Switching power supply circuit system based on intelligent monitoring and protection

Abstract

The invention relates to the technical field of power electronics, and discloses a switching power supply circuit system based on intelligent monitoring and protection, and the system comprises a power supply monitoring module which is used for collecting the voltage, current, temperature and environment data of a power supply system in real time; the intelligent decision-making and learning module is used for receiving the voltage, current, temperature and environment data transmitted by the power supply monitoring module, analyzing the working state of the power supply system through reinforcement learning and a neural network algorithm based on the data and optimizing the operation parameters of the power supply so as to generate a protection and optimization strategy; and the multi-element collaborative protection module is used for receiving the protection and optimization strategy output by the intelligent decision-making and learning module. Through deep fusion of the intelligent monitoring module and the reinforcement learning algorithm, the system can sense key indexes such as voltage, current, temperature and environment in real time, autonomously identify the power supply operation state and intelligently adjust the operation strategy, so that dynamic optimization and accurate control of the power supply system are realized, and the adaptability and stability of the system are enhanced.

Description

Technical Field

[0001] The present invention relates to the field of power electronics technology, and specifically to a switching power supply circuit system based on intelligent monitoring and protection. Background Art

[0002] With the continuous improvement of the performance requirements for power supply systems in fields such as industrial control, communication equipment, and consumer electronics, switching power supplies are widely used due to their high efficiency, small size, and high reliability. However, existing switching power supply systems generally rely on fixed control parameters and traditional threshold protection mechanisms during actual operation, and their response capabilities to external environmental changes and load dynamic fluctuations are weak, prone to problems such as adjustment hysteresis, protection lag, and even mis-triggering, especially difficult to ensure the operation stability and safety of the system in complex application scenarios.

[0003] In addition, traditional power supply systems often lack real-time data learning and feedback mechanisms, unable to effectively model and evolve strategies based on long-term operation data, which greatly limits the improvement of power efficiency and the ability of fault prediction. At the same time, the computing resources of the power supply system are limited, and it is difficult to achieve efficient real-time operation when facing complex algorithm models or multi-parameter control requirements, thus affecting the response speed and optimization accuracy. In terms of remote operation and maintenance, existing solutions mainly focus on passive monitoring, lacking remote adaptive adjustment capabilities based on global perception and intelligent decision-making, which is not conducive to building a highly available and reliable energy management system. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides a switching power supply circuit system based on intelligent monitoring and protection, which solves the problems of weak dynamic adjustment ability, lagging protection response, fixed control parameters, and lack of intelligent remote operation and maintenance in existing switching power supply systems.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: A switching power supply circuit system based on intelligent monitoring and protection, including: A power supply monitoring module, used to collect voltage, current, temperature, and environmental data of the power supply system in real time; An intelligent decision-making and learning module, receiving the voltage, current, temperature, and environmental data transmitted by the power supply monitoring module, analyzing the working state of the power supply system based on the data through reinforcement learning and neural network algorithms, and optimizing the operating parameters of the power supply to generate protection and optimization strategies; A multi-element collaborative protection module, receiving the protection and optimization strategies output by the intelligent decision-making and learning module, and performing overvoltage, overcurrent, over-temperature, and short-circuit protection operations in real time based on real-time monitoring data and decision results; Optimization algorithm module, which receives the real-time data from the power supply monitoring module and the optimization strategy from the intelligent decision-making and learning module, and adaptively optimizes the control parameters of the power supply system based on the genetic algorithm or the particle swarm optimization algorithm; Hardware acceleration and computing module, which is used to accelerate the computing processes of the intelligent decision-making and learning module and the optimization algorithm module; Cloud platform and remote monitoring module, which uploads the real-time data of the power supply monitoring module and the intelligent decision-making and learning module to the cloud platform through the Internet of Things technology for remote monitoring, data analysis and global optimization, and at the same time feeds back the optimization results and protection strategies to the power supply system.

[0006] Preferably, the power supply monitoring module includes: Voltage monitoring unit, which is used to collect the power supply output voltage signal in real time and transmit it to the intelligent decision-making and learning module; Current monitoring unit, which is used to detect the power supply output current in real time through a Hall sensor or a shunt resistor and transmit it to the intelligent decision-making and learning module; Temperature monitoring unit, which is used to monitor the internal temperature of the power supply system and transmit it to the intelligent decision-making and learning module; Environmental monitoring unit, which is used to collect environmental data, including humidity, air pressure, etc., and transmit it to the intelligent decision-making and learning module.

[0007] Preferably, the intelligent decision-making and learning module includes: Reinforcement learning unit, which is used to optimize the power supply protection threshold and working strategy according to the monitoring signal through the reinforcement learning algorithm; Neural network unit, which is used to adjust the power supply working state based on historical data and real-time monitoring signals; The update formula of the reinforcement learning algorithm is: ; Wherein, represents the value of taking action in state ; is the learning rate; is the discount factor; is the immediate reward; represents the return value of selecting the optimal action in the next state.

[0008] Preferably, the multi-element collaborative protection module includes: Overvoltage protection unit, which is used to trigger the overvoltage protection operation when the power supply output voltage exceeds the set threshold; Overcurrent protection unit, which is used to trigger the overcurrent protection operation when the power supply output current exceeds the set value; Over-temperature protection unit, which is used to trigger the over-temperature protection operation when the internal temperature of the power supply exceeds the set value; The short - circuit protection unit is used to immediately cut off the power output when a short - circuit occurs.

[0009] Preferably, the optimization algorithm module includes: The genetic algorithm unit is used to optimize the control parameters of the power supply through the genetic algorithm; The particle swarm optimization unit is used to dynamically adjust the power supply operating frequency and PWM control signal through the particle swarm optimization algorithm; The velocity update formula of the particle swarm optimization algorithm is: ; Where, represents the velocity of particle at the th generation; is the inertia weight; and are the learning factors; and are random numbers; is the historical best position of particle ; is the global best position of the particle swarm.

[0010] Preferably, the hardware acceleration and computing module includes: The FPGA acceleration unit is used to accelerate the calculation of machine learning algorithms and optimization algorithms; The AI chip acceleration unit is used to accelerate the execution of neural network models.

[0011] Preferably, the cloud platform and remote monitoring module includes: The data upload unit is used to transmit the real - time data of the power supply system to the cloud platform through Internet of Things technology; The remote monitoring and management unit is used to achieve remote monitoring, fault diagnosis and adjustment of the system through a Web interface or a mobile application.

[0012] Preferably, the intelligent decision - making and learning module adjusts the power supply protection threshold in real - time through a reinforcement learning algorithm to adapt to different workloads and environmental conditions.

[0013] Preferably, when the power supply load changes, the optimization algorithm module adjusts the control parameters in real - time to maximize the power supply efficiency and stability.

[0014] The fitness function optimization formula of the genetic algorithm is: ; Where, represents the optimal control parameters obtained by optimizing through the genetic algorithm; is the fitness function of the power supply system.

[0015] The present invention also provides a switching power supply circuit method based on intelligent monitoring and protection, and the method includes the following steps: S1. Real-time collect the output voltage, current, temperature and environmental data of the power supply; S2. Input the collected monitoring signals into the intelligent decision-making and learning module, and optimize the power supply working state through reinforcement learning and neural network algorithms; S3. According to the optimization result, trigger the protection operation in real time through the multi-element collaborative protection module to ensure the safety of the power supply; S4. Optimize the working efficiency of the power supply through the optimization algorithm module, and adjust the control parameters to improve the system performance; S5. Accelerate the calculation process through the hardware acceleration and calculation module to ensure the real-time response of the system; S6. Upload the monitoring data and optimization results to the cloud platform through Internet of Things technology for remote monitoring and data analysis The present invention provides a switching power supply circuit system based on intelligent monitoring and protection. It has the following beneficial effects: 1. Through the deep integration of the intelligent monitoring module and the reinforcement learning algorithm, the system can real-time sense key indicators such as voltage, current, temperature and environment, autonomously identify the operation state of the power supply and intelligently adjust the operation strategy, so as to realize the dynamic optimization and precise control of the power supply system, and enhance the adaptability and stability of the system.

[0016] 2. Through the multi-element collaborative protection module according to the optimization strategy provided by the intelligent decision-making module, the system can realize the rapid identification and response to various faults such as overvoltage, overcurrent, overheating and short circuit, establish a multi-dimensional and multi-level protection mechanism, and significantly enhance the self-protection ability of the power supply system in abnormal environments.

[0017] 3. By integrating the genetic algorithm and the particle swarm optimization algorithm, the system can continuously and adaptively optimize parameters such as the working frequency and control signal of the power supply, so as to improve the overall energy efficiency performance, reduce energy consumption, adapt to different load changes, and realize the coordinated unity of high performance and energy saving.

[0018] 4. Through the hardware acceleration and calculation module, by virtue of the parallel acceleration characteristics of the FPGA and the AI chip, the system provides efficient computing power support for the reinforcement learning and optimization algorithms, ensuring that the system has a millisecond-level response ability in the face of sudden load changes, and avoiding protection failure or operation disorder caused by calculation delay.

[0019] 5. Through the cloud platform and remote monitoring module constructed by Internet of Things technology, the power supply system can be analyzed and adjusted remotely all-weather, enabling maintenance personnel to timely master the system operation state, realizing fault prediction, diagnosis and remote control, reducing the maintenance cost, and improving the intelligent level of system operation and maintenance. Brief Description of the Drawings

[0020] Figure 1 is the system architecture diagram of the present invention; Figure 2 is the schematic structural diagram of the power supply monitoring module of the present invention; Figure 3 is the schematic structural diagram of the intelligent decision-making and learning module of the present invention; Figure 4 is the schematic structural diagram of the multi-element collaborative protection module of the present invention; Figure 5 is the schematic structural diagram of the optimization algorithm module of the present invention; Figure 6 is the schematic structural diagram of the hardware acceleration and computing module of the present invention; Figure 7 is the schematic structural diagram of the cloud platform and remote monitoring module of the present invention; Figure 8 is the method flow chart of the present invention. Detailed Description of the Preferred Embodiments

[0021] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0022] Please refer to the attached Figure 1 - attached Figure 7 , the embodiments of the present invention provide a switched-mode power supply circuit system based on intelligent monitoring and protection, including the following modules: A power supply monitoring module for real-time acquisition of voltage, current, temperature and environmental data of the power supply system; In this embodiment, the power supply monitoring module is used to real-time collect various operating state data of the power supply system, including but not limited to voltage, current, temperature, and environmental parameters, etc. As the basic unit of the system, this module is responsible for providing feedback information to ensure that the intelligent decision-making and learning module can make optimized decisions based on accurate and real-time data. The core design of the power supply monitoring module is high-precision and low-latency real-time data acquisition and transmission to ensure that the system can respond to any abnormal state in a timely manner.

[0023] In this embodiment, the power supply monitoring module is composed of multiple sub-units, which are respectively responsible for collecting different types of data. Specifically, it includes a voltage monitoring unit, a current monitoring unit, a temperature monitoring unit, and an environmental monitoring unit. The functions of each unit are independent of each other, and at the same time, they cooperate with each other to jointly provide complete monitoring information for the system.

[0024] The voltage monitoring unit is responsible for the real-time monitoring of the power supply output voltage. This unit typically uses a high-precision voltage sensor, combined with an analog-to-digital converter (ADC) to convert the voltage signal into a digital signal for subsequent processing. By setting a predetermined voltage threshold, the voltage monitoring unit can send a warning signal to the system when the voltage exceeds or falls below the safe range. The acquisition of voltage values can adopt an accurate feedback control loop to ensure that the system can quickly respond to voltage fluctuations.

[0025] The current monitoring unit detects the current value of the power supply output in real time through a Hall effect sensor or a shunt resistor. Under high-load or abnormal load conditions, the change in the current value is particularly important. The current monitoring unit can transmit the digital signal to the intelligent decision-making and learning module when the current value exceeds the set safe range, thereby triggering relevant protection mechanisms. In addition, the current monitoring unit can also provide trend information on load changes, which helps to predict the working state of the power supply system.

[0026] The temperature monitoring unit uses a high-precision temperature sensor (such as an NTC thermistor) to monitor the temperature inside and around the power supply system in real time. During the operation of the power supply system, the change in temperature directly affects its efficiency and safety. Especially under high-power output or harsh working environment conditions, temperature monitoring is particularly important. The collected temperature data will be uploaded to the intelligent decision-making module in real time as a reference for generating protection strategies.

[0027] The environmental monitoring unit is used to monitor factors such as air pressure and humidity in the power supply working environment. Environmental data has a certain impact on the power supply performance. Especially under high humidity, high pressure, or extreme temperature conditions, the power supply may become unstable or malfunction. The environmental monitoring unit provides additional reference information for the intelligent decision-making and learning module by collecting data on environmental changes. By collecting environmental data in real time, the system can more comprehensively evaluate the operating state of the power supply and give early warnings of potential risks.

[0028] All the collected data, including voltage, current, temperature, and environmental data, are transmitted to the intelligent decision-making and learning module through a high-speed data bus. The data transmission process should have high reliability and low latency to ensure that information can be fed back to the decision-making system in real time, thereby achieving real-time control and adjustment.

[0029] To ensure the accuracy and reliability of data, a multiple redundancy design is adopted for the power monitoring module in this embodiment. For example, the voltage monitoring unit and the current monitoring unit will simultaneously and independently monitor the same power output, and compare the values collected by each. If there are significant differences in the monitoring results of the two, the system will correct them through a self-calibration mechanism to avoid the impact of sensor failures or data anomalies on the system. In addition, the system also has a fault detection mechanism. Once a monitoring unit fails, the system can automatically switch to a backup monitoring channel to ensure continuous data collection.

[0030] The power monitoring module of this embodiment has a flexible configuration function, supporting the adjustment of monitoring accuracy and response speed according to the requirements of different power systems. By adopting high-precision sensors and advanced data acquisition technologies, the power monitoring module can continuously provide high-quality data without increasing the system burden, providing solid data support for intelligent decision-making and the generation of protection strategies. The intelligent decision-making and learning module receives the voltage, current, temperature, and environmental data transmitted by the power monitoring module, analyzes the working state of the power system based on the data through reinforcement learning and neural network algorithms, and optimizes the operating parameters of the power supply to generate protection and optimization strategies. In this embodiment, the intelligent decision-making and learning module is the core part of the system, responsible for dynamically optimizing the working state of the power system according to the real-time data provided by the power monitoring module through machine learning algorithms, and generating optimization decisions and protection strategies. The design of this module relies on two advanced intelligent algorithms, reinforcement learning and neural networks, to process multi-dimensional data from each monitoring unit and make scientific and reasonable control decisions to ensure the efficient and safe operation of the power system.

[0031] In this embodiment, the intelligent decision-making and learning module is mainly composed of two core sub-units: the reinforcement learning unit and the neural network unit. These two units cooperate with each other and provide continuous optimization decision support for the power system through continuous learning and adaptive adjustment.

[0032] The reinforcement learning unit uses reinforcement learning algorithms to adjust the operating strategy of the power system. The basic principle of the reinforcement learning algorithm is to optimize the decision-making process through feedback from the environment. Specifically, the system will perform certain control operations based on the real-time collected voltage, current, temperature, and environmental data, and give rewards or punishments according to the changes in the system state. Through multiple iterations, the reinforcement learning model continuously adjusts the power control parameters to reach the optimal working state. For example, in the case of large current fluctuations, the reinforcement learning unit will adjust the current output limit according to the system feedback to avoid overcurrent damage to the power supply.

[0033] The mathematical model of the reinforcement learning process can be expressed by the following formula: ; Among them, represents the value of taking action under state ; is the learning rate; is the discount factor; is the immediate reward; represents the return value of choosing the optimal action in the next state.

[0034] Neural network units are used to handle more complex non - linear data relationships. Through deep - learning algorithms, the neural network can learn the complex characteristics of the power system from a large amount of historical data and use these learned characteristics for prediction and decision - making. When processing power - system data, the neural network can generate the best working parameters according to real - time data. For example, when the load of the power system changes, the neural network can predict future load changes based on historical data and adjust the power output in advance to ensure that the power remains stable under load fluctuations.

[0035] The mathematical model of the neural network is implemented through a multi - layer perceptron (MLP), and its basic structure is: ; Among them, is the input data; is the weight of the network layer; represents the activation function; is the network output. Through multi - level learning and processing of the input data, the neural network can output corresponding power - control strategies to optimize the working efficiency of the power system.

[0036] The output of the intelligent decision - making and learning module is the optimized power - operation parameters and protection strategies. These outputs will be transmitted to the downstream multi - variable collaborative protection module and optimization - algorithm module. According to the actual situation, the intelligent decision - making and learning module can dynamically adjust the control parameters of the power supply, such as current, voltage range, operating frequency, etc., to ensure that the power system can maintain a stable and efficient working state under various environments and loads.

[0037] In this embodiment, the intelligent decision - making and learning module also has a certain self - adaptive learning ability. When the power system enters a new working environment or load mode, the system can adjust the optimization strategy through self - learning. The system is trained by continuously interacting with the environment, enabling the decision - making module to adapt to new operating conditions without manual intervention.

[0038] It is worth noting that the intelligent decision-making and learning module can process the data from the power monitoring module more efficiently through the combined use of reinforcement learning and neural networks, and make real-time decisions based on the feedback of the system. The combination of the two algorithms enables the system to maintain high robustness and flexibility in the face of changing working environments and complex power loads. Through these intelligent decision-making supports, the power system can achieve optimized operations under various conditions, ensuring both the efficient operation of the system and the effective avoidance of potential risks.

[0039] The multi-element collaborative protection module receives the protection and optimization strategies output by the intelligent decision-making and learning module, and based on the real-time monitoring data and decision results, it performs overvoltage, overcurrent, overtemperature, and short-circuit protection operations in real time; In this embodiment, the multi-element collaborative protection module is responsible for providing real-time protection functions during the operation of the power system. This module collaboratively performs a series of protection operations such as overvoltage, overcurrent, overtemperature, and short-circuit protection according to the optimization strategy from the intelligent decision-making and learning module and the real-time data provided by the power monitoring module, ensuring that the power system can respond quickly in case of abnormalities and avoid irreparable damage to the power supply and load devices.

[0040] The multi-element collaborative protection module realizes the real-time protection of various parameters of the power system through multiple independent protection units. Each protection unit responds to specific types of abnormal situations and performs protection actions under the guidance of the system's protection strategy. Through the collaborative action of multiple protection mechanisms, this module can ensure that the power system is always in a safe and stable state under different working conditions.

[0041] In this embodiment, the overvoltage protection unit is responsible for monitoring the real-time value of the power supply output voltage. When the voltage exceeds the preset safe range, the overvoltage protection unit will immediately respond and initiate the corresponding protection operation. The working principle of overvoltage protection is to compare the real-time voltage value with the set maximum voltage threshold. If the voltage value exceeds this threshold, the voltage is reduced by controlling the output port of the power supply or adjusting the voltage regulating device to prevent damage to the power supply or load device caused by excessive voltage.

[0042] The overcurrent protection unit prevents the system from experiencing overcurrent by monitoring the power supply output current in real time. When the current value exceeds the safe operating range, the overcurrent protection unit will handle it by current limiting, disconnecting the power supply output, or other means. The working principle of the overcurrent protection unit is to connect to the current sensor to continuously detect the power supply output current and perform protection operations according to the set current threshold. The current threshold can be adjusted according to the load requirements of different application scenarios to ensure that the power supply can respond in a timely manner when the load changes.

[0043] The over-temperature protection unit monitors the temperature changes of the power supply and its components through a temperature sensor. When the internal or external temperature of the power supply exceeds the preset safety threshold, the over-temperature protection unit will prevent system damage caused by overheating by taking measures such as reducing power output, automatically shutting down, or other cooling measures. Over-temperature protection is usually based on the temperature data collected by the temperature sensor. When the temperature reaches or exceeds the set safety threshold, the protection mechanism will be immediately activated to ensure that the system does not malfunction due to overheating.

[0044] The function of the short-circuit protection unit is to monitor the short-circuit state of the power supply output port in real time. When a short-circuit fault occurs in the power supply system, the short-circuit protection unit will immediately identify and trigger a protection action, immediately disconnect the power supply output, and prevent damage to the power supply and load caused by excessive current. The short-circuit protection unit usually adopts current mutation detection technology. When the output current changes beyond the normal range, the system will judge whether there is a short-circuit fault and quickly take corresponding protection measures.

[0045] With the coordinated operation of all protection units, the power supply system can take appropriate protection measures in a timely manner under any abnormal conditions to avoid potential faults. The multi-element coordinated protection module in this embodiment does not rely solely on a single protection mechanism, but through the linkage of multiple protection units, jointly ensuring the safety and stability of the power supply system. This coordinated effect enables the system to provide more comprehensive protection when facing complex and changing working environments.

[0046] In the control logic of the multi-element coordinated protection module, when multiple monitoring data exceed the set thresholds, the module can execute protection actions according to the priorities. For example, when the current exceeds the set threshold and the power supply temperature is also too high, the protection module will judge the optimal protection plan according to the preset strategy. The priority of this plan may be dynamically adjusted according to load requirements, power supply working modes, and other parameters to ensure that the power supply system is always in the best working state.

[0047] The protection actions of this module are not limited to simple power-off operations, but can also be adjusted more precisely according to actual needs. For example, in over-voltage protection, the system can reduce the voltage by automatically adjusting the voltage regulator according to the type of power supply and load requirements, rather than directly cutting off the power, to avoid unnecessary shutdown of the system.

[0048] By closely collaborating with the intelligent decision-making and learning module, the multi-element coordinated protection module can provide more accurate and flexible protection during the operation of the power supply system. For example, the intelligent decision-making and learning module dynamically analyzes the working state of the power supply system based on real-time data and transmits protection strategies. The multi-element coordinated protection module adjusts the parameters or action types of the protection mechanism in real time according to these strategies to ensure that the system can operate efficiently and safely under different working conditions.

[0049] Optimization algorithm module, which receives the real-time data from the power supply monitoring module and the optimization strategy from the intelligent decision-making and learning module, and adaptively optimizes the control parameters of the power supply system based on the genetic algorithm or the particle swarm optimization algorithm; In this embodiment, the optimization algorithm module adaptively optimizes the control parameters of the power supply system by receiving the real-time data from the power supply monitoring module and the optimization strategy output by the intelligent decision-making and learning module, using an advanced optimization algorithm to improve the working efficiency, stability and overall performance of the system. The core task of this module is to adjust the power supply working parameters according to the real-time state of the system to ensure the efficient and stable operation of the power supply under different working conditions.

[0050] The optimization algorithm module mainly includes two optimization algorithm units: the genetic algorithm and the particle swarm optimization algorithm. Each algorithm has its unique advantages in different scenarios, so the combination of the two can effectively improve the global and local accuracy of the optimization process.

[0051] The genetic algorithm unit is based on the principles of natural selection and genetics, and solves optimization problems by simulating the biological evolution process. Its working principle can be described through the following steps: Initializing the population: First, the genetic algorithm generates a set of random candidate solutions. Each candidate solution (individual) is represented as a solution vector, which contains a set of control parameters of the power supply system (such as working frequency, voltage, current, etc.).

[0052] Selection operation: According to a certain fitness function (usually the operating efficiency or stability of the power supply system), select the individuals with higher fitness for "reproduction". The fitness function will be dynamically adjusted according to the real-time data of the system and the optimization strategy provided by the intelligent decision-making and learning module.

[0053] Crossover operation: Perform a crossover operation on the selected individuals, simulating gene exchange in genetics. Through the crossover operation, new candidate solutions are generated to explore a wider solution space.

[0054] Mutation operation: Mutate some individuals, simulating the process of gene mutation to avoid the algorithm falling into a local optimal solution. The introduction of the mutation operation enhances the exploration ability of the algorithm, enabling the system to discover potential global optimal solutions.

[0055] Update operation: Through multiple generations of evolution, gradually improve the fitness of the candidate solutions, and finally obtain an optimal solution, which represents the best combination of control parameters of the power supply system.

[0056] The mathematical model of the genetic algorithm can be expressed through the fitness function. Assuming the fitness function of the system is , where is a solution vector of control parameters, and the genetic algorithm iteratively optimizes ,Finally, the optimal solution is found: ; Among them, represents the optimal control parameters obtained by genetic algorithm optimization; is the fitness function of the power system.

[0057] The particle swarm optimization algorithm unit is based on the theory of swarm intelligence and simulates the process of bird flocks foraging or fish schools swimming. The basic idea of particle swarm optimization is to jointly search for the optimal solution through the cooperation and competition of each particle in the swarm. Each particle represents a solution, and its position and velocity represent the control parameters and control speed of the power system respectively.

[0058] The working principle of particle swarm optimization is achieved through the following steps: Initializing the particle swarm: Similar to the genetic algorithm, the particle swarm optimization algorithm first randomly initializes the particle swarm, and each particle corresponds to a solution vector (power control parameters). The size of the particle swarm is usually set according to the complexity of the specific problem.

[0059] Updating velocity and position: Each particle updates its position and velocity according to its current velocity, individual historical optimal position, and the global historical optimal position of the swarm. The velocity update formula for the particle is: ; Among them, represents the velocity of particle in the generation; is the inertia weight; and are learning factors; and are random numbers; is the historical optimal position of particle ; is the global optimal position of the particle swarm.

[0060] Position update: After each particle updates its velocity, it updates its position, that is, the control parameter value: ; Among them, represents the position of particle in the generation (iteration), that is, the updated control parameter value; is the position of particle in the generation, that is, the current control parameter value; is the position of particle in the The speed in the generation is calculated through the speed update formula and is used to determine the variation amplitude and direction of the current position.

[0061] Optimal solution selection: Each particle evaluates the fitness of its current position and compares it with its historical optimal position. If the fitness of the current position is better, the individual's optimal position is updated. Through continuous iteration, the particle swarm finally converges to the global optimal solution.

[0062] The goal of particle swarm optimization is to find an optimal combination of control parameters such that the performance of the power supply system (such as efficiency, stability, etc.) reaches the optimal: ; wherein, represents the best control parameters optimized by the genetic algorithm; is the fitness function of the power supply system.

[0063] During the optimization process, the genetic algorithm and the particle swarm optimization algorithm can complement each other. The genetic algorithm mainly avoids local optima through global search, while the particle swarm optimization pays more attention to the accuracy and fast convergence of local search. Through the combination of the two, the optimization algorithm module can more effectively find the best control parameters of the power supply system.

[0064] The output of the optimization algorithm module is the optimized control parameters, which are passed to the power control unit to adjust the working state of the power supply in real time. The optimization results include but are not limited to adjusting key parameters such as the output current, voltage, and working frequency of the power supply to ensure the best performance of the power supply system under different loads and environmental conditions.

[0065] The hardware acceleration and computing module is used to accelerate the computing processes of the intelligent decision-making and learning module and the optimization algorithm module; In this embodiment, the core function of the hardware acceleration and computing module is to improve the computing efficiency of the intelligent decision-making and learning module and the optimization algorithm module to ensure that the system can operate efficiently and stably in real-time data processing and complex computing tasks. This module adopts hardware acceleration technology, especially using programmable logic devices (FPGA) and dedicated artificial intelligence chips (such as AI processors or tensor processing units TPU) to accelerate the key computing tasks in the system and ensure the minimization of the system response time.

[0066] The design of the hardware acceleration and computing module depends on targeted optimization for computing bottlenecks. Especially in the processing of deep learning and optimization algorithms, traditional software processing often cannot meet the efficient and real-time computing requirements. Therefore, this embodiment introduces efficient hardware acceleration technology, enabling the decision-making and optimization processes of the power supply system to be executed quickly.

[0067] The FPGA acceleration unit is a key part of the hardware acceleration module and is particularly suitable for parallel computing tasks. FPGA (Field Programmable Gate Array) can customize the hardware circuit according to specific tasks. Compared with the general Central Processing Unit (CPU), FPGA has significant performance advantages when executing parallel tasks. FPGA can accelerate machine learning algorithms, optimization algorithms, and signal processing tasks at the hardware level, thus achieving faster computing and response. Through the hardware acceleration of FPGA, the system can process more data streams and quickly complete complex computing tasks. Especially during the iterative process of optimization algorithms, FPGA can greatly improve the system processing speed.

[0068] The working principle of FPGA acceleration is mainly reflected in the parallel processing of data streams. Since FPGA can design specific hardware circuits according to specific algorithms, it can process multiple data channels simultaneously, greatly improving the efficiency of data processing. In deep learning and optimization algorithms, the parallel processing of data is particularly important. FPGA can perform a large number of matrix multiplications and convolution operations in parallel, which is crucial for the training and optimization process of neural networks. Through the customized hardware circuit, FPGA can provide high-throughput and low-latency computing capabilities, thus meeting the strict requirements for real-time and efficiency in power systems.

[0069] Another important hardware acceleration unit is the AI chip, especially the dedicated processing unit for deep learning, such as the Tensor Processing Unit (TPU). TPU is a hardware accelerator designed specifically to accelerate deep neural network computing tasks and has significant advantages in tasks such as matrix calculations and convolution operations. The AI chip can perform a large number of parallel calculations quickly through a dedicated hardware architecture, thus accelerating the inference and training processes of neural networks. In the intelligent decision-making and learning module, the AI chip can accelerate the inference process of neural networks, help the system process a large amount of input data in real time, and output the optimal power control parameters.

[0070] The working principle of the AI chip is based on its dedicated hardware architecture, which can perform highly optimized parallel calculations for each layer and each neuron in the neural network. For example, during the inference process of a Convolutional Neural Network (CNN), the AI chip can perform a large number of convolution operations in parallel, reducing the computing time and accelerating the processing of data streams. The mathematical model of this process can be expressed as: ; where, is the weight matrix; is the input data; is the bias term; is the activation function. The AI chip can accelerate these mathematical operations to ensure optimized power control parameters are obtained in a short time.

[0071] In addition, the hardware acceleration and computing module also includes other dedicated computing units, such as digital signal processors (DSPs) and graphics processing units (GPUs). These units can efficiently process data according to specific tasks, further improving the computing power and response speed of the system. In some specific application scenarios, DSPs and GPUs can process high-frequency signals and image data to optimize the operating efficiency of the power system.

[0072] The hardware acceleration and computing module works closely with the intelligent decision-making and learning module and the optimization algorithm module. The reinforcement learning and neural network algorithms used in the intelligent decision-making and learning module usually require a large amount of computing, especially during real-time optimization and policy adjustment. Through hardware acceleration, FPGAs and AI chips can significantly improve the execution speed of the algorithms, ensuring that the intelligent decision-making and learning module can make optimization decisions in real time based on power monitoring data. Similarly, the genetic algorithm and particle swarm optimization algorithm in the optimization algorithm module can also be rapidly iterated through hardware acceleration, thus shortening the time of the optimization process.

[0073] The design of the hardware acceleration and computing module also takes into account data transmission and collaborative work between modules. Data exchange between modules is carried out through a high-speed bus to ensure that data can be quickly transmitted from the monitoring module to the decision-making module, then to the optimization module, and finally to the control unit for operation. The high-speed data bus and efficient data stream processing mechanism ensure the efficient operation of the system, enabling the power system to respond in a timely manner and make optimization adjustments in a complex and dynamic working environment.

[0074] The cloud platform and remote monitoring module upload the real-time data of the power monitoring module and the intelligent decision-making and learning module to the cloud platform through Internet of Things technology for remote monitoring, data analysis, and global optimization, and at the same time feedback the optimization results and protection strategies to the power system; In this embodiment, the cloud platform and remote monitoring module upload the real-time monitoring data, optimization results, and operating status of the power system to the cloud platform through Internet of Things technology, thereby realizing centralized management, real-time monitoring, and global optimization of the power system. This module can not only provide the remote management function for the power system, but also deeply optimize the system through data analysis and cloud computing technology, further improving the reliability, efficiency, and operability of the power system.

[0075] The cloud platform and the remote monitoring module consist of two main parts: the data upload unit and the remote monitoring and management unit. The main function of the data upload unit is to upload the real-time data generated by the power monitoring module and the intelligent decision-making and learning module to the cloud platform through the network, ensuring that all monitoring data and system status information can be transmitted and stored in the cloud in real time. The remote monitoring and management unit, through the Web interface or mobile application, allows users to access the cloud platform, view the operating status, historical data, and alarm information of the power system in real time, and at the same time provides remote adjustment and control functions.

[0076] The data upload unit connects the power system and the cloud platform through Internet of Things technology. To ensure the stability and real-time nature of data transmission, this unit uses efficient data transmission protocols such as MQTT (Message Queuing Telemetry Transport) or CoAP (Constrained Application Protocol). These protocols can ensure that data is uploaded to the cloud with minimal latency, guaranteeing the response speed of the system. The uploaded data includes information such as the voltage, current, temperature, environmental parameters, system operating status, and protection actions of the power supply. In addition, the uploaded data also includes the optimization results and control parameters from the intelligent decision-making and learning module and the optimization algorithm module. These data provide an important basis for subsequent analysis, monitoring, and optimization.

[0077] Through real-time data upload, the cloud platform can centrally monitor the power system. The cloud platform is not only a data storage and display platform, but also through cloud computing and big data analysis technologies, it can process, analyze, and predict data, thus providing a global optimization strategy for the power system. During the operation of the power system, the cloud platform will conduct real-time analysis based on the data from the power monitoring module and the decision-making module, timely detect potential faults or performance degradation, and provide optimization suggestions. In this way, the cloud platform can help users comprehensively understand the operating status of the power system and make corresponding adjustments and optimizations.

[0078] The remote monitoring and management unit provides a convenient operation interface for users through the Web interface or mobile application. Users can view the operating status of the power system, real-time monitoring data, alarm information, and historical data of the system through these interfaces. Users can take timely measures according to the alarm information of the system to prevent the further expansion of faults. For example, if a certain parameter of the power system exceeds the safe range, the system will automatically generate an alarm message through the cloud platform and display it to the user through the interface. Users can view the detailed alarm information and select the corresponding processing method, such as manually adjusting the power parameters or starting a certain protection mechanism.

[0079] In terms of remote management, users can not only view the system status, but also adjust system parameters through the cloud platform, such as control parameters like the output voltage and current of the power supply. The realization of this function is through the connection between the remote control module and the power control unit, and users can operate via the Internet anywhere, greatly improving the flexibility and manageability of the system.

[0080] The cloud platform and the remote monitoring module also have data analysis and prediction functions. By analyzing the historical data uploaded to the cloud platform, the cloud platform can provide users with health status reports, efficiency analysis, and future operation trend predictions of the power supply system. The cloud platform can also, through big data algorithms, based on historical and real-time data, predict possible faults and performance degradation of the power supply system, so as to take preventive measures in advance. The prediction results can help users perform maintenance before the power supply system fails, reducing the likelihood of failures and enhancing the reliability and stability of the system.

[0081] In addition, the cloud platform globally optimizes the power supply system through cloud computing technology. By summarizing and analyzing the data of all power supply systems, the cloud platform can dynamically adjust the working strategies of each power supply according to the mutual relationship between different power supplies, load distribution, and environmental conditions to achieve the optimal performance of the entire system. For example, when the load of a certain power supply is high, the cloud platform can automatically adjust the working parameters of this power supply according to the optimization algorithm, or transfer the load to other power supplies to improve the overall efficiency of the system and extend the service life of the power supply.

[0082] To ensure data security and privacy, the cloud platform and the remote monitoring module adopt high-standard encryption technologies and authentication mechanisms. All uploaded data is transmitted through encryption to ensure that the data is not tampered with or stolen. In addition, the remote operation permissions of users are managed through identity verification and permission control to ensure that only authorized users can access and control the power supply system.

[0083] Please refer to the appendix Figure 8 , the present invention also provides a switching power supply circuit method based on intelligent monitoring and protection, and the method includes the following steps: S1. Real-time collect the output voltage, current, temperature, and environmental data of the power supply; S2. Input the collected monitoring signals into the intelligent decision-making and learning module, and optimize the working state of the power supply through reinforcement learning and neural network algorithms; S3. According to the optimization results, trigger protection operations in real-time through the multi-element collaborative protection module to ensure the safety of the power supply; S4. Optimize the working efficiency of the power supply through the optimization algorithm module, and adjust the control parameters to improve the system performance; S5. Accelerate the calculation process through the hardware acceleration and computing module to ensure the real-time response of the system; S6. Upload the monitoring data and optimization results to the cloud platform through Internet of Things technology for remote monitoring and data analysis.

[0084] The method of this embodiment can be used to implement the above method embodiment, and its principle and technical effects are similar, so they will not be elaborated here.

[0085] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A switching power supply circuit system based on intelligent monitoring and protection, characterized in that, Including: A power supply monitoring module for collecting voltage, current, temperature, and environmental data of the power supply system in real time; An intelligent decision-making and learning module that receives the voltage, current, temperature, and environmental data transmitted by the power supply monitoring module, analyzes the working state of the power supply system based on the data through reinforcement learning and neural network algorithms, and optimizes the operating parameters of the power supply to generate protection and optimization strategies; A multi-element collaborative protection module that receives the protection and optimization strategies output by the intelligent decision-making and learning module, and based on real-time monitoring data and decision results, performs overvoltage, overcurrent, overtemperature, and short-circuit protection operations in real time; An optimization algorithm module that receives the real-time data of the power supply monitoring module and the optimization strategies of the intelligent decision-making and learning module, and adaptively optimizes the control parameters of the power supply system based on genetic algorithms or particle swarm optimization algorithms; A hardware acceleration and computing module for accelerating the computing processes of the intelligent decision-making and learning module and the optimization algorithm module; A cloud platform and remote monitoring module that uploads the real-time data of the power supply monitoring module and the intelligent decision-making and learning module to the cloud platform through Internet of Things technology for remote monitoring, data analysis, and global optimization, and at the same time feeds back the optimization results and protection strategies to the power supply system.

2. The switching power supply circuit system based on intelligent monitoring and protection according to claim 1, wherein, The power supply monitoring module includes: A voltage monitoring unit for collecting the power supply output voltage signal in real time and transmitting it to the intelligent decision-making and learning module; A current monitoring unit for detecting the power supply output current in real time through a Hall sensor or a shunt resistor and transmitting it to the intelligent decision-making and learning module; A temperature monitoring unit for monitoring the internal temperature of the power supply system and transmitting it to the intelligent decision-making and learning module; An environmental monitoring unit for collecting environmental data, including humidity and air pressure, and transmitting it to the intelligent decision-making and learning module.

3. The switching power supply circuit system based on intelligent monitoring and protection according to claim 1, wherein The intelligent decision-making and learning module includes: A reinforcement learning unit for optimizing the power supply protection threshold and working strategy according to the monitoring signal through a reinforcement learning algorithm; A neural network unit for adjusting the power supply working state based on historical data and real-time monitoring signals; The update formula of the reinforcement learning algorithm is: ； Among them, represents the value of taking action in state ; is the learning rate; is the discount factor; is the immediate reward; represents the return value of choosing the optimal action in the next state.

4. The switching power supply circuit system based on intelligent monitoring and protection according to claim 1, characterized in that, The multi-element collaborative protection module includes: An overvoltage protection unit for triggering an overvoltage protection operation when the power supply output voltage exceeds the set threshold; An overcurrent protection unit for triggering an overcurrent protection operation when the power supply output current exceeds the set value; An overtemperature protection unit for triggering an overtemperature protection operation when the internal temperature of the power supply exceeds the set value; A short-circuit protection unit for immediately cutting off the power supply output when a short circuit occurs.

5. The switching power supply circuit system based on intelligent monitoring and protection according to claim 1, wherein The optimization algorithm module includes: A genetic algorithm unit for optimizing the control parameters of the power supply through a genetic algorithm; A particle swarm optimization unit for dynamically adjusting the power supply working frequency and PWM control signal through a particle swarm optimization algorithm; The speed update formula of the particle swarm optimization algorithm is: ； Among them, represents the velocity of the particle in the generation; is the inertia weight; and are the learning factors; and are random numbers; is the historical optimal position of the particle ; is the global optimal position of the particle swarm.

6. The switch power supply circuit system based on intelligent monitoring and protection according to claim 1, characterized in that, The hardware acceleration and computing module includes: An FPGA acceleration unit for accelerating the computing of machine learning algorithms and optimization algorithms; An AI chip acceleration unit for accelerating the execution of neural network models.

7. The switch power supply circuit system based on intelligent monitoring and protection according to claim 1, characterized in that, The platform cloud and remote monitoring module includes: A data uploading unit for transmitting the real-time data of the power supply system to the cloud platform through Internet of Things technology; Remote monitoring and management unit, used to remotely monitor, diagnose faults and adjust the system through a Web interface or a mobile application.

8. The switch power supply circuit system based on intelligent monitoring and protection according to claim 1, characterized in that, The intelligent decision-making and learning module adjusts the power protection threshold in real time through a reinforcement learning algorithm to adapt to different workloads and environmental conditions.

9. The switch power supply circuit system based on intelligent monitoring and protection according to claim 1, characterized in that, When the power load changes, the optimization algorithm module adjusts the control parameters in real time to maximize power efficiency and stability. The fitness function optimization formula of the genetic algorithm is: ； Among them, represents the optimal control parameters obtained by genetic algorithm optimization; is the fitness function of the power supply system.

10. A switching power supply circuit method based on intelligent monitoring and protection, applied to the switching power supply circuit system based on intelligent monitoring and protection according to any one of claims 1-9, characterized in that, The method includes the following steps: S1. Real-time collect the power output voltage, current, temperature and environmental data; S2. Input the collected monitoring signals into the intelligent decision-making and learning module, and optimize the power working state through reinforcement learning and neural network algorithms; S3. According to the optimization results, trigger protection operations in real time through the multi-element collaborative protection module to ensure the safety of the power supply; S4. Optimize the power working efficiency through the optimization algorithm module and adjust the control parameters to improve the system performance; S5. Accelerate the calculation process through the hardware acceleration and computing module to ensure the real-time response of the system; S6. Upload the monitoring data and optimization results to the cloud platform through Internet of Things technology for remote monitoring and data analysis.

Images