CN-121356358-B - Self-adaptive control method and system for inverter

CN121356358BCN 121356358 BCN121356358 BCN 121356358BCN-121356358-B

Abstract

The invention discloses a self-adaptive control method and a self-adaptive control system of an inverter, wherein the method comprises the steps of acquiring current operation sampling data, working condition parameters and rated device parameters of an arc load in real time; performing fast Fourier transformation to obtain frequency spectrum characteristic data, performing optimization parameter adjustment based on an arc load model library to obtain primary control parameters, inputting current operation sampling data into a support vector machine model which is trained in advance to obtain target control parameters, performing load adjustment based on a fuzzy logic control algorithm to obtain adjustment current operation sampling data, performing distortion analysis to obtain adjustment results, adjusting the primary control parameters to obtain initial control parameters, and performing working condition analysis and adjustment according to the environment temperature, the rated parameters of the device and the initial control parameters to obtain final control parameters. The method realizes the self-adaptive control of the inverter to the arc load and the sensitive and accurate control of stable combustion of the arc.

Inventors

ZHANG XUELIN
ZHANG YUANXING
WANG MING

Assignees

重庆荣凯川仪仪表有限公司

Dates

Publication Date: 20260508
Application Date: 20251217

Claims (8)

1. An adaptive control method of an inverter, comprising: S1, acquiring current operation sampling data, working condition parameters and device rated parameters of an arc load, wherein the current operation sampling data comprise real-time voltage and real-time current, and the working condition parameters comprise ambient temperature, solder supply speed, solder parameters and solder supply acceleration parameters, and the acceleration parameters correspond to acceleration duration time and acceleration magnitude; S2, performing fast Fourier transform on the current operation sampling data to obtain frequency spectrum characteristic data; S3, matching the current arc load type by adopting a preset arc load model library according to the spectrum characteristic data, and extracting a preliminary control parameter set from the matched model; S4, inputting the preliminary control parameter set, the working condition parameters and the solder supply acceleration parameters into a support vector machine model which is trained in advance to obtain target control parameters, wherein the training process of the support vector machine model comprises the following steps: Recording a history control parameter set, history working condition parameters and corresponding history target control parameters under different arc working conditions; Normalizing the historical control parameter set and the historical working condition parameters, taking the normalized historical control parameter set and the historical working condition parameters as input, taking the historical target control parameters as output, constructing a support vector regression model, and training a kernel function of the support vector regression model; When the loss function value of the support vector regression model is smaller than a preset loss threshold value, obtaining a support vector machine model after training; s5, inputting the target control parameters into a preset fuzzy logic controller for fuzzification processing to obtain fuzzy control quantity, and defuzzifying the fuzzy control quantity to generate an adjustment instruction so that the inverter can be adjusted according to the adjustment instruction; s6, monitoring the output voltage waveform and the output current waveform of the inverter in real time, detecting distortion components in the waveform by wavelet transformation, and judging the stability of arc combustion according to the distortion components; s7, when unstable combustion of the electric arc is detected, calculating fluctuation amplitude according to rated parameters of the device, and determining the type of the current working condition; And S8, searching from a preset working condition parameter mapping table according to the current working condition type to obtain an optimal parameter combination, and correcting the current control parameter according to the optimal parameter combination so as to ensure that the output characteristic of the inverter is kept stable under different working conditions.
2. The adaptive control method of an inverter according to claim 1, characterized by comprising, in step S3: extracting spectral features of the spectral feature data to obtain a spectral feature vector; performing similarity calculation on the frequency spectrum feature vector and a model feature vector in a preset arc load model library to obtain similarity, wherein each arc load model comprises the model feature vector and a preliminary control parameter corresponding to the arc load type; selecting the model with the highest similarity as the current arc load model; And extracting control parameters corresponding to the current arc load model to obtain a preliminary control parameter set.
3. The adaptive control method of an inverter according to claim 2, wherein a cosine similarity is used to calculate a similarity between the spectral feature vector and the model feature vector to obtain a similarity; The calculation formula of the similarity is as follows: in the formula, Representing the similarity, n represents the dimension of the feature vector, Representing the ith component of the spectral feature vector, Representing the ith component of the model feature vector.
4. The adaptive control method of an inverter according to claim 1, wherein in step S5, the derivation of the fuzzy control amount includes the steps of: determining a fuzzy set of target control parameters; and calculating the membership degree of the target control parameter to the fuzzy set, wherein the membership degree calculation formula is as follows: Wherein, the Representing membership; representing the normalized target control parameters; 、 And All are set constants; and according to the membership degree, combining a preset fuzzy rule to obtain a fuzzy control quantity.
5. The method according to claim 1, wherein the step of detecting a distortion component in the waveform by wavelet transformation to determine the stability of arc combustion comprises: decomposing the output voltage waveform and the output current waveform by adopting wavelet transformation, and extracting distortion components; Acquiring amplitude-frequency characteristic data from the distortion components, and judging the burning state of the arc; and when the amplitude frequency characteristic data exceeds a preset amplitude frequency threshold value, judging that the arc combustion is unstable.
6. The method for adaptively controlling an inverter according to claim 1, wherein said calculating a fluctuation range according to said device rating parameter, determining a current operating condition type, comprises: Calculating the current voltage fluctuation amplitude and current fluctuation amplitude according to the rated parameters of the device; the calculation formulas of the voltage fluctuation amplitude and the current fluctuation amplitude are as follows: Wherein, the Representing the voltage fluctuation amplitude; Representing a peak value of a current voltage waveform; representing a current voltage waveform valley; representing the rated voltage of the device; representing the current fluctuation amplitude; representing the peak value of the current waveform; representing a present current waveform valley; indicating the rated current of the device; and respectively comparing the voltage fluctuation amplitude and the current fluctuation amplitude with a preset fluctuation amplitude range, and determining the type of the current working condition.
7. The method according to claim 1, wherein the correcting the current control parameter according to the optimal parameter combination to ensure that the output characteristic of the inverter remains stable under different conditions comprises: correcting the control parameters according to the optimal parameter combination to obtain corrected control parameters; Updating inverter settings by adopting the corrected control parameters, and judging whether the output characteristics are stable; And when the output characteristic is not kept stable, acquiring the optimal parameter combination again according to the working condition type, and correcting the control parameter again until the output characteristic is kept stable.
8. An adaptive control system of an inverter, configured to implement the adaptive control method of an inverter of claim 1, comprising: the data acquisition module is used for acquiring current operation sampling data, working condition parameters and device rated parameters of the arc load; the data processing module is used for carrying out fast Fourier transform on the current operation sampling data to obtain frequency spectrum characteristic data; The initial parameter acquisition module comprises a preset arc load model library and corresponding control parameters, matches the current arc load type according to the frequency spectrum characteristic data, and extracts a corresponding initial control parameter set; The target parameter acquisition module comprises a support vector machine model which is trained in advance, and outputs the preliminary control parameter set, the working condition parameters and the solder supply acceleration parameters which are input into the support vector machine model as target control parameters; The adjusting module comprises a fuzzy logic controller with preset adjusting logic, and performs fuzzification processing on the target control parameters input into the fuzzy logic controller to obtain fuzzy control quantity, and performs defuzzification on the fuzzy control quantity to generate an adjusting instruction so as to enable the inverter to adjust according to the adjusting instruction; The stability judging module is used for monitoring the output voltage waveform and the output current waveform of the inverter in real time, detecting distortion components in the waveform by adopting wavelet transformation, and judging the stability and the working condition type of arc combustion according to the distortion components; And the parameter correction module is used for searching from a preset working condition parameter mapping table according to the current working condition type to obtain an optimal parameter combination, and correcting the current control parameter according to the optimal parameter combination.

Description

Self-adaptive control method and system for inverter Technical Field The present invention relates to the field of inverter control technologies, and in particular, to an adaptive control method and system for an inverter. Background Arc welding inverter power supplies are currently in wide use in control of a variety of industrial applications, particularly in situations where matching to arc loads is required, such as electric welding and arc heating systems. However, the dynamic characteristics of the arc load are extremely complex and affected by a variety of factors, such as electrode materials, electrode spacing, gas composition, gas flow rate, etc., which make the voltage and current characteristics of the arc constantly changing, and present a great challenge to the control of the inverter. The existing arc welding inverter is usually controlled by adopting fixed parameters, such as PID gain, proportional gain and the like, which are optimized only based on specific working conditions. However, the voltage-current characteristics of an arc load have significant nonlinearity, time-variability, and randomness, and the negative resistance characteristics of an arc (voltage drop when current increases) conflict with the linear assumption of fixed parameter control, which is prone to positive feedback instability. In addition, the dynamic behavior of the arc is affected by a number of factors, and once the arc characteristics change, such as electrode ablation leading to increased spacing, the original fixed parameters such as proportional gain cannot compensate for the new error characteristics, and too high a proportional gain may cause oscillations and too low a proportional gain leads to response hysteresis. Therefore, the traditional inverter control method cannot adapt to the dynamic change of the arc load, so that the real-time control precision is insufficient, and the stability of arc combustion is insufficient. Disclosure of Invention The invention provides a self-adaptive control method and a self-adaptive control system for an inverter, which are used for realizing the self-adaptive control of the inverter on an arc load and solving the problems of insufficient sensitivity and unstable arc combustion of accurate control. In order to solve the above technical problems, the present invention provides an adaptive control method of an inverter, including: S1, acquiring current operation sampling data, working condition parameters and device rated parameters of an arc load, wherein the current operation sampling data comprise real-time voltage and real-time current; S2, performing fast Fourier transform on the current operation sampling data to obtain frequency spectrum characteristic data; S3, matching the current arc load type by adopting a preset arc load model library according to the spectrum characteristic data, and extracting a preliminary control parameter set from the matched model; S4, inputting the preliminary control parameter set, the working condition parameters and the solder supply acceleration parameters into a support vector machine model which is trained in advance to obtain target control parameters; s5, inputting the target control parameters into a preset fuzzy logic controller for fuzzification processing to obtain fuzzy control quantity, and defuzzifying the fuzzy control quantity to generate an adjustment instruction so that the inverter can be adjusted according to the adjustment instruction; s6, monitoring the output voltage waveform and the output current waveform of the inverter in real time, detecting distortion components in the waveform by wavelet transformation, and judging the stability of arc combustion according to the distortion components; s7, when unstable combustion of the electric arc is detected, calculating fluctuation amplitude according to rated parameters of the device, and determining the type of the current working condition; And S8, searching from a preset working condition parameter mapping table according to the current working condition type to obtain an optimal parameter combination, and correcting the current control parameter according to the optimal parameter combination so as to ensure that the output characteristic of the inverter is kept stable under different working conditions. In an optional implementation manner, the matching the current arc load type with a preset arc load model base according to the spectrum characteristic data, and extracting a preliminary control parameter set from the matched model includes: extracting spectral features of the spectral feature data to obtain a spectral feature vector; performing similarity calculation on the frequency spectrum feature vector and a model feature vector in a preset arc load model library to obtain similarity, wherein each arc load model comprises the model feature vector and a preliminary control parameter corresponding to the arc load type; selecting the model with the highest simi