on the switch feedback signal and the abnormal arc signal when the execution module switches the connection state and transmitting the result to the decision module.

Images