A soft start method and system for inverter

Abstract

The present invention proposes a soft start method and system for an inverter. It belongs to the field of power electronics technology. After the inverter is powered on, a hardware self-check is performed, and the voltage, current and external power supply status of the DC side are monitored through sensors; and the startup configuration parameters and preset strategy library stored in the non-volatile memory are read and analyzed; before starting, a preliminary judgment is made on the load type and size through a small current pre-start or voltage detection method, and an initial startup strategy is selected from the preset strategy library according to the load characteristics. Through small current pre-start or voltage detection, the large current impact at the startup moment is avoided, the inverter and the connected load are protected, and the life of the equipment is extended.

Description

Technical Field

[0001] The invention provides a soft starting method and system for an inverter, belonging to the technical field of power electronics. Background Art

[0002] When the off-grid system inverter is started, due to the direct connection between the DC power supply and the load, a large current surge is often generated, which will not only damage the power devices inside the inverter, but also may affect the stable operation of the entire system. Traditional soft-start methods mostly rely on hardware circuits (such as starting resistors) to achieve current limitation, but these methods have the disadvantages of high cost, large size and low efficiency. Therefore, it is particularly important to develop an efficient, intelligent soft-start method that does not require additional hardware. Summary of the invention

[0003] The present invention provides a soft start method and system for an inverter, which are used to solve the problems mentioned in the above background technology:

[0004] The present invention provides a soft starting method for an inverter, the method comprising:

[0005] S1. After the inverter is powered on, it performs a hardware self-check and monitors the DC side voltage, current and external power supply status through sensors; and reads and analyzes the startup configuration parameters and preset strategy library stored in the non-volatile memory;

[0006] S2. Before starting, make a preliminary judgment on the load type (resistive, inductive, capacitive, etc.) and size through small current pre-start or voltage detection, and select the initial startup strategy from the preset strategy library according to the load characteristics;

[0007] S3, using a lower PWM duty cycle and a slow rise rate, gradually increasing the inverter output voltage and current, and monitoring the changes in output voltage, current, DC side voltage and load current in real time through closed-loop control;

[0008] S4. When the output voltage and current reach the preset stability threshold and the fluctuation does not exceed the preset range within the preset time, it is judged that the system has entered a stable state; a switching signal is sent to switch the inverter from the soft start mode to the normal working mode and adjust it to the optimal operating parameters;

[0009] S5. Record key parameters and events during the entire startup process. When startup fails or an abnormal situation is detected, trigger a fault alarm and try to perform recovery operations or automatically restart.

[0010] Furthermore, the S1 includes:

[0011] S11, verifying the input voltage, determining whether the input voltage is within an allowable range, and checking the working state of the power conversion circuit;

[0012] S12, perform a switch test on the IGBT module through a built-in test program, and detect the capacitance and inductance;

[0013] S13, calibrating the voltage, current and temperature sensors, and verifying the integrity and reliability of the communication interface by sending test signals and receiving responses;

[0014] S14, monitor the DC bus voltage in real time, and predict the energy reserve situation, analyze the DC side current fluctuation based on the prediction result, and identify potential power supply fluctuation or load change based on the analysis result;

[0015] S15, monitor the output voltage and current of the solar panel, calculate the MPPT efficiency, and predict the energy output according to weather conditions;

[0016] S16. Setting temperature sensors inside and outside the inverter to monitor temperature changes in real time;

[0017] S17, reading the configuration parameters of the last successful startup from the non-volatile memory, parsing the preset strategy library, and understanding the load type, environmental conditions, and advantages and disadvantages applicable to each strategy;

[0018] S18, preliminarily selecting a startup strategy based on current environmental conditions, historical data, and load expectations;

[0019] S19. Before selecting a strategy, conduct simulation tests to verify the feasibility and effectiveness of the launch strategy, and make fine adjustments as needed.

[0020] Furthermore, the S14 includes:

[0021] The ADC continuously monitors the DC bus voltage, captures any instantaneous voltage fluctuations, removes noise through a digital filtering algorithm, and calculates the effective value, peak value, and fluctuation range of the voltage in real time;

[0022] Dynamically adjust the upper and lower limits of the safe operating range of the DC bus voltage according to the current operating status of the system, and use the machine learning model to predict the optimal safety threshold;

[0023] Integrated energy management system, combining historical energy consumption data, weather forecast and load forecast, and using deep learning algorithms to predict energy demand and supply in the future;

[0024] According to the prediction results, the charging and discharging strategy of the energy storage device is intelligently scheduled, and the cycle life of the energy storage device is optimized;

[0025] Analyze the DC side current waveform through Fourier transform, identify the harmonic components and their sources, and evaluate their impact on system stability and efficiency;

[0026] Through the time-frequency analysis algorithm, the time-varying characteristics of current fluctuations are monitored in real time, and potential power supply fluctuations or load mutations can be quickly responded to;

[0027] Based on the existing machine learning anomaly detection model, the DC bus voltage and current data are continuously monitored to identify abnormal patterns and provide early warning of potential power failures or load anomalies.

[0028] Based on the early warning information, preventive maintenance tasks are automatically triggered.

[0029] Furthermore, the S19 includes:

[0030] S191. Use digital twin technology to build a high-precision virtual model of the inverter and its working environment, and reproduce various scenarios during the inverter startup process based on the virtual environment;

[0031] S192, inputting the initially selected startup strategy into the virtual test environment to simulate the startup process of the inverter under the guidance of the strategy;

[0032] S193. Real-time monitoring of various performance indicators of the virtual inverter during the startup process, comprehensive evaluation of the results of the simulation operation, and analysis of the gap between the actual effect of the startup strategy and the expected target;

[0033] S194. Use big data analysis to identify performance bottlenecks and potential problem areas during the startup process, and make fine adjustments to key parameters in the startup strategy based on performance evaluation results;

[0034] S195, automatically searching for the optimal parameter combination through an intelligent optimization algorithm, and inputting the adjusted startup strategy into the virtual test environment again for iterative testing;

[0035] S196. Repeat the above steps S192 to S195 until the performance index of the startup strategy reaches or exceeds the preset standard.

[0036] Furthermore, the S2 includes:

[0037] S21, start the inverter with a small current, observe the initial response characteristics of the load, and preliminarily distinguish the load type by analyzing the shape and phase difference of the current response waveform;

[0038] S22, combining data from multiple pre-start cycles, making a comprehensive judgment on the load type, gradually increasing the current and monitoring voltage changes during the pre-start process, and analyzing the current-voltage relationship curve;

[0039] S23, calculating the equivalent impedance or power demand of the load according to the current-voltage relationship curve, and correcting the estimation result based on the dynamic changes of the load;

[0040] S24, matching a startup strategy from a preset strategy library according to the load type and size; fine-tuning the parameters of the selected strategy according to the load characteristics and system requirements;

[0041] S25. Verify the adjusted strategy through simulation software and evaluate its impact on system stability and efficiency.

[0042] Furthermore, the S24 includes:

[0043] On the basis of preliminary classification of load types, the detailed characteristics of the load are further analyzed, and the historical load data is learned through machine learning algorithms to establish a load characteristic model;

[0044] According to load characteristics and system requirements, a matching startup strategy is retrieved from the preset strategy library;

[0045] Perform preliminary parameter setting on the retrieved startup strategy, and use adaptive control theory to preliminarily set parameters based on load characteristics and system status;

[0046] Predict the dynamic changes of load during the startup process and dynamically adjust the startup strategy based on the prediction results;

[0047] Based on the load characteristics and system requirements, fine-tune the parameters of the selected strategy, and use simulation tools to test the system performance under different parameter combinations;

[0048] Global optimization of strategy parameters through multi-objective optimization algorithm;

[0049] During the startup process, the load characteristics and system status are monitored in real time, and the startup strategy is adjusted in real time based on the monitoring results.

[0050] Furthermore, the S3 includes:

[0051] S31, selecting an initial PWM duty cycle, and gradually increasing the PWM duty cycle in a linear or exponential growth manner;

[0052] S32, collecting output voltage, current, DC side voltage and load current data in real time through high-precision sensors, filtering and denoising the collected data, and extracting information;

[0053] S33, using a control algorithm to perform closed-loop control on the inverter output; dynamically adjusting the PWM duty cycle and rise rate according to the changes in voltage, current and load current monitored in real time;

[0054] S34. During the startup process, the dynamic changes of the load are continuously evaluated and the control strategy is adjusted accordingly;

[0055] S35. Set up an abnormality detection mechanism and take corresponding measures immediately once an abnormal situation is detected.

[0056] Furthermore, the S4 includes:

[0057] S41, setting the stable threshold and fluctuation range of the output voltage and current, monitoring the changes of the output voltage and current in real time within a preset time window, and calculating the fluctuation range;

[0058] S42, comparing the real-time monitored data with the set stability threshold to determine whether the system has entered a stable state;

[0059] S43, if it is confirmed that the system has entered a stable state, immediately send a switching signal to switch the inverter from the soft start mode to the normal working mode;

[0060] S44. Under normal operating mode, further optimize and adjust the operating parameters of the inverter according to load characteristics and system requirements; and evaluate the performance of the inverter.

[0061] Furthermore, the S5 includes:

[0062] S51. During the startup process, detailed information of key parameters and events is recorded in detail;

[0063] S52, save the recorded data to a non-volatile memory and perform regular backup.

[0064] S53. During the startup process or in the normal working mode, once an abnormal situation or fault is detected, the fault alarm mechanism is immediately triggered and the user or technician is notified through multiple channels;

[0065] S54, diagnose and locate the fault according to the fault alarm information, combined with the startup record and log analysis, and perform corresponding recovery operations or automatically restart the inverter according to the fault diagnosis result;

[0066] S55. Record the fault information, diagnosis results and recovery operations in the fault report for subsequent analysis and improvement.

[0067] The present invention provides a soft start system for an inverter, the system comprising:

[0068] Self-check module: After the inverter is powered on, it performs hardware self-check and monitors the DC side voltage, current and external power supply status through sensors; and reads and analyzes the startup configuration parameters and preset strategy library stored in the non-volatile memory;

[0069] Selection module: Before starting, the load type (resistance, inductance, capacitance, etc.) and size are preliminarily judged through low current pre-start or voltage detection, and the initial startup strategy is selected from the preset strategy library according to the load characteristics;

[0070] Monitoring module: Uses a lower PWM duty cycle and a slow rise rate to gradually increase the inverter output voltage and current. Through closed-loop control, it monitors the changes in output voltage, current, DC side voltage and load current in real time.

[0071] Adjustment module: When the output voltage and current reach the preset stability threshold and the fluctuation does not exceed the preset range within the preset time, it is judged that the system has entered a stable state; a switching signal is sent to switch the inverter from the soft start mode to the normal working mode and adjust it to the optimal operating parameters;

[0072] Recording module: records key parameters and events during the entire startup process. When startup fails or an abnormal situation is detected, it triggers a fault alarm and attempts to perform recovery operations or automatically restart.

[0073] Beneficial effects of the present invention: Through small current pre-start or voltage detection, the large current shock at the moment of startup is avoided, the inverter and the connected load are protected, and the life of the equipment is extended. Through detailed load type judgment and the use of preset strategy library, the most suitable startup strategy can be selected according to the specific load characteristics, which increases the success rate of startup. Closed-loop control and real-time monitoring ensure the smooth transition of output voltage and current, so that the system can reach a stable state with minimal disturbance, and improve the overall system stability. The fault alarm and recovery mechanism can detect and handle problems in time, reduce the downtime caused by faults, and improve the reliability and availability of the system. Strategy adjustment based on data analysis and machine learning can dynamically optimize operating parameters so that the inverter can operate efficiently under different conditions. Through the analysis and prediction of historical data, potential problems can be identified in advance, preventive measures can be taken, and the risk of sudden failures can be reduced. Through digital twins and machine learning algorithms, the inverter can self-learn and optimize the startup strategy, which improves the intelligent level of control. Detailed recording and analysis functions are helpful for troubleshooting and system optimization, and provide a basis for improvement and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 It is a step diagram of the method of the present invention;

[0075] Figure 2This is a system module diagram of the present invention. DETAILED DESCRIPTION

[0076] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0077] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. The embodiments described 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0078] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0079] One embodiment of the present invention, as Figure 1 As shown, a soft start method for an inverter, the method comprising:

[0080] S1. After the inverter is powered on, a hardware self-check is performed to ensure that each module is normal. At the same time, the DC side voltage, current and external power supply status (such as solar panel output) are monitored through sensors; and the startup configuration parameters and preset strategy library stored in the non-volatile memory are read and analyzed;

[0081] S2. Before starting, make a preliminary judgment on the load type (resistance, inductance, capacitance, etc.) and size through small current pre-start or voltage detection, and select the appropriate initial startup strategy from the preset strategy library according to the load characteristics;

[0082] S3, using a lower PWM duty cycle and a slow rise rate, gradually increasing the inverter output voltage and current, and monitoring the changes in output voltage, current, DC side voltage and load current in real time through closed-loop control;

[0083] S4. When the output voltage and current reach the preset stability threshold and the fluctuation does not exceed the preset range within the preset time, it is judged that the system has entered a stable state; a switching signal is sent to switch the inverter from the soft start mode to the normal working mode and adjust it to the optimal operating parameters;

[0084] S5. Record key parameters and events during the entire startup process. When startup fails or an abnormal situation is detected, trigger a fault alarm and try to perform recovery operations or automatically restart.

[0085] The working principle of the above technical solution is as follows: after the inverter is powered on, it first performs a hardware self-check. This step is to ensure that all modules of the inverter (such as control module, power module, sensor, etc.) are in normal working state and there is no hardware failure. At the same time, the inverter monitors the voltage and current of the DC side and the output status of the external power supply (such as solar panels) in real time through sensors to ensure the stability and reliability of the input power supply. In addition, the inverter will also read and parse the startup configuration parameters and preset strategy library stored in the non-volatile memory. These parameters and strategies will guide the subsequent startup process. Before the formal startup, the inverter uses a small current pre-start or voltage detection method to make a preliminary judgment on the load. By monitoring the response characteristics of the load during the pre-start process (such as the change trend of voltage and current), the inverter can roughly judge the type (resistance, inductance, capacitance, etc.) and size of the load. Based on this judgment result, the inverter selects the most appropriate initial startup strategy from the preset strategy library. This step is crucial to ensure that the inverter can smoothly transition to a normal working state during the startup process. The inverter uses a lower PWM (pulse width modulation) duty cycle and a slow rise rate to gradually increase the output voltage and current. This gradual startup method helps to reduce the impact current and voltage fluctuations during the startup process, protecting the inverter and the load from damage. At the same time, the inverter monitors the changes in output voltage, current, DC side voltage and load current in real time through closed-loop control, and adjusts the startup strategy in real time according to these feedback signals. This dynamic adjustment mechanism ensures that the inverter can quickly adapt to load changes and maintain stable operation during the startup process. When the output voltage and current reach the preset stability threshold and the fluctuation does not exceed the preset range within the preset time, the inverter determines that the system has entered a stable state. At this time, the inverter will send a switching signal to switch itself from soft start mode to normal working mode and adjust to the optimal operating parameters to meet the load requirements. This step marks the end of the inverter startup process and the beginning of normal working state. During the entire startup process, the inverter will record key parameters and events for subsequent analysis and optimization. If the startup fails or an abnormal situation is detected (such as overload, short circuit, overheating, etc.), the inverter will trigger a fault alarm and try to perform recovery operations or automatically restart. This fault handling mechanism helps to detect and solve problems in a timely manner, ensuring the reliability and stability of the inverter.

[0086] The effects of the above technical solution are: ensuring that all modules of the inverter are in normal state before power-on, avoiding startup failure or damage caused by hardware failure; effectively reducing the impact current during the startup process through small current pre-start and gradually increasing the output voltage / current, protecting the electronic components inside the inverter and the external load, and avoiding damage caused by instantaneous large current; selecting a suitable startup strategy through preliminary judgment of the load, so that the inverter can better adapt to different types of loads and improve the success rate of startup; during the startup process, real-time monitoring and adjustment of the startup strategy through closed-loop control ensure that the inverter can quickly respond to load changes, further improving the stability and success rate of startup; after the inverter enters the normal working mode, by adjusting to the optimal operating parameters, the overall performance of the system is optimized, and the conversion efficiency and output stability of the inverter are improved; The method of increasing the output voltage and current step by step reduces unnecessary energy consumption and improves energy utilization compared to direct full-power startup; it records key parameters and events during the startup process in real time, and triggers a fault alarm when an abnormal situation is detected, which helps to find and solve problems in time and enhance the reliability and stability of the system; when the startup fails or an abnormal situation is detected, the inverter will try to perform recovery operations or automatically restart, reducing downtime caused by faults and improving system availability and reliability; the inverter transitions smoothly during the startup process, reducing voltage and current fluctuations, providing users with a more stable and reliable power supply and improving user experience; by reading and parsing the startup configuration parameters and preset strategy library stored in the non-volatile memory, the intelligent management of the inverter is realized, the need for manual intervention is reduced, and the automation level of the system is improved.

[0087] In one embodiment of the present invention, the S1 includes:

[0088] S11, verifying the input voltage, determining whether the input voltage is within an allowable range, and checking the working state of the power conversion circuit;

[0089] S12, perform a switch test on the IGBT module through a built-in test program, and detect the capacitance and inductance;

[0090] S13, calibrating the voltage, current and temperature sensors, and verifying the integrity and reliability of the communication interface by sending test signals and receiving responses;

[0091] S14, monitor the DC bus voltage in real time, and predict the energy reserve situation, analyze the DC side current fluctuation based on the prediction result, and identify potential power supply fluctuation or load change based on the analysis result;

[0092] S15, monitor the output voltage and current of the solar panel, calculate the MPPT efficiency, and predict the energy output according to weather conditions;

[0093] S16. Setting temperature sensors inside and outside the inverter to monitor temperature changes in real time;

[0094] S17, reading the configuration parameters of the last successful startup from the non-volatile memory, including voltage setting, current limit, PWM frequency, etc., parsing the preset strategy library, and understanding the load type, environmental conditions, and advantages and disadvantages of each strategy;

[0095] S18. Preliminarily select the most appropriate startup strategy based on current environmental conditions (such as light intensity, temperature), historical data, and load expectations;

[0096] S19. Before selecting a strategy, conduct simulation tests to verify the feasibility and effectiveness of the launch strategy, and make fine adjustments as needed.

[0097] The working principle of the above technical solution is as follows: the inverter first verifies whether the input voltage is within the allowed range, which is to ensure that the inverter can safely receive and process the input power. At the same time, the working status of the power conversion circuit is checked to ensure that the circuit can normally convert the input voltage to the voltage level required inside the inverter. The IGBT (insulated gate bipolar transistor) module, which is a key power conversion component in the inverter, is tested by the built-in test program. The test is to ensure that the IGBT module has no short circuit or open circuit phenomenon to ensure that it can reliably control the on and off of the current. At the same time, the capacitors and inductors are tested to determine whether they have leakage or damage signs. These components play an important role in energy storage and filtering in the inverter. The voltage, current and temperature sensors are calibrated to ensure that they can accurately measure physical quantities inside and outside the inverter. The integrity and reliability of the communication interface are verified by sending test signals and receiving responses, which is the key to data transmission between modules inside the inverter and between the inverter and the external system. The DC bus voltage is monitored in real time, which is a key parameter on the input side of the inverter. By ensuring that the DC bus voltage is within the safe operating range, inverter damage or performance degradation caused by excessively high or low voltage can be avoided. At the same time, based on the analysis of DC side current fluctuations, potential power fluctuations or load changes are predicted to provide a basis for the subsequent startup strategy selection. The output voltage and current of the solar panel are monitored, and the MPPT (maximum power point tracking) efficiency is calculated. MPPT is an optimization technology used to ensure that the solar panel outputs the maximum power under given conditions. Predicting energy output based on weather conditions (such as light intensity and temperature) helps the inverter better manage its energy conversion process. Temperature sensors are set inside and outside the inverter to monitor temperature changes in real time. Temperature is one of the important factors affecting the performance and reliability of the inverter, so monitoring it helps to detect and deal with potential thermal problems in a timely manner. The configuration parameters of the last successful startup are read from the non-volatile memory, including voltage setting, current limit, PWM frequency, etc. These parameters are the basis for the normal operation of the inverter. At the same time, the preset strategy library is parsed to understand the load type, environmental conditions and advantages and disadvantages applicable to each strategy, so as to prepare for the subsequent startup strategy selection. Based on the current environmental conditions (such as light intensity, temperature), historical data, and load expectations, the most appropriate startup strategy is initially selected. This step takes into account a variety of factors to ensure that the inverter can start smoothly under various conditions. Before selecting a strategy, a simulation test is performed to verify its feasibility and effectiveness. The simulation test can simulate various situations in the actual startup process to evaluate the performance of the startup strategy. Based on the results of the simulation test, the startup strategy is fine-tuned as necessary to ensure that it can achieve the best results in actual applications.

[0098] The effects of the above technical solution are: ensuring that the input voltage received by the inverter is within the allowable range to prevent equipment damage or performance degradation caused by excessively high or low voltage; ensuring the normal operation of the power conversion circuit to avoid safety hazards caused by circuit failures; timely discovering and handling problems such as short circuits, open circuits, leakage or damage to prevent equipment damage or safety accidents such as fire caused by these failures; ensuring accurate measurement of key parameters such as voltage, current and temperature to provide reliable data support for the stable operation of the inverter; ensuring the integrity and reliability of data transmission within the inverter and between the inverter and the external system to avoid system instability or data loss caused by communication failures; through real-time monitoring and prediction of DC bus voltage, energy reserves, solar panel output, etc., timely discovering and responding to potential power fluctuations or load changes to improve the adaptability and stability of the system; calculating according to the output voltage and current of the solar panel MPPT efficiency and predicting energy output based on weather conditions help the inverter to utilize solar energy resources more efficiently and improve the overall performance of the system; preliminarily select the most appropriate startup strategy based on current environmental conditions, historical data and load expectations, and verify its feasibility and effectiveness through simulation tests, and then make necessary fine-tuning to ensure that the inverter can start smoothly and quickly reach the optimal operating state; through a comprehensive hardware self-check and preparation process, potential problems can be discovered and handled in a timely manner, reducing downtime caused by failures, improving system availability and user satisfaction; using historical data and preset strategy libraries stored in non-volatile memory to achieve intelligent management of inverters, reduce the need for manual intervention, and improve the automation level of the system; through MPPT technology and energy output prediction, the inverter can utilize solar energy resources more efficiently, reduce dependence on traditional energy, and promote energy conservation, emission reduction and sustainable development.

[0099] In one embodiment of the present invention, the S14 includes:

[0100] The DC bus voltage is continuously monitored through the ADC to capture any instantaneous voltage fluctuations, and the noise is removed through a digital filtering algorithm (such as the Kalman filter), and the effective value, peak value and fluctuation range of the voltage are calculated in real time;

[0101] Dynamically adjust the upper and lower limits of the safe operating range of the DC bus voltage according to the current operating status of the system (such as load level and ambient temperature), and use machine learning models to predict the optimal safety threshold;

[0102] Integrated energy management system (EMS), combining historical energy consumption data, weather forecasts and load forecasts, and using deep learning algorithms to predict energy demand and supply over a period of time;

[0103] Based on the prediction results, intelligently dispatch the charging and discharging strategies of energy storage devices (such as battery packs) and optimize the cycle life of energy storage devices;

[0104] Analyze the DC side current waveform through Fourier transform (FFT), identify the harmonic components and their sources, and evaluate their impact on system stability and efficiency;

[0105] Through time-frequency analysis algorithms, such as short-time Fourier transform (STFT) or wavelet transform, the time-varying characteristics of current fluctuations can be monitored in real time to quickly respond to potential power supply fluctuations or load mutations;

[0106] Based on the existing machine learning anomaly detection model, the DC bus voltage and current data are continuously monitored to identify abnormal patterns and provide early warning of potential power failures or load anomalies.

[0107] Based on the early warning information, preventive maintenance tasks are automatically triggered.

[0108] The working principle of the above technical solution is: the DC bus voltage is continuously monitored through a high-precision analog-to-digital converter (ADC) to capture any instantaneous voltage fluctuations. Subsequently, digital filtering algorithms such as Kalman filters are used to remove noise and improve the accuracy of voltage monitoring. At the same time, the effective value, peak value and fluctuation range of the voltage are calculated in real time to provide basic data for subsequent safety range adjustment and energy management; the upper and lower limits of the safe operating range of the DC bus voltage are dynamically adjusted according to the current operating status of the system (such as load level, ambient temperature). The optimal safety threshold is predicted using machine learning models (such as prediction models based on historical data and real-time data) to maximize system efficiency while ensuring system safety; the integrated energy management system (EMS) combines historical energy consumption data, weather forecasts and load forecasts to predict energy demand and supply in the future through deep learning algorithms. This prediction helps to optimize the charging and discharging strategy of the energy storage device and improve energy utilization efficiency; based on the prediction results of the EMS, the charging and discharging strategy of the energy storage device (such as a battery pack) is intelligently scheduled. During peak demand, the DC bus voltage is stabilized by discharging the energy storage device; during trough demand, the excess energy is used for charging to optimize the cycle life and overall energy efficiency of the energy storage device; the DC side current waveform is analyzed by Fourier transform (FFT) to identify the harmonic components and their sources. These harmonic components may have a negative impact on system stability and efficiency, so they need to be evaluated and corresponding suppression measures taken; the time-varying characteristics of current fluctuations are monitored in real time using time-frequency analysis algorithms such as short-time Fourier transform (STFT) or wavelet transform. This real-time monitoring can quickly respond to potential power supply fluctuations or load mutations to ensure stable system operation; based on existing machine learning anomaly detection models, the DC bus voltage and current data are continuously monitored. These models can identify abnormal patterns, such as abnormal voltage fluctuations, excessive current harmonics, etc., so as to provide early warning of potential power failures or load anomalies; based on the warning information, preventive maintenance tasks are automatically triggered. These tasks may include adjusting control parameters, replacing aging components, or performing cleaning maintenance to reduce unplanned downtime and improve system reliability and service life.

[0109] The effects of the above technical solutions are: real-time monitoring and denoising of the DC bus voltage through ADC and digital filtering algorithms ensures the accuracy and real-time performance of voltage monitoring, helps to timely detect and handle voltage anomalies, and prevents system instability or safety accidents caused by voltage fluctuations. Dynamically adjusting the safe operating range of the DC bus voltage and using machine learning models to predict the optimal safety threshold can maximize system efficiency while ensuring system safety and improve the overall stability and safety of the system. The integrated energy management system (EMS) and deep learning algorithm can accurately predict the energy demand and supply situation in the future, providing strong support for the intelligent scheduling of energy storage devices. The intelligent scheduling of the charging and discharging strategy of the energy storage device ensures that the DC bus voltage remains stable during peak demand, which not only improves the power supply reliability of the system, but also optimizes the cycle life of the energy storage equipment, further improving the overall efficiency of the system. Analyzing the DC side current waveform through Fourier transform (FFT) to identify harmonic components and their sources helps to evaluate the impact of harmonics on system stability and efficiency, and take corresponding suppression measures to improve the performance and stability of the system. The application of time-frequency analysis algorithms (such as STFT or wavelet transform) can monitor the time-varying characteristics of current fluctuations in real time, quickly respond to potential power supply fluctuations or load mutations, and ensure that the system can quickly adapt to changes and maintain stable operation. Based on the existing machine learning anomaly detection model, the DC bus voltage and current data are continuously monitored, which can automatically identify abnormal patterns and provide early warnings for potential power failures or load anomalies, reducing the risk of system failures. Preventive maintenance tasks are automatically triggered based on early warning information, such as adjusting control parameters, replacing aging components, or performing cleaning maintenance, which effectively reduces unplanned downtime and improves system reliability and availability. Through the above measures, the system can operate more stably and efficiently, reducing downtime and maintenance costs caused by failures, and improving user satisfaction and economic benefits. At the same time, intelligent monitoring and maintenance methods also reduce reliance on manual intervention, reduce the burden on operation and maintenance personnel, and improve operation and maintenance efficiency.

[0110] In one embodiment of the present invention, the S19 includes:

[0111] S191. Use digital twin technology to build a high-precision virtual model of the inverter and its working environment. The model includes dynamic simulation of key parameters such as input voltage, load characteristics, and environmental conditions (such as light and temperature). Based on the virtual environment, various scenarios during the inverter startup process are reproduced, including normal startup, sudden failures, and extreme environmental conditions;

[0112] S192. Input the initially selected startup strategy into the virtual test environment to simulate the startup process of the inverter under the guidance of the strategy.

[0113] S193. Real-time monitoring of various performance indicators of the virtual inverter during the startup process, such as voltage stability, current fluctuation, startup time, energy conversion efficiency, etc. Comprehensive evaluation of the results of the simulation operation, and analysis of the gap between the actual effect of the startup strategy and the expected goal;

[0114] S194. Use big data analysis to identify performance bottlenecks and potential problem areas during the startup process, such as voltage overshoot, current harmonics, and excessive startup time. Based on the performance evaluation results, fine-tune key parameters in the startup strategy, including adjusting the voltage set point, optimizing the current limit strategy, and adjusting the PWM frequency and duty cycle.

[0115] S195, automatically searching for the optimal parameter combination through an intelligent optimization algorithm (such as a genetic algorithm or a particle swarm optimization algorithm), and inputting the adjusted startup strategy into the virtual test environment again for iterative testing;

[0116] S196. Repeat the above steps S192 to S195 until the performance index of the startup strategy reaches or exceeds the preset standard.

[0117] The working principle of the above technical solution is as follows: using digital twin technology to build a high-precision virtual model of the inverter and its working environment. The model integrates dynamic simulation of key parameters such as input voltage, load characteristics, and environmental conditions (such as light and temperature), and can reproduce various scenarios of the inverter in actual work, including normal startup, sudden failures, and extreme environmental conditions; the initially selected startup strategy is input into the virtual test environment to simulate the startup process of the inverter under the guidance of the strategy. Through the virtual model, the dynamic response and performance of the inverter under different startup strategies can be observed; various performance indicators of the virtual inverter during the startup process, such as voltage stability, current fluctuations, startup time, energy conversion efficiency, etc., are monitored in real time. These performance indicators reflect the actual effect of the startup strategy. The results of the simulation operation are comprehensively evaluated to analyze the gap between the actual effect of the startup strategy and the expected goal. By comparing the measured values ​​of the performance indicators with the preset standards, it is determined whether the startup strategy meets the requirements; using big data analysis technology, the performance data during the startup process is deeply mined to identify performance bottlenecks and potential problem areas. For example, voltage overshoot, current harmonics, and long startup time may affect the performance of the inverter. According to the performance evaluation results, the key parameters in the startup strategy are finely adjusted. These parameters may include voltage set point, current limit strategy, PWM frequency and duty cycle. By adjusting these parameters, the startup performance of the inverter can be optimized; intelligent optimization algorithms (such as genetic algorithm and particle swarm optimization algorithm) are used to automatically search for the optimal parameter combination. These algorithms can quickly find the optimal solution or approximate optimal solution in a complex parameter space, thereby further improving the startup performance of the inverter. The adjusted startup strategy is input into the virtual test environment again for iterative testing. By continuously repeating steps S192 to S195, the performance indicators of the startup strategy are gradually optimized until they reach or exceed the preset standards. After multiple iterative tests, the optimal startup strategy finally obtained will be fed back to the actual inverter system for application. In actual applications, the startup strategy can also be further adjusted and optimized based on real-time data and user feedback to ensure that the inverter can operate stably and efficiently under various working conditions.

[0118] The effect of the above technical solution is: the high-precision virtual model built by digital twin technology can simulate the startup process of the inverter under various complex working conditions, including normal startup, sudden failure and extreme environmental conditions. This simulation enables engineers to comprehensively evaluate the effect of the startup strategy without actually affecting the inverter hardware; in the virtual environment, various performance indicators during the inverter startup process can be monitored and evaluated in real time, so as to identify potential performance bottlenecks and problem areas in advance. This helps to optimize the startup strategy before actual application to avoid failures or performance degradation in actual operation; traditional inverter startup strategy optimization requires a lot of physical testing and experiments, which is not only time-consuming and labor-intensive, but also costly. The S19 technical solution builds a virtual test environment, allowing engineers to repeatedly test and optimize the startup strategy in a virtual environment, greatly reducing the cost of research and development and testing; due to the convenience of the virtual test environment, engineers can quickly adjust and optimize the startup strategy and perform iterative testing. This ability to quickly iterate accelerates the improvement process of the startup strategy and shortens the product development cycle; through big data analysis and intelligent optimization algorithms, the S19 technical solution can identify key parameters in the startup process and make fine adjustments to them. This adjustment can minimize voltage overshoot, current harmonics and other problems, thereby optimizing the inverter's startup performance and improving energy conversion efficiency; after multiple iterations of testing, the performance indicators of the startup strategy can meet or exceed the preset standards. This means that the inverter can show better startup performance and energy conversion efficiency in actual applications, thereby meeting users' demand for high-quality electricity; the optimized startup strategy can ensure that the inverter can start and operate stably and reliably under various working conditions, providing users with a stable and reliable power supply. This helps to improve users' trust and satisfaction with power equipment; as the reliability and stability of the startup strategy are improved, the inverter's failure rate and maintenance costs will also be reduced accordingly. This saves users maintenance costs and time costs and improves the overall user experience.

[0119] In one embodiment of the present invention, the S2 includes:

[0120] S21. Start the inverter with a small current, observe the initial response characteristics of the load, and preliminarily distinguish the load type (resistive, inductive, capacitive) by analyzing the shape and phase difference of the current response waveform;

[0121] S22, combining data from multiple pre-start cycles, making a comprehensive judgment on the load type, gradually increasing the current and monitoring voltage changes during the pre-start process, and analyzing the current-voltage relationship curve;

[0122] S23, calculating the equivalent impedance or power demand of the load according to the current-voltage relationship curve, and correcting the estimation result based on possible dynamic changes of the load (such as the inrush current when the motor starts);

[0123] S24. Match the most appropriate startup strategy from the preset strategy library according to the load type and size; fine-tune the parameters of the selected strategy according to the load characteristics and system requirements; the fine-tuning includes adjusting the initial value and rise rate of the PWM duty cycle.

[0124] S25. Verify the adjusted strategy through simulation software and evaluate its impact on system stability and efficiency.

[0125] The working principle of the above technical solution is as follows: the inverter is first started with a small current. This step is intended to reduce the sudden impact on the load and allow the system to start working in a gentler way. During the startup process, the inverter monitoring system preliminarily distinguishes the load type (resistive, inductive, capacitive) by observing the initial response characteristics of the load, especially the shape and phase difference of the current response waveform. These characteristics reflect the response mode of the load to current and voltage, and are an important basis for distinguishing different types of loads. Combining the data of multiple pre-start cycles, the system makes a more comprehensive judgment on the load type. In this stage, the inverter gradually increases the current and monitors the voltage change, and analyzes the current-voltage relationship curve to gain a deep understanding of the electrical characteristics of the load; the current-voltage relationship curve provides important information on the load impedance or power demand. By analyzing this curve, the system can further confirm the load type and estimate its equivalent impedance or power demand; considering the possible dynamic changes of the load (such as the impact current when the motor starts), the system makes necessary corrections to the estimation results to ensure the accuracy and effectiveness of the subsequent startup strategy; according to the load type and size, the system matches the most appropriate startup strategy from the preset strategy library. Subsequently, the parameters of the selected strategy are fine-tuned according to the load characteristics and system requirements. These fine-tuning may include adjusting the initial value and rise rate of the PWM duty cycle to optimize the output waveform and response speed of the inverter; the adjusted strategy is verified through simulation software to simulate the operation of the inverter under actual load conditions. This step is to evaluate the impact of the new strategy on system stability and efficiency to ensure its effectiveness and reliability in actual applications; based on the simulation results, the system further evaluates and optimizes the startup strategy. If any problems or deficiencies are found, the system will return to the previous steps to make necessary adjustments and improvements until the optimal startup strategy is found.

[0126] The effects of the above technical solution are as follows: preliminarily distinguishing the load type through small current starting, and making a comprehensive judgment based on the data of multiple pre-start cycles. This method not only improves the accuracy of load identification, but also completes the identification process in a shorter time, thereby improving efficiency; taking into account the possible dynamic changes of the load, such as the impact current when the motor starts, the estimation result is corrected, which makes the load identification closer to the actual situation and improves the accuracy of identification; according to the load type and size, the most suitable starting strategy is matched from the preset strategy library, and the parameters of the selected strategy are fine-tuned according to the load characteristics and system requirements. This personalized starting strategy can better adapt to the needs of different loads, reduce energy loss during the starting process, and improve the overall performance of the system; by adjusting the initial value and rise rate of the PWM duty cycle and other parameters, the output waveform and response speed of the inverter can be accurately controlled, the starting process can be further optimized, and the stability and efficiency of the system can be improved; during the pre-start process, the current is gradually increased and the voltage change is monitored, and the current-voltage relationship curve is analyzed, which helps to timely discover potential electrical problems or load abnormalities and avoid serious failures during the actual starting process; the adjusted strategy is verified by simulation software to evaluate its impact on system stability and efficiency. This verification method can comprehensively evaluate the effect of the startup strategy without actually affecting the system hardware, reducing system risks. The optimized startup strategy can ensure that the inverter responds quickly and outputs stably during the startup process, providing users with a more reliable and efficient power supply. This helps to improve users' trust and satisfaction with power equipment; due to the optimization of the startup strategy and the improvement of system stability, the failure rate and maintenance cost of the inverter will also be reduced accordingly. This saves users maintenance costs and time costs and improves the overall user experience.

[0127] In one embodiment of the present invention, the S24 includes:

[0128] On the basis of preliminary classification of load types (resistive, inductive, capacitive), further analysis is conducted on detailed characteristics of the load, such as frequency response, dynamic range, and thermal characteristics of impedance. Through machine learning algorithms, historical load data is learned to establish a load characteristic model.

[0129] According to load characteristics and system requirements (such as startup time, stability, efficiency, etc.), search for matching startup strategies in the preset strategy library;

[0130] Perform preliminary parameter setting for the retrieved startup strategy, including the initial value of the PWM duty cycle, the rise rate, the voltage set point, the current limit, etc. The parameters are preliminarily set according to the load characteristics and the system status through adaptive control theory;

[0131] Predict the dynamic changes of the load during the startup process, such as the inrush current and load mutation when the motor starts. Based on the prediction results, dynamically adjust the startup strategy, including adjusting the rising rate of the PWM duty cycle, adding current buffering links, optimizing voltage regulation strategies, etc.

[0132] Based on the load characteristics and system requirements, fine-tune the parameters of the selected strategy, and use simulation tools to test the system performance under different parameter combinations;

[0133] Through multi-objective optimization algorithms (such as NSGA-II), multiple optimization objectives such as startup time, stability, and efficiency are comprehensively considered to globally optimize the strategy parameters;

[0134] During the startup process, the load characteristics and system status are monitored in real time, such as current, voltage, temperature and other key parameters. Based on the monitoring results, the startup strategy is adjusted in real time.

[0135] The working principle of the above technical solution is as follows: on the basis of preliminary distinction of load types (resistance, inductance, capacitance), further analyze the detailed characteristics of the load, such as the frequency response, dynamic range, thermal characteristics, etc. of impedance. These characteristics are crucial for the selection and adjustment of subsequent startup strategies; through machine learning algorithms, historical load data is learned to establish a load characteristic model. This model can capture the complex characteristics and behavior patterns of the load, providing strong support for the selection and adjustment of the startup strategy; according to the load characteristics and system requirements (such as startup time, stability, efficiency, etc.), the matching startup strategy is retrieved from the preset strategy library. This process is similar to finding qualified records in a database; preliminary parameter settings are made for the retrieved startup strategy, including the initial value of the PWM duty cycle, the rise rate, the voltage set point, the current limit, etc. The setting of these parameters is based on adaptive control theory, and preliminary adjustments are made according to the load characteristics and system status; the dynamic changes of the load during the startup process are predicted, such as the inrush current and load mutation when the motor starts. These prediction results help to adjust the startup strategy in advance to deal with potential problems. Based on the prediction results, the startup strategy is dynamically adjusted. This includes adjusting the rate of rise of the PWM duty cycle, adding current buffering links, optimizing voltage regulation strategies, etc. These adjustments are aimed at improving the stability and efficiency of the startup process; fine-tuning the parameters of the selected strategy in combination with load characteristics and system requirements. Using simulation tools, the system performance under different parameter combinations is tested to find the optimal parameter combination; through multi-objective optimization algorithms (such as NSGA-II), multiple optimization goals such as startup time, stability, and efficiency are comprehensively considered to globally optimize the strategy parameters. This process aims to find a set of parameters that can better meet each optimization goal; during the startup process, the load characteristics and system status, such as current, voltage, temperature and other key parameters, are monitored in real time. Based on the monitoring results, the startup strategy is adjusted in real time. This real-time adjustment ensures that the system can maintain stable and efficient operation under various complex working conditions.

[0136] The effect of the above technical solution is: by deeply analyzing the detailed characteristics of the load and combining with the system requirements, the matching startup strategy is retrieved from the preset strategy library. This precise matching method can ensure that the selected strategy is highly consistent with the load characteristics and system requirements, thereby improving the efficiency and stability of the startup process; after the initial parameter setting of the retrieved startup strategy, fine-tuning and dynamic adjustment are performed according to the load characteristics and system status. This personalized adjustment can cope with the complexity and dynamic changes of the load, ensuring that the startup strategy can maintain optimal performance under different working conditions; predict the dynamic changes of the load during the startup process, such as the inrush current and load mutation when the motor starts. This prediction capability can identify potential problems in advance and take corresponding adjustment measures to avoid emergencies and failures during the startup process; dynamically adjust the startup strategy based on the prediction results, such as adjusting the rising rate of the PWM duty cycle, adding a current buffer link, etc. These preventive measures can reduce the risk and cost of the startup process and improve the reliability and stability of the system; the strategy parameters are globally optimized through a multi-objective optimization algorithm (such as NSGA-II), taking into account multiple optimization goals such as startup time, stability, and efficiency. This optimization method can find a set of optimal parameter combinations, so that the system performance is significantly improved in many aspects; the system performance under different parameter combinations is tested using simulation tools to ensure that the adjusted strategy can achieve the expected effect in actual applications. This simulation test can reduce the cost of trial and error and improve the efficiency and accuracy of adjustment; during the startup process, the load characteristics and system status, such as current, voltage, temperature and other key parameters, are monitored in real time. This real-time monitoring can promptly discover potential problems and provide a basis for subsequent adjustments; the startup strategy is adjusted in real time based on the monitoring results. This real-time feedback mechanism can ensure that the system can maintain optimal performance under different working conditions and improve the adaptability and robustness of the system; through precise matching and personalized adjustment, it ensures that the inverter can start quickly and stably to provide users with reliable power supply. This fast and stable startup process can improve the user experience and satisfaction; due to the optimization of the startup strategy and the improvement of system stability, the failure rate and maintenance cost of the inverter will also be reduced accordingly. This saves maintenance costs and time costs for users and further improves user satisfaction.

[0137] In one embodiment of the present invention, S3 includes:

[0138] S31, selecting an extremely low initial PWM duty cycle, and gradually increasing the PWM duty cycle in a linear or exponential growth manner;

[0139] S32, collecting output voltage, current, DC side voltage and load current data in real time through high-precision sensors, filtering and denoising the collected data, and extracting information;

[0140] S33, using a control algorithm (such as PID control, fuzzy control, neural network control, etc.) to perform closed-loop control on the inverter output; dynamically adjusting the PWM duty cycle and rise rate according to the changes in the voltage, current and load current monitored in real time;

[0141] S34. During the startup process, the dynamic changes of the load, such as sudden increase or decrease in load, are continuously evaluated, and the control strategy is adjusted accordingly to ensure system stability and efficiency.

[0142] If it is detected that the DC side voltage drops too quickly, the PWM duty cycle increase rate is reduced or the increase is suspended and then continued after the voltage stabilizes.

[0143] If the load changes suddenly (such as suddenly connecting a high-power device), the PWM duty cycle is quickly adjusted according to the load change to keep the output voltage stable.

[0144] An island detection mechanism is introduced to monitor the voltage and frequency of the grid access point during the startup process to ensure safe startup even in an island state.

[0145] S35. Set up an abnormality detection mechanism, such as overcurrent protection, overvoltage protection, short circuit protection, etc., and take corresponding measures immediately once an abnormal situation is detected.

[0146] The working principle of the above technical solution is as follows: In the S31 stage, a very low initial PWM duty cycle is selected to ensure that the inverter output voltage and current are started within a completely safe range; the PWM duty cycle is gradually increased in a linear or exponential growth manner, thereby slowly increasing the output voltage and current. This method helps to reduce the impact and instability during the startup process; in the S32 stage, the output voltage, current, DC side voltage and load current data are collected in real time through high-precision sensors. These data are important bases for subsequent control algorithms and adjustment strategies; the collected data are filtered and denoised to improve the accuracy and reliability of the data; key information is extracted from the processed data for subsequent control algorithms and strategy adjustments; in the S33 stage, control algorithms (such as PID control, fuzzy control, neural network control, etc.) are used to perform closed-loop control on the inverter output. These algorithms can dynamically adjust the PWM duty cycle and rise rate according to the changes in voltage, current and load current monitored in real time; through closed-loop control, the inverter output is ensured to be stable and meet the preset requirements. When the voltage, current or load current changes are detected, the control algorithm will respond quickly and adjust the control parameters to keep the system stable; the S34 stage continuously evaluates the dynamic changes of the load, such as sudden increase or decrease of the load. This evaluation helps to predict the system status in advance and take corresponding adjustment measures; quickly adjust the control parameters such as PWM duty cycle and rise rate according to the load changes. For example, when it is detected that the DC side voltage drops too fast, reduce the PWM duty cycle increase rate or pause the increase; when the load changes suddenly, quickly adjust the PWM duty cycle to keep the output voltage stable. The island detection mechanism is introduced to monitor the voltage and frequency of the grid access point during the startup process. Once the island state is detected, take corresponding measures immediately to ensure the safe startup and operation of the inverter; the S35 stage sets abnormal detection mechanisms, such as overcurrent protection, overvoltage protection, short circuit protection, etc. These mechanisms can monitor the operating status of the system in real time and detect potential abnormal conditions; once an abnormal condition is detected, take corresponding measures immediately to protect the inverter and load from damage. For example, reduce the output, cut off the power supply or trigger an alarm signal.

[0147] The effects of the above technical solution are: by selecting an extremely low initial PWM duty cycle, the output voltage and current of the inverter are ensured to be within the safe range at the initial startup, effectively preventing equipment damage or grid shock caused by excessive startup current; the abnormal detection mechanism (such as overcurrent protection, overvoltage protection, short circuit protection, etc.) set in S35 can monitor the operating status of the system in real time, and take corresponding measures (such as reducing output, cutting off power, etc.) immediately when abnormal conditions are detected to protect the inverter and load equipment from damage. This instant response mechanism greatly improves the safety and reliability of the system; the PWM duty cycle is gradually increased in a linear or exponential growth manner, so that the output voltage and current can be increased steadily, avoiding sudden changes in voltage and current during the startup process, thereby ensuring the stability of the inverter output; by adopting a control algorithm (such as PID control, fuzzy control, neural network control, etc.) to perform closed-loop control on the inverter output, the PWM duty cycle and rise rate can be dynamically adjusted according to the changes in voltage, current and load current monitored in real time. This closed-loop control strategy ensures that the inverter output can accurately track the preset requirements, improving the control accuracy and efficiency of the system; during the startup process, the dynamic changes of the load (such as sudden increase or decrease in load) are continuously evaluated, and the control strategy is adjusted accordingly. This adaptive adjustment mechanism can ensure that the inverter can maintain stable output performance under different load conditions, improving the adaptability and stability of the system; the introduction of the island detection mechanism monitors the voltage and frequency of the grid access point during the startup process to ensure safe startup even in the island state. This mechanism is of great significance for improving the grid-connected capability of the inverter and ensuring the safety of the grid; the output voltage, current, DC side voltage and load current data are collected in real time through high-precision sensors, and filtering, denoising and information extraction are performed. This process provides accurate data support for subsequent control algorithms and strategy adjustments, making the entire startup process more intelligent and automated; whether it is the monitoring and adjustment of the DC side voltage or the rapid response to load mutations, the S3 technical solution has demonstrated a strong dynamic adjustment capability. This capability enables the inverter to maintain optimal performance under complex and changing working conditions.

[0148] In one embodiment of the present invention, the S33 includes:

[0149] Set the proportional (P), integral (I), and differential (D) coefficients of the PID controller and perform preliminary debugging based on the preliminary response characteristics of the system;

[0150] PID control is used as the core framework for basic voltage and current regulation. Based on expert experience and system dynamic characteristics, a fuzzy rule set for the input (error, error change rate) and output (PID parameter adjustment amount) of the fuzzy logic controller is established.

[0151] Based on PID control, PID parameters are dynamically adjusted through fuzzy logic, and the neural network model is trained using historical data and simulation data to learn the nonlinear relationship between system input and output;

[0152] Predict the future state of the system online, fine-tune the PID and fuzzy control outputs based on the prediction results, and further optimize the control effect.

[0153] Combine historical load data with current load change trends, use time series analysis or machine learning algorithms to predict future load changes, and adjust the PWM duty cycle in advance based on the load prediction results;

[0154] Construct a sliding surface to ensure that the system state variables can still slide along the predetermined trajectory when subjected to external disturbances or parameter changes, and dynamically adjust the switching control law according to the distance between the system state and the sliding surface;

[0155] Use the extended Kalman filter (EKF) to identify system parameters online and automatically adjust the internal parameters of PID, fuzzy control or neural network based on the identification results;

[0156] Build a system operation case library to store control strategies and effects under different working conditions. When encountering new working conditions, use the case-based reasoning (CBR) method to quickly match similar cases, learn from their successful control strategies, and make fine adjustments based on the current status.

[0157] Combining multi-dimensional data such as voltage, current, temperature, and power factor, the system operating status is comprehensively evaluated. When an abnormal situation is detected, the protection mechanism (such as overcurrent protection, overvoltage protection, overheating protection, etc.) is automatically triggered, and the most appropriate response measures are intelligently selected according to the type of abnormality.

[0158] The working principle of the above technical solution is as follows: first, the proportional (P), integral (I), and differential (D) coefficients of the PID controller are set according to the preliminary response characteristics of the system, and preliminary debugging is performed to ensure the basic control performance; on the basis of PID control, a fuzzy logic controller is introduced. Through expert experience and system dynamic characteristics, a fuzzy rule set is established, and the error and error change rate are used as the input of the fuzzy logic controller to output the adjustment amount of the PID parameters. This fusion method can dynamically adjust the PID parameters to adapt to system changes and improve control accuracy and stability; the neural network model is trained using historical data and simulation data so that it can learn the nonlinear relationship between the input and output of the system; the neural network model predicts the future state of the system online, and fine-tunes the output of PID and fuzzy control based on the prediction results. This advanced control strategy can further optimize the control effect and improve the dynamic response capability of the system; combined with historical load data and current load change trends, time series analysis or machine learning algorithms are used to predict future load changes; according to the load prediction results, the PWM duty cycle is adjusted in advance to offset the impact of load changes on the output voltage. This strategy can significantly improve the dynamic response capability and stability of the system; construct a sliding surface to ensure that the system state variables can still slide along the predetermined trajectory when subjected to external disturbances or parameter changes; dynamically adjust the switching control law according to the distance between the system state and the sliding surface to quickly suppress disturbances and maintain system stability. Sliding mode control enhances the robustness and anti-interference ability of the system; use EKF to identify system parameters online, including system dynamic characteristics and uncertainty factors; automatically adjust the internal parameters of PID, fuzzy control or neural network according to the identification results, so that the control strategy always adapts to the current state of the system. This adaptive adjustment mechanism improves the flexibility and adaptability of the system; build a system operation case library to store control strategies and effects under different working conditions. When encountering new working conditions, the CBR method is used to quickly match similar cases, learn from their successful control strategies and fine-tune them in combination with the current state. This intelligent control method can significantly improve the decision-making efficiency and accuracy of the control system; comprehensively evaluate the system operation status by combining multi-dimensional data such as voltage, current, temperature, and power factor; automatically trigger protection mechanisms (such as overcurrent protection, overvoltage protection, overheating protection, etc.) when abnormal conditions are detected, and intelligently select the most appropriate response measures according to the abnormal type. This protection mechanism ensures the security and reliability of the system.

[0159] The effects of the above technical solution are: by dynamically adjusting the PID parameters through fuzzy logic, it can adapt to the dynamic changes of the system more accurately and improve the accuracy and stability of control. PID as the basic framework guarantees the basic control performance, while fuzzy logic enhances the adaptive ability of the system; the neural network model is used to predict the future state of the system online, and the control output is fine-tuned based on the prediction results, so as to realize the advanced control and further improve the accuracy and stability of the control; the future load is predicted by combining the historical load data and the current load change trend, and the PWM duty cycle is adjusted in advance accordingly, which effectively offsets the impact of the load change on the output voltage and significantly improves the dynamic response ability of the system; the sliding surface is constructed and the switching control law is dynamically adjusted when the system is subjected to external disturbances or parameter changes, which can quickly suppress the disturbance and keep the system stable, further enhancing the robustness and dynamic performance of the system; the control strategy is always adapted to the current state of the system by identifying the system parameters online and automatically adjusting the internal parameters of the control strategy, so as to realize adaptive control. This adaptive ability enables the system to maintain the best control effect under different working conditions; the system operation case library is constructed and similar cases are quickly matched through the case reasoning method, and its successful control strategy is learned and fine-tuned in combination with the current state to realize intelligent control. This intelligent control method not only improves the decision-making efficiency of the control system, but also reduces the dependence on expert experience; it combines multi-dimensional data such as voltage, current, temperature, and power factor to comprehensively evaluate the system operation status, which can provide a more comprehensive understanding of the system health status; it automatically triggers the protection mechanism (such as overcurrent protection, overvoltage protection, overheating protection, etc.) when an abnormal situation is detected, and intelligently selects the most appropriate response measures according to the type of abnormality. This instant response mechanism effectively protects the inverter and load equipment from damage and improves the safety and reliability of the system; by organically integrating multiple control strategies and algorithms, a collaborative overall control framework is formed. This comprehensive control strategy not only optimizes the performance of each control link, but also improves the overall performance of the entire system.

[0160] In one embodiment of the present invention, the S4 includes:

[0161] S41. According to system requirements and application scenarios, set the stable threshold and fluctuation range of output voltage and current, monitor the changes of output voltage and current in real time within a preset time window, and calculate the fluctuation range;

[0162] S42, comparing the real-time monitored data with the set stability threshold to determine whether the system has entered a stable state;

[0163] S43, if it is confirmed that the system has entered a stable state, immediately send a switching signal to switch the inverter from the soft start mode to the normal working mode;

[0164] S44. In normal working mode, according to load characteristics and system requirements, further optimize and adjust the operating parameters of the inverter; such as adjusting PWM frequency, optimizing voltage regulation range, setting more accurate current limit, etc. And evaluate the performance of the inverter, including efficiency, stability, response time and other indicators, to ensure that the system reaches the best operating state.

[0165] The working principle of the above technical solution is as follows: First, according to the system requirements and the specific needs of the application scenario, the stability thresholds of the output voltage and current and their allowed fluctuation ranges are set. These thresholds and ranges are determined based on the performance indicators and safety requirements of the system design; within the preset time window, the system monitors the changes in the output voltage and current of the inverter in real time. This is usually achieved through high-precision sensors or measurement circuits that can capture the instantaneous values ​​of voltage and current; based on the real-time monitored data, the system calculates the fluctuation amplitude of the output voltage and current, that is, the degree of their deviation from the set value. This step is the key to evaluating the stability of the system; the voltage and current data monitored in real time are compared with the set stability thresholds. This includes checking whether the voltage and current are kept within their respective threshold ranges, and whether their fluctuation amplitudes are less than the set fluctuation range; if both the voltage and current meet the stability requirements (i.e., they are within the threshold range and the fluctuation amplitude is within the allowable range), the system is judged to have entered a stable state; once it is confirmed that the system has entered a stable state, the system immediately sends a switching signal. This signal is the instruction to control the inverter to switch from soft start mode to normal working mode; after receiving the switching signal, the inverter gradually adjusts its internal working state according to the preset program and logic, and smoothly transitions from soft start mode to normal working mode. During this process, the inverter will gradually increase the output power while maintaining the stability of the output voltage and current; in normal working mode, the system will further optimize the operating parameters of the inverter according to the load characteristics and system requirements. This may include adjusting the PWM (pulse width modulation) frequency to improve the output waveform quality, optimizing the voltage regulation range to adapt to different load requirements, setting more precise current limits to prevent overcurrent, etc.; evaluate the performance of the inverter in normal working mode, including key indicators such as efficiency, stability, and response time. These evaluation results are used to verify whether the operating state of the inverter meets the design requirements and serve as the basis for subsequent adjustments and optimizations.

[0166] The effect of the above technical solution is: by accurately setting the stable threshold and fluctuation range of the output voltage and current according to the system requirements and application scenarios, and monitoring its changes in real time within the preset time window, the system can quickly capture any possible fluctuations or anomalies. This real-time monitoring mechanism helps to discover and deal with potential problems in a timely manner, thereby improving the stability and reliability of the system; by comparing the real-time monitoring data with the set stability threshold, it can accurately determine whether the system has entered a stable state. This step ensures that the inverter is in a stable state before switching to the normal working mode, avoiding system crashes or damage caused by unstable states; in the normal working mode, dynamically adjusting the operating parameters of the inverter (such as PWM frequency, voltage adjustment range, current limit, etc.) according to the load characteristics and system requirements helps to optimize the output characteristics of the inverter and improve the overall performance of the system. This dynamic adjustment mechanism enables the inverter to better adapt to different load conditions and system requirements; the performance of the inverter is evaluated, including key indicators such as efficiency, stability, and response time, providing strong data support for system optimization. Through continuous evaluation and optimization, the system can gradually approach the optimal operating state, achieve higher performance and lower energy consumption; by confirming that the system has entered a stable state before sending a switching signal, the inverter is smoothly switched from soft start mode to normal operating mode, reducing the voltage or current shock that may occur during the switching process. This smooth transition mechanism helps protect other devices and components in the system and avoid damage or failure caused by transient processes generated during the switching process; the dynamic adjustment and optimization mechanism in the S4 technical solution enables the inverter to automatically adjust operating parameters according to load characteristics and system requirements, thereby achieving adaptive control. This adaptability enhances the system's adaptability to different application scenarios and load conditions, and improves the system's flexibility and scalability; by ensuring that the inverter operates in a stable state and continuously optimizing its performance, the S4 technical solution improves the stability and reliability of the system and reduces failures and downtime. This helps to improve user experience and satisfaction, because users can use the system more confidently and expect it to provide high-quality power output in a continuous and stable manner.

[0167] In one embodiment of the present invention, S5 includes:

[0168] S51. During the startup process, detailed information of key parameters (such as voltage, current, temperature, PWM duty cycle, etc.) and events (such as strategy adjustment, mode switching, abnormal conditions, etc.) is recorded in detail;

[0169] S52, save the recorded data to a non-volatile memory and perform regular backup.

[0170] S53. During the startup process or in the normal working mode, once an abnormal situation or fault is detected, the fault alarm mechanism is immediately triggered, and the user or technician is notified through the indicator light, buzzer or communication interface.

[0171] S54, diagnose and locate the fault according to the fault alarm information, combined with the startup record and log analysis, and perform corresponding recovery operations (such as reset, reconfiguration, etc.) or automatically restart the inverter according to the fault diagnosis result;

[0172] S55. Record the fault information, diagnosis results and recovery operations in the fault report for subsequent analysis and improvement.

[0173] The working principle of the above technical solution is as follows: during the startup process of the inverter, the system starts to record key parameters (such as voltage, current, temperature, PWM duty cycle, etc.) and important events (such as strategy adjustment, mode switching, abnormal conditions, etc.) in detail. These parameters and events are important bases for subsequent fault diagnosis and performance analysis; the recording operation needs to be carried out in real time to ensure that all key information in the startup process is captured; the recorded data is saved in a non-volatile memory to ensure that the data will not be lost after the system is powered off or restarted; in order to further enhance the security of the data, the system will regularly back up the stored data to prevent data damage or loss; during the startup process or in normal working mode, the system continuously monitors the operating status of the inverter, and once an abnormal situation or fault (such as abnormal voltage, excessive current, high temperature, etc.) is detected, the fault alarm mechanism is immediately triggered; the user or technician is notified in time by flashing an indicator light (such as an LED light), sounding a buzzer, or sending an alarm signal through a communication interface (such as RS485, CAN bus, etc.); based on the fault alarm information and startup records and log analysis, the system diagnoses and locates the fault. This usually involves an in-depth analysis of the recorded data to find the root cause of the fault; based on the fault diagnosis results, the system performs the corresponding recovery operations. This may include resetting the inverter, reconfiguring key parameters, adjusting the control strategy, etc. In some cases, if the fault cannot be resolved by software means, the system may attempt to automatically restart the inverter to restore normal operation of the system; record the fault information, diagnostic results and recovery operations in the fault report. These reports not only provide a detailed record for the resolution of the current fault, but also provide valuable data support for subsequent system analysis and improvement; through in-depth analysis of fault reports, technicians can identify potential problems in the system and take preventive measures to avoid similar faults from happening again. In addition, these analyses also help improve the design and control strategy of the inverter, and enhance the overall performance and reliability of the system.

[0174] The effect of the above technical solution is: by recording the key parameters and events during the startup process in detail, the system can establish a comprehensive operating baseline. This allows any parameters or events that deviate from the normal range to be detected in time, which helps prevent the occurrence of potential faults; continuous monitoring of the inverter status during startup and normal working modes can detect abnormal conditions or faults early, thereby avoiding the expansion of faults and reducing system downtime; once an abnormal condition or fault is detected, the system can immediately trigger the alarm mechanism and notify the user or technician in a variety of ways (such as indicator lights, buzzers, communication interfaces). This immediate response mechanism can shorten the fault handling time and improve the efficiency of system maintenance; combined with startup records and log analysis, the system can accurately diagnose and locate faults. This helps technicians quickly find the root cause of the problem and reduce the blindness of troubleshooting; based on the fault diagnosis results, the system can automatically perform corresponding recovery operations (such as reset, reconfiguration, etc.) or try to automatically restart the inverter to restore normal operation of the system. This automated operation reduces the need for manual intervention and improves the efficiency of fault recovery. Through rapid response and efficient recovery mechanisms, the system can minimize downtime and ensure the continuity of production or application. The recorded data is saved in non-volatile memory to ensure that the data will not be lost after the system is powered off or restarted. This helps to recover and analyze data after a fault occurs. Regular data backup further enhances data security and prevents the risk of data corruption or loss. Fault information, diagnostic results, and recovery operations are recorded in fault reports, providing valuable data support for subsequent analysis and improvement. Through in-depth analysis of fault reports, technicians can identify potential problems in the system and take preventive measures to avoid similar faults from happening again. Based on the analysis results of fault reports, technicians can optimize the inverter system to improve its performance and reliability. This continuous improvement mechanism helps to improve user experience and overall system performance.

[0175] One embodiment of the present invention, as Figure 2 As shown, a soft start system for an inverter, the system comprising:

[0176] Self-check module: After the inverter is powered on, it performs hardware self-check and monitors the DC side voltage, current and external power supply status (such as solar panel output) through sensors; and reads and analyzes the startup configuration parameters and preset strategy library stored in the non-volatile memory;

[0177] Selection module: Before starting, the load type (resistance, inductance, capacitance, etc.) and size are preliminarily judged through low current pre-start or voltage detection, and the appropriate initial startup strategy is selected from the preset strategy library according to the load characteristics;

[0178] Monitoring module: Use a lower PWM duty cycle and a slow rise rate to gradually increase the inverter output voltage and current. Through closed-loop control, monitor the changes in output voltage, current, DC side voltage and load current in real time; adjust the startup strategy in real time based on the changes;

[0179] Adjustment module: When the output voltage and current reach the preset stability threshold and the fluctuation does not exceed the preset range within the preset time, it is judged that the system has entered a stable state; a switching signal is sent to switch the inverter from the soft start mode to the normal working mode and adjust it to the optimal operating parameters;

[0180] Recording module: records key parameters and events during the entire startup process. When startup fails or an abnormal situation is detected, it triggers a fault alarm and attempts to perform recovery operations or automatically restart.

[0181] The working principle of the above technical solution is as follows: after the inverter is powered on, it first performs a hardware self-check. This step is to ensure that all modules of the inverter (such as control module, power module, sensor, etc.) are in normal working state and there is no hardware failure. At the same time, the inverter monitors the voltage and current of the DC side and the output status of the external power supply (such as solar panels) in real time through sensors to ensure the stability and reliability of the input power supply. In addition, the inverter will also read and parse the startup configuration parameters and preset strategy library stored in the non-volatile memory. These parameters and strategies will guide the subsequent startup process. Before the formal startup, the inverter uses a small current pre-start or voltage detection method to make a preliminary judgment on the load. By monitoring the response characteristics of the load during the pre-start process (such as the change trend of voltage and current), the inverter can roughly judge the type (resistance, inductance, capacitance, etc.) and size of the load. Based on this judgment result, the inverter selects the most appropriate initial startup strategy from the preset strategy library. This step is crucial to ensure that the inverter can smoothly transition to a normal working state during the startup process. The inverter uses a lower PWM (pulse width modulation) duty cycle and a slow rise rate to gradually increase the output voltage and current. This gradual startup method helps to reduce the impact current and voltage fluctuations during the startup process, protecting the inverter and the load from damage. At the same time, the inverter monitors the changes in output voltage, current, DC side voltage and load current in real time through closed-loop control, and adjusts the startup strategy in real time according to these feedback signals. This dynamic adjustment mechanism ensures that the inverter can quickly adapt to load changes and maintain stable operation during the startup process. When the output voltage and current reach the preset stability threshold and the fluctuation does not exceed the preset range within the preset time, the inverter determines that the system has entered a stable state. At this time, the inverter will send a switching signal to switch itself from soft start mode to normal working mode and adjust to the optimal operating parameters to meet the load requirements. This step marks the end of the inverter startup process and the beginning of normal working state. During the entire startup process, the inverter will record key parameters and events for subsequent analysis and optimization. If the startup fails or an abnormal situation is detected (such as overload, short circuit, overheating, etc.), the inverter will trigger a fault alarm and try to perform recovery operations or automatically restart. This fault handling mechanism helps to detect and solve problems in a timely manner, ensuring the reliability and stability of the inverter.

[0182] The effects of the above technical solution are: ensuring that all modules of the inverter are in normal state before power-on, avoiding startup failure or damage caused by hardware failure; effectively reducing the impact current during the startup process through small current pre-start and gradually increasing the output voltage / current, protecting the electronic components inside the inverter and the external load, and avoiding damage caused by instantaneous large current; selecting a suitable startup strategy through preliminary judgment of the load, so that the inverter can better adapt to different types of loads and improve the success rate of startup; during the startup process, real-time monitoring and adjustment of the startup strategy through closed-loop control ensure that the inverter can quickly respond to load changes, further improving the stability and success rate of startup; after the inverter enters the normal working mode, by adjusting to the optimal operating parameters, the overall performance of the system is optimized, and the conversion efficiency and output stability of the inverter are improved; The method of increasing the output voltage and current step by step reduces unnecessary energy consumption and improves energy utilization compared to direct full-power startup; it records key parameters and events during the startup process in real time, and triggers a fault alarm when an abnormal situation is detected, which helps to find and solve problems in time and enhance the reliability and stability of the system; when the startup fails or an abnormal situation is detected, the inverter will try to perform recovery operations or automatically restart, reducing downtime caused by faults and improving system availability and reliability; the inverter transitions smoothly during the startup process, reducing voltage and current fluctuations, providing users with a more stable and reliable power supply and improving user experience; by reading and parsing the startup configuration parameters and preset strategy library stored in the non-volatile memory, the intelligent management of the inverter is realized, the need for manual intervention is reduced, and the automation level of the system is improved.

[0183] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A soft start method for an inverter, characterized in that: The method comprises: S1. After the inverter is powered on, it performs a hardware self-check and monitors the DC side voltage, current and external power supply status through sensors; and reads and analyzes the startup configuration parameters and preset strategy library stored in the non-volatile memory; S2. Before starting, make a preliminary judgment on the load type and size through low current pre-start or voltage detection, and select the initial startup strategy from the preset strategy library according to the load characteristics; S3, using a lower PWM duty cycle and a slow rise rate, gradually increasing the inverter output voltage and current, and monitoring the changes in output voltage, current, DC side voltage and load current in real time through closed-loop control; S4. When the output voltage and current reach the preset stability threshold and the fluctuation does not exceed the preset range within the preset time, it is judged that the system has entered a stable state; a switching signal is sent to switch the inverter from the soft start mode to the normal working mode and adjust it to the optimal operating parameters; S5. Record key parameters and events during the entire startup process. When startup fails or an abnormal situation is detected, trigger a fault alarm and try to perform recovery operations or automatically restart.

2. The soft start method of an inverter according to claim 1, characterized in that: Said S1 comprises: S11, verifying the input voltage, determining whether the input voltage is within an allowable range, and checking the working state of the power conversion circuit; S12, perform a switch test on the IGBT module through a built-in test program, and detect the capacitance and inductance; S13, calibrating the voltage, current and temperature sensors, and verifying the integrity and reliability of the communication interface by sending test signals and receiving responses; S14, monitor the DC bus voltage in real time, and predict the energy reserve situation, analyze the DC side current fluctuation based on the prediction result, and identify potential power supply fluctuation or load change based on the analysis result; S15, monitor the output voltage and current of the solar panel, calculate the MPPT efficiency, and predict the energy output according to weather conditions; S16. Setting temperature sensors inside and outside the inverter to monitor temperature changes in real time; S17, reading the configuration parameters of the last successful startup from the non-volatile memory, parsing the preset strategy library, and understanding the load type, environmental conditions, and advantages and disadvantages applicable to each strategy; S18, preliminarily selecting a startup strategy based on current environmental conditions, historical data, and load expectations; S19. Before selecting a strategy, conduct simulation tests to verify the feasibility and effectiveness of the launch strategy, and make fine adjustments as needed.

3. A soft start method for an inverter according to claim 2, characterized in that: The S14 comprises: The ADC continuously monitors the DC bus voltage, captures any instantaneous voltage fluctuations, removes noise through a digital filtering algorithm, and calculates the effective value, peak value, and fluctuation range of the voltage in real time; Dynamically adjust the upper and lower limits of the safe operating range of the DC bus voltage according to the current operating status of the system, and use the machine learning model to predict the optimal safety threshold; Integrated energy management system, combining historical energy consumption data, weather forecast and load forecast, and using deep learning algorithms to predict energy demand and supply in the future; According to the prediction results, the charging and discharging strategy of the energy storage device is intelligently scheduled, and the cycle life of the energy storage device is optimized; Analyze the DC side current waveform through Fourier transform, identify the harmonic components and their sources, and evaluate their impact on system stability and efficiency; Through the time-frequency analysis algorithm, the time-varying characteristics of current fluctuations are monitored in real time, and potential power supply fluctuations or load mutations can be quickly responded to; Based on the existing machine learning anomaly detection model, the DC bus voltage and current data are continuously monitored to identify abnormal patterns and provide early warning of potential power failures or load anomalies. Based on the early warning information, preventive maintenance tasks are automatically triggered.

4. A soft start method for an inverter according to claim 2, characterized in that: The S19 comprises: S191. Use digital twin technology to build a high-precision virtual model of the inverter and its working environment, and reproduce various scenarios during the inverter startup process based on the virtual environment; S192, inputting the initially selected startup strategy into the virtual test environment to simulate the startup process of the inverter under the guidance of the strategy; S193. Real-time monitoring of various performance indicators of the virtual inverter during the startup process, comprehensive evaluation of the results of the simulation operation, and analysis of the gap between the actual effect of the startup strategy and the expected target; S194. Use big data analysis to identify performance bottlenecks and potential problem areas during the startup process, and make fine adjustments to key parameters in the startup strategy based on performance evaluation results; S195, automatically searching for the optimal parameter combination through an intelligent optimization algorithm, and inputting the adjusted startup strategy into the virtual test environment again for iterative testing; S196. Repeat the above steps S192 to S195 until the performance index of the startup strategy reaches or exceeds the preset standard.

5. The soft start method of an inverter according to claim 1, characterized in that: The S2 comprises: S21, start the inverter with a small current, observe the initial response characteristics of the load, and preliminarily distinguish the load type by analyzing the shape and phase difference of the current response waveform; S22, combining data from multiple pre-start cycles, making a comprehensive judgment on the load type, gradually increasing the current and monitoring voltage changes during the pre-start process, and analyzing the current-voltage relationship curve; S23, calculating the equivalent impedance or power demand of the load according to the current-voltage relationship curve, and correcting the estimation result based on the dynamic changes of the load; S24, matching a startup strategy from a preset strategy library according to the load type and size; fine-tuning the parameters of the selected strategy according to the load characteristics and system requirements; S25. Verify the adjusted strategy through simulation software and evaluate its impact on system stability and efficiency.

6. The soft starting method of an inverter according to claim 1, characterized in that: The S24 comprises: On the basis of preliminary classification of load types, the detailed characteristics of the load are further analyzed, and the historical load data is learned through machine learning algorithms to establish a load characteristic model; According to load characteristics and system requirements, a matching startup strategy is retrieved from the preset strategy library; Perform preliminary parameter setting on the retrieved startup strategy, and use adaptive control theory to preliminarily set parameters based on load characteristics and system status; Predict the dynamic changes of load during the startup process and dynamically adjust the startup strategy based on the prediction results; Based on the load characteristics and system requirements, fine-tune the parameters of the selected strategy, and use simulation tools to test the system performance under different parameter combinations; Global optimization of strategy parameters through multi-objective optimization algorithm; During the startup process, the load characteristics and system status are monitored in real time, and the startup strategy is adjusted in real time based on the monitoring results.

7. The soft starting method of an inverter according to claim 1, characterized in that: The S3 includes: S31, selecting an initial PWM duty cycle, and gradually increasing the PWM duty cycle in a linear or exponential growth manner; S32, collecting output voltage, current, DC side voltage and load current data in real time through high-precision sensors, filtering and denoising the collected data, and extracting information; S33, using a control algorithm to perform closed-loop control on the inverter output; dynamically adjusting the PWM duty cycle and rise rate according to the changes in voltage, current and load current monitored in real time; S34. During the startup process, the dynamic changes of the load are continuously evaluated and the control strategy is adjusted accordingly; S35. Set up an abnormality detection mechanism and take corresponding measures immediately once an abnormal situation is detected.

8. The soft starting method of an inverter according to claim 1, characterized in that: The S4 comprises: S41, setting the stable threshold and fluctuation range of the output voltage and current, monitoring the changes of the output voltage and current in real time within a preset time window, and calculating the fluctuation range; S42, comparing the real-time monitored data with the set stability threshold to determine whether the system has entered a stable state; S43, if it is confirmed that the system has entered a stable state, immediately send a switching signal to switch the inverter from the soft start mode to the normal working mode; S44. Under normal operating mode, further optimize and adjust the operating parameters of the inverter according to load characteristics and system requirements; and evaluate the performance of the inverter.

9. The soft starting method of an inverter according to claim 1, characterized in that: The S5 comprises: S51. During the startup process, detailed information of key parameters and events is recorded in detail; S52, save the recorded data to a non-volatile memory and perform regular backup. S53. During the startup process or in the normal working mode, once an abnormal situation or fault is detected, the fault alarm mechanism is immediately triggered and the user or technician is notified through multiple channels; S54, diagnose and locate the fault according to the fault alarm information, combined with the startup record and log analysis, and perform corresponding recovery operations or automatically restart the inverter according to the fault diagnosis result; S55. Record the fault information, diagnosis results and recovery operations in the fault report for subsequent analysis and improvement.

10. A soft start system for an inverter, characterized in that: The system comprises: Self-check module: After the inverter is powered on, it performs hardware self-check and monitors the DC side voltage, current and external power supply status through sensors; and reads and analyzes the startup configuration parameters and preset strategy library stored in the non-volatile memory; Selection module: Before starting, the load type and size are preliminarily judged through low current pre-start or voltage detection, and the initial startup strategy is selected from the preset strategy library according to the load characteristics; Monitoring module: Uses a lower PWM duty cycle and a slow rise rate to gradually increase the inverter output voltage and current. Through closed-loop control, it monitors the changes in output voltage, current, DC side voltage and load current in real time. Adjustment module: When the output voltage and current reach the preset stability threshold and the fluctuation does not exceed the preset range within the preset time, it is judged that the system has entered a stable state; a switching signal is sent to switch the inverter from the soft start mode to the normal working mode and adjust it to the optimal operating parameters; Recording module: records key parameters and events during the entire startup process. When startup fails or an abnormal situation is detected, it triggers a fault alarm and attempts to perform recovery operations or automatically restart.

Images