Power quality testing methods and equipment in frequency converters

Abstract

This invention discloses a method and equipment for power quality detection in frequency converters, relating to the field of frequency converters. The method includes: constructing a power quality map; performing multi-dimensional fault prediction on the frequency converter based on the power quality map to obtain a frequency converter fault map; optimizing the frequency converter's frequency control decisions to obtain a fault suppression frequency converter regulation library; optimizing the output stability of the frequency converter's control decisions to obtain a stability frequency converter regulation library; and performing global propagation optimization on the fault suppression and stability frequency converter regulation libraries based on frequency converter optimality evaluation conditions to obtain the frequency converter regulation optimization result. This invention solves the technical problems of existing technologies, such as single-dimensional power quality detection, insufficient fault prediction capability, and poor adaptability of frequency converter control regulation. It achieves the technical effects of improving the accuracy of power quality detection in frequency converters, enhancing fault prediction and suppression capabilities, and realizing stable adaptive optimization control of the frequency converter.

Description

Technical Field

[0001] This application relates to the field of frequency converters, and in particular to power quality testing methods and equipment in frequency converters. Background Technology

[0002] Inverters, as a crucial component of power electronic equipment, are widely used in motor control, wind power generation, and grid regulation. In actual operation, the power quality of inverters directly affects the safety, efficiency, and lifespan of the equipment. Existing inverter power quality detection methods often rely on single monitoring indicators, such as voltage, current, or harmonic content, lacking a multi-dimensional, dynamic, and comprehensive evaluation system. This fails to fully reflect the power quality status of inverters under complex load conditions. Furthermore, traditional methods primarily depend on post-event analysis, lacking the ability to predict and proactively control potential accidents. This makes inverters prone to power quality anomalies or unstable output when operating under varying loads, grid connection, or multi-device collaborative operation.

[0003] Therefore, there is an urgent need for a technical solution that can realize multi-dimensional power quality detection, dynamic accident prediction and adaptive optimization control to improve the safety, reliability and intelligence level of frequency converter operation. Summary of the Invention

[0004] This application provides a power quality detection method and equipment for frequency converters, solving the technical problems of limited power quality detection dimensions, insufficient fault prediction capabilities, and poor control and adjustment adaptability in existing technologies. It achieves the technical effects of improving the accuracy of power quality detection, enhancing fault prediction and suppression capabilities, and realizing stable adaptive optimization control of the frequency converter.

[0005] This application provides a power quality detection method for frequency converters. The method includes: real-time monitoring of the frequency converter of a target device to obtain frequency converter monitoring data; multi-dimensional power quality detection of the frequency converter based on the frequency converter monitoring data to construct a power quality map; multi-dimensional fault prediction of the frequency converter based on the power quality map to obtain a frequency converter fault map; adjustment and optimization of the frequency converter control decision based on the frequency converter fault map and a frequency converter risk prediction model to obtain a fault suppression frequency converter adjustment library; output stability adjustment and optimization of the frequency converter control decision based on the load prediction data of the target device to obtain a stability frequency converter adjustment library; global propagation optimization of the fault suppression frequency converter adjustment library and the stability frequency converter adjustment library based on frequency converter optimality evaluation conditions to obtain frequency converter adjustment optimization results, and performing frequency converter optimization control based on the frequency converter adjustment optimization results.

[0006] Preferably, performing multi-dimensional fault prediction on the frequency converter based on the power quality map to obtain a frequency converter fault map includes: predicting voltage fluctuation faults on the frequency converter based on the power quality map to obtain voltage fluctuation fault paths; predicting current surge faults on the frequency converter based on the power quality map to obtain current surge fault paths; predicting harmonic distortion faults on the frequency converter based on the power quality map to obtain harmonic distortion fault paths; and generating the frequency converter fault map based on the voltage fluctuation fault paths, the current surge fault paths, and the harmonic distortion fault paths.

[0007] Preferably, predicting voltage fluctuation incidents for the frequency converter based on the power quality map to obtain voltage fluctuation incident paths includes: searching for voltage fluctuation incidents among frequency converters in the same family to obtain a historical voltage fluctuation incident set; training an incident tree based on the historical voltage fluctuation incident set to obtain a voltage fluctuation incident tree model; performing adversarial training on the voltage fluctuation incident tree model based on the historical voltage fluctuation incident set to obtain a voltage fluctuation incident inference model; and outputting the voltage fluctuation incident path based on the power quality map and the voltage fluctuation incident inference model.

[0008] Preferably, based on the inverter fault map, the inverter control decisions of the inverter are adjusted and optimized according to the inverter risk prediction model to obtain a fault suppression inverter regulation library, including: performing correlation feature analysis on the inverter control decisions based on the inverter fault map to obtain an fault inverter control correlation graph; performing adaptive adjustment on the inverter control decisions based on the fault inverter control correlation graph to obtain a first inverter regulation library; performing risk prediction on the first inverter regulation library according to the inverter risk prediction model to obtain an inverter regulation risk library; and performing inverter regulation risk optimization on the first inverter regulation library based on the inverter regulation risk library and inverter regulation risk constraints to obtain the fault suppression inverter regulation library.

[0009] Preferably, optimizing the output stability of the frequency converter control decision based on the load prediction data of the target device to obtain a stability frequency converter adjustment library includes: adjusting the frequency converter control decision based on the load prediction data to obtain a second frequency converter adjustment library; extracting a Y-th frequency converter adjustment scheme from the second frequency converter adjustment library, where Y is a positive integer; performing simulated adjustment on the frequency converter based on the load prediction data and the Y-th frequency converter adjustment scheme to obtain Y-th frequency converter simulation data; obtaining a Y-th output stability based on the Y-th frequency converter simulation data and an output stability evaluation network; and adding the Y-th frequency converter adjustment scheme to the stability frequency converter adjustment library if the Y-th output stability is greater than or equal to a predetermined output stability.

[0010] Preferably, the frequency conversion optimization is performed on the accident suppression frequency conversion control library and the stability frequency conversion control library according to the frequency conversion optimization evaluation conditions to obtain the frequency conversion control optimization result, including: performing intersection identification on the accident suppression frequency conversion control library and the stability frequency conversion control library to obtain the first global frequency conversion control library;

[0011] Based on the first global frequency converter regulation library, the accident suppression frequency converter regulation library and the stability frequency converter regulation library are subjected to global multiplication optimization to obtain a second global frequency converter regulation library; based on the second global frequency converter regulation library, the accident suppression frequency converter regulation library and the stability frequency converter regulation library are subjected to global multiplication optimization until multiple global frequency converter regulation libraries that meet the number of global multiplication optimizations are obtained; based on the frequency converter optimality evaluation conditions, the multiple global frequency converter regulation libraries are subjected to frequency converter optimality maximization optimization to obtain the frequency converter regulation optimization result.

[0012] Preferably, based on the first global frequency converter regulation library, a second global frequency converter regulation library is obtained by performing global propagation optimization on the accident suppression frequency converter regulation library and the stability frequency converter regulation library, including: detecting differences in frequency converter regulation parameters in the accident suppression frequency converter regulation library based on the first global frequency converter regulation library to obtain a first frequency converter difference vector set; performing cross-mutation on the accident suppression frequency converter regulation library based on the first frequency converter difference vector set to obtain a third frequency converter regulation library; performing risk prediction optimization on the third frequency converter regulation library based on the frequency converter risk prediction model to obtain a fourth frequency converter regulation library; performing mutation adjustment on the stability frequency converter regulation library based on the first global frequency converter regulation library to obtain a fifth frequency converter regulation library; performing output stability adjustment optimization on the fifth frequency converter regulation library based on the output stability evaluation network to obtain a sixth frequency converter regulation library; and performing intersection identification on the fourth and sixth frequency converter regulation libraries to generate the second global frequency converter regulation library.

[0013] Preferably, the inverter risk prediction model includes inverter risk factors, which include voltage fluctuation risk, current surge risk, and harmonic distortion risk.

[0014] Preferably, the frequency conversion optimization evaluation conditions include output stability weight, voltage fluctuation risk weight, current surge risk weight, and harmonic distortion risk weight.

[0015] This application also provides an electronic device, which includes: a memory for storing executable instructions; and a processor for executing the executable instructions stored in the memory to implement the power quality detection method in the frequency converter provided in this application.

[0016] This application proposes a method and equipment for power quality detection in frequency converters. It involves real-time monitoring of the frequency converter in the target device to obtain monitoring data; using this data, multi-dimensional power quality detection is performed to construct a power quality map; this map is then used to perform multi-dimensional fault prediction on the frequency converter, resulting in a fault map; combined with a risk prediction model, the frequency converter's control decisions are optimized to obtain a fault suppression frequency converter regulation library; based on load prediction data of the target device, output stability adjustment is optimized for the frequency converter control decisions, resulting in a stability frequency converter regulation library; and based on optimal frequency converter evaluation conditions, the fault suppression and stability frequency converter regulation libraries are globally optimized to obtain the frequency converter regulation optimization results, which are then used to execute optimized frequency converter control. This solves the technical problems of existing technologies, such as single-dimensional power quality detection, insufficient fault prediction capabilities, and poor adaptability of frequency converter control regulation. This achieves the technical effects of improving the accuracy of power quality detection in frequency converters, enhancing the ability to predict and suppress accidents, and realizing stable adaptive optimization control of frequency converters. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0018] Figure 1 This is a flowchart illustrating the power quality detection method in a frequency converter provided in an embodiment of this application.

[0019] Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0020] Explanation of reference numerals in the attached drawings: Processor 31, Memory 32, Input device 33, Output device 34. Detailed Implementation

[0021] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application.

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" may be the same or different subsets of all possible embodiments and may be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0024] This application provides a method for power quality detection in frequency converters, such as... Figure 1 As shown, the method includes:

[0025] Step S100: Perform real-time monitoring of the inverter in the target equipment to obtain inverter monitoring data. It should be noted that the target equipment includes load devices such as motors, pumps, and fans. The monitoring device continuously collects operating status data from the inverter in the target equipment to obtain inverter monitoring data, providing basic data for subsequent power quality analysis. The monitoring device includes multiple sensors such as voltage sensors and current sensors. The inverter monitoring data includes real-time parameters such as real-time voltage parameters, real-time current parameters, real-time power parameters, and real-time temperature parameters.

[0026] Step S200: Perform multi-dimensional power quality detection on the inverter based on the inverter monitoring data to construct a power quality map. It should be noted that multi-dimensional power quality detection refers to a comprehensive evaluation of the inverter's output power from multiple perspectives, including but not limited to voltage fluctuations, current fluctuations, harmonic characteristics, and power factor. The specific process of multi-dimensional power quality detection includes: First, performing Fourier transform on the voltage and current signals in the inverter monitoring data to extract harmonic characteristics, including the amplitude and frequency distribution of each harmonic, to assess potential power interference. Second, analyzing voltage fluctuations and transient impacts by identifying events such as voltage dips, surges, and flicker, and quantifying their amplitude and duration to determine the stability of the inverter's power supply to the load. Then, calculating the power factor, active power ratio, and reactive power ratio to assess the inverter's energy efficiency. Finally, integrating the above multi-dimensional power quality indicators and generating a power quality map through visualization. The power quality profile includes data on the inverter's voltage fluctuation characteristics, current fluctuation characteristics, harmonic characteristics, and power factor characteristics. By performing multi-dimensional power quality detection on the inverter, the accuracy of power quality detection is effectively improved.

[0027] Step S300: Perform multi-dimensional fault prediction on the frequency converter based on the power quality map to obtain a frequency converter fault map. In one possible implementation, step S300 further includes steps S310 to S340. Step S310: Perform voltage fluctuation fault prediction on the frequency converter based on the power quality map to obtain a voltage fluctuation fault path. Step S310 includes steps S311 to S314. Step S311: Search for voltage fluctuation faults based on the frequency converter's family of frequency converters to obtain a historical voltage fluctuation fault set; Step S312: Train a fault tree based on the historical voltage fluctuation fault set to obtain a voltage fluctuation fault tree model; Step S313: Perform adversarial training on the voltage fluctuation fault tree model based on the historical voltage fluctuation fault set to obtain a voltage fluctuation fault inference model; Step S314: Output the voltage fluctuation fault path based on the power quality map and the voltage fluctuation fault inference model.

[0028] A family of frequency converters refers to a group of frequency converters that are similar in model, control method, and operating conditions. By searching historical operating records in the operational database of these family of frequency converters, a large amount of historical information on voltage fluctuation incidents can be collected. The historical voltage fluctuation incident set includes multiple types of historical voltage fluctuation incidents, such as voltage dips, voltage sags, voltage overshoots, and voltage flicker. The voltage fluctuation incident search can employ big data retrieval algorithms to automatically label and categorize historical voltage fluctuation incident information under similar operating conditions, ensuring the comprehensiveness and representativeness of the historical voltage fluctuation incident set.

[0029] Subsequently, a fault tree model was constructed and trained using a historical voltage fluctuation incident set. A fault tree is a fault causal model based on logical relationships, representing the formation mechanism of an incident through a hierarchical structure of "event-condition-effect." Logistic regression algorithms or Bayesian networks can be used to mine the causal relationships within the historical voltage fluctuation incident set, establishing correspondences between various voltage fluctuation events and potential triggers (such as harmonic distortion, grid fluctuations, load surges, etc.), thereby generating the voltage fluctuation fault tree model.

[0030] Furthermore, the robustness and generalization ability of the voltage fluctuation accident tree model are improved by adversarial training using a historical voltage fluctuation accident set. Specifically, an adversarial example generator is used to inject perturbations into the historical voltage fluctuation accident set. Perturbation injection includes superimposing random noise into the voltage fluctuation amplitude signal, nonlinearly deforming the voltage drop curve, anomalously enhancing the load impact data, and simulating grid harmonic interference and short-term voltage interruptions. Through these perturbation operations, an adversarial example set can be generated, which is used to simulate the complex behavior of voltage fluctuation accidents under extreme operating conditions. The adversarial example generator is a tool in the prior art that introduces perturbations into data samples to generate new "adversarial examples".

[0031] Adversarial sample sets are input into the voltage fluctuation fault tree model for iterative training. This enables the model to learn not only the fault formation path under normal conditions but also to maintain correct reasoning about the fault mechanism under disturbance, noise, and extreme load disturbance conditions. Gradient descent optimization algorithms, loss function weighting mechanisms, and robustness constraints are employed during training to improve the generalization and anti-interference capabilities of the voltage fluctuation fault prediction model. Finally, the voltage fluctuation fault tree model, after adversarial training, evolves into a voltage fluctuation fault prediction model. This model not only contains the causal structure of voltage fluctuation faults but also predicts the dynamic evolution trajectory of voltage fluctuation faults under different disturbance conditions, thereby enabling the prediction and source analysis of voltage fluctuation fault occurrence paths. Furthermore, power quality maps are input into the voltage fluctuation fault prediction model to obtain voltage fluctuation fault paths. A voltage fluctuation fault path is the logical chain of the occurrence, development, and propagation of voltage fluctuation events. It includes information on the voltage fluctuation trigger source, the voltage fluctuation propagation process, the response information of key nodes, and the fault outcome. By understanding the voltage fluctuation incident path, maintenance personnel can intuitively grasp the development trend of voltage fluctuation incidents and implement incident prevention measures in advance, thereby improving the stability and safety of inverter operation.

[0032] Step S320: Based on the power quality map, predict the current surge accident of the frequency converter to obtain the current surge accident path. Step S330: Based on the power quality map, predict the harmonic distortion accident of the frequency converter to obtain the harmonic distortion accident path. Step S340: Generate the frequency converter accident map based on the voltage fluctuation accident path, the current surge accident path, and the harmonic distortion accident path. Specifically, based on the power quality map, continue to predict the current surge accident and harmonic distortion accident of the frequency converter to obtain the current surge accident path and the harmonic distortion accident path. The specific implementation of steps S320 and S330 is similar to steps S311 to S314, which are all accident prediction processes of "retrieving historical accident sets → constructing an accident tree model → adversarial training → deducing and outputting accident paths", and will not be described again here. The current surge accident path includes the full-process prediction information of the occurrence, development, and propagation of the current surge accident. The harmonic distortion accident path includes the process prediction information of the harmonic distortion accident.

[0033] After acquiring the voltage fluctuation fault path, current surge fault path, and harmonic distortion fault path, the trigger source information of these fault paths is first uniformly encoded, including data such as inverter operating status, load changes, and external disturbances. Then, the propagation process information of these fault paths is modeled in detail, recording the changes in current and voltage over time. Next, the response information of key nodes is analyzed, including protection device actions, circuit breaker tripping, and oscillation amplitude, with the time sequence and amplitude of each node's response labeled. Simultaneously, the fault outcome information, such as equipment overcurrent, harmonic exceedances, and power quality degradation, is normalized. Finally, the above four types of information are correlated and mapped to form a complete inverter fault map, which is used for subsequent inverter control regulation and optimization decision analysis.

[0034] Step S400: Based on the inverter fault map, adjust and optimize the inverter control decision according to the inverter risk prediction model to obtain a fault suppression inverter regulation library. In one possible implementation, step S400 further includes steps S410 to S440. Step S410: Perform correlation feature analysis on the inverter control decision according to the inverter fault map to obtain a fault inverter control correlation map; Step S420: Perform adaptive adjustment on the inverter control decision according to the fault inverter control correlation map to obtain a first inverter regulation library; Step S430: Perform risk prediction on the first inverter regulation library according to the inverter risk prediction model to obtain an inverter regulation risk library; wherein, the inverter risk prediction model includes inverter risk factors, and the inverter risk factors include voltage fluctuation risk, current surge risk, and harmonic distortion risk. Step S440: Based on the variable frequency control risk library, optimize the variable frequency control risk of the first variable frequency control library according to the variable frequency control risk constraints to obtain the accident suppression variable frequency control library.

[0035] It should be noted that the frequency converter control decision includes multiple real-time control parameters of the frequency converter. Based on the frequency converter fault map, the frequency converter control decision is analyzed for correlation features. Specifically, multi-dimensional fault features such as voltage fluctuations, current surges, and harmonic distortion are first extracted from the frequency converter fault map. These fault features are then mapped to each real-time control parameter within the frequency converter control decision, quantifying the impact of each real-time control parameter on fault triggering, propagation, and critical node response. Real-time control parameters with high impact and direct effects on critical nodes in the fault path are identified as high-risk parameters. This establishes a correspondence between each real-time control parameter and potential faults, represented by a graph structure to form a fault frequency converter control correlation map. The fault frequency converter control correlation map includes the correlation and influence relationships between the frequency converter control decision and voltage fluctuation fault paths, current surge fault paths, and harmonic distortion fault paths. Furthermore, based on the fault frequency converter control correlation map, the frequency converter control decision is adaptively adjusted, offsetting high-risk parameters while keeping low-risk parameters unchanged, forming the first frequency converter regulation library. The first frequency converter regulation library includes multiple frequency converter control regulation strategies. Each frequency converter control adjustment strategy includes a set of adjusted control parameters.

[0036] Each frequency converter control strategy in the first frequency converter regulation library is input into the frequency converter risk prediction model. The frequency converter risk prediction model performs risk prediction on each frequency converter control strategy according to the frequency converter risk factors, thus obtaining the frequency converter regulation risk library. The frequency converter risk factors include voltage fluctuation risk, current surge risk, and harmonic distortion risk. The frequency converter regulation risk library includes multiple frequency converter regulation risk prediction results. Each frequency converter regulation risk prediction result includes the voltage fluctuation risk coefficient, current surge risk coefficient, and harmonic distortion risk coefficient corresponding to each frequency converter control strategy. The frequency converter risk prediction model is trained using a neural network, specifically including the following process: collecting multiple sets of frequency converter risk records as training datasets. Each set of frequency converter risk records includes historical frequency converter control strategies, and the corresponding historical voltage fluctuation risk coefficient, current surge risk coefficient, and harmonic distortion risk coefficient. Using historical frequency converter control strategies as input labels and historical risk coefficients for voltage fluctuations, current surges, and harmonic distortion as output labels, a frequency converter risk prediction model is obtained by continuously training a neural network to convergence based on multiple sets of frequency converter risk records. The frequency converter risk prediction model includes an input layer, a hidden layer, and an output layer, and is capable of multi-dimensional risk prediction based on frequency converter risk factors.

[0037] Subsequently, frequency converter regulation risk optimization is performed on the first frequency converter regulation library. Specifically, it is determined whether each frequency converter regulation risk prediction result in the library meets the frequency converter regulation risk constraints. If a frequency converter regulation risk prediction result meets the constraints, the corresponding frequency converter control regulation strategy is added to the fault suppression frequency converter regulation library. This effectively enhances the fault suppression capability of the frequency converter regulation and reduces the frequency converter fault risk. Frequency converter regulation risk constraints include pre-defined voltage fluctuation risk constraints, current surge risk constraints, and harmonic distortion risk constraints. The fault suppression frequency converter regulation library includes multiple frequency converter control regulation strategies that meet the frequency converter regulation risk constraints.

[0038] Step S500: Optimize the output stability of the frequency converter control decision based on the load prediction data of the target device to obtain a stability frequency converter adjustment library. In one possible implementation, step S500 further includes steps S510 to S550. Step S510: Adjust the frequency converter control decision based on the load prediction data to obtain a second frequency converter adjustment library; Step S520: Extract the Y-th frequency converter adjustment scheme from the second frequency converter adjustment library, where Y is a positive integer; Step S530: Based on the load prediction data, perform simulated adjustment on the frequency converter according to the Y-th frequency converter adjustment scheme to obtain the Y-th frequency converter simulation data; Step S540: Based on the Y-th frequency converter simulation data, obtain the Y-th output stability according to the output stability evaluation network; Step S550: If the Y-th output stability is greater than or equal to a predetermined output stability, add the Y-th frequency converter adjustment scheme to the stability frequency converter adjustment library.

[0039] Specifically, the load forecast data includes instantaneous load change rate, load curve trend, and load peak prediction information. Based on the load forecast data, various control parameters within the frequency converter control decision are adjusted, such as output voltage amplitude, output frequency, acceleration / deceleration time, overload protection threshold, and current limit. Multiple feasible control schemes are generated through different parameter combinations. These schemes constitute the second frequency converter regulation library, providing a candidate set for subsequent stability assessment. The second frequency converter regulation library includes multiple frequency converter regulation schemes. Each frequency converter regulation scheme is a feasible adjustment strategy corresponding to the load forecast data. Each frequency converter regulation scheme includes parameters such as voltage, frequency, acceleration / deceleration time, and protection threshold.

[0040] Extract any frequency converter control scheme from the second frequency converter control library as the Y-th frequency converter control scheme. Then, input the Y-th frequency converter control scheme into the frequency converter simulation platform, along with the load prediction data. Under the load prediction data, the frequency converter simulation platform simulates the frequency converter according to the Y-th frequency converter control scheme, and collects simulated information such as the frequency converter's simulated output voltage, simulated current waveform, and simulated frequency changes as the Y-th frequency converter simulation data. The frequency converter simulation platform is a hardware and software combined platform used in the prior art for simulating and testing frequency converter control strategies. By constructing virtual models of the frequency converter and the target equipment, the frequency converter simulation platform allows the adjustment of control parameters to be verified and optimized under non-actual operating conditions.

[0041] The simulated data of the Y-th frequency converter is used as input information and fed into the output stability evaluation network to obtain the Y-th output stability. The output stability evaluation network is trained using a neural network; the specific training process is similar to that of the frequency converter risk prediction model and will not be elaborated here. The output stability evaluation network includes an input layer, a hidden layer, and an output layer. The Y-th output stability is a quantitative evaluation of the ability of the Y-th frequency converter regulation scheme to maintain stable operation of the frequency converter under the load prediction data. The larger the Y-th output stability, the stronger the ability of the Y-th frequency converter regulation scheme to maintain stable operation of the frequency converter under the load prediction data. Then, it is determined whether the Y-th output stability is greater than or equal to a predetermined output stability. If the Y-th output stability is greater than or equal to the predetermined output stability, the Y-th frequency converter regulation scheme is added to the stability frequency converter regulation library. Using the stability frequency converter regulation library can effectively improve the load adaptability of the frequency converter regulation and ensure the output stability of the frequency converter under the load prediction data. If the Y-th output stability is less than the predetermined output stability, the Y-th frequency converter regulation scheme is discarded. The predetermined output stability can be adaptively set and determined. The stability variable frequency control library includes multiple variable frequency control schemes with a stability greater than or equal to the predetermined output stability.

[0042] Step S600: Globally optimize the accident suppression frequency converter library and the stability frequency converter library according to the frequency converter optimization evaluation conditions to obtain the frequency converter optimization result, and perform frequency converter optimization control based on the frequency converter optimization result. In one possible implementation, step S600 further includes steps S610 to S640. Step S610: Perform intersection identification on the accident suppression frequency converter library and the stability frequency converter library to obtain a first global frequency converter library. Specifically, when performing intersection identification on the accident suppression frequency converter library and the stability frequency converter library, firstly, all schemes in the accident suppression frequency converter library and the stability frequency converter library are standardized according to a unified format, and then all schemes in the accident suppression frequency converter library and the stability frequency converter library are compared one by one, and the intersection schemes are added to the first global frequency converter library. The first global frequency converter library includes multiple global frequency converter schemes. Each global frequency converter scheme is the intersection scheme of the accident suppression frequency converter library and the stability frequency converter library. By identifying the intersection of the accident suppression frequency converter control library and the stability frequency converter control library, the generated first global frequency converter control library takes into account both the safety of frequency converter operation and the output stability. It achieves the goal of maintaining the stability of the frequency converter while reducing the risks of voltage fluctuations, current surges and harmonic distortion, and effectively improves the control reliability of the frequency converter.

[0043] Step S620: Based on the first global frequency converter control library, perform global multiplication and optimization on the accident suppression frequency converter control library and the stability frequency converter control library to obtain a second global frequency converter control library. In one possible implementation, step S620 further includes steps S621 to S626. Step S621: Based on the first global frequency converter control library, perform frequency converter control parameter difference detection on the accident suppression frequency converter control library to obtain a first frequency converter difference vector set. Specifically, compare each global frequency converter control scheme in the first global frequency converter control library with each scheme in the accident suppression frequency converter control library parameter by parameter to identify the control difference parameters between the two libraries, forming a frequency converter difference vector, which is used as a reference for subsequent cross-mutation operations. The first frequency converter difference vector set includes multiple frequency converter difference vectors. Each frequency converter difference vector includes multiple control difference parameters between each global frequency converter control scheme in the first global frequency converter control library and each scheme in the accident suppression frequency converter control library.

[0044] Step S622: Perform cross-mutation on the accident suppression frequency converter regulation library according to the first frequency converter difference vector set to obtain a third frequency converter regulation library. The cross-mutation process of the accident suppression frequency converter regulation library includes: adjusting the amplitude of each frequency converter difference vector in the first frequency converter difference vector set multiple times to obtain multiple cross-candidate parameters. Combine the multiple cross-candidate parameters with the corresponding parameters of other schemes in the accident suppression frequency converter regulation library to generate multiple frequency converter regulation mutation schemes. The third frequency converter regulation library includes multiple frequency converter regulation mutation schemes. At the same time, the third frequency converter regulation library can also be slightly and randomly fine-tuned to cover more feasible solution space and expand the optimization range. Step S623: Perform risk prediction optimization on the third frequency converter regulation library according to the frequency converter risk prediction model to obtain a fourth frequency converter regulation library. Perform multi-dimensional risk prediction on each frequency converter regulation mutation scheme in the third frequency converter regulation library through the frequency converter risk prediction model, and select frequency converter regulation mutation schemes that meet the frequency converter regulation risk constraints to add to the fourth frequency converter regulation library. The specific implementation process of step S623 is similar to steps S410 to S440, and will not be repeated here.

[0045] Step S624: The stability frequency converter adjustment library is mutated and adjusted according to the first global frequency converter adjustment library to obtain the fifth frequency converter adjustment library. The process of mutating and adjusting the stability frequency converter adjustment library according to the first global frequency converter adjustment library is similar to steps S621 and S622, involving frequency converter adjustment parameter difference detection and cross-mutation, etc., and will not be elaborated here. Step S625: The output stability adjustment of the fifth frequency converter adjustment library is optimized according to the output stability evaluation network to obtain the sixth frequency converter adjustment library. The output stability evaluation network is used to evaluate the output stability of each scheme in the fifth frequency converter adjustment library, and schemes with a stability greater than or equal to a predetermined output stability are added to the sixth frequency converter adjustment library. The specific implementation process of step S625 is similar to steps S510 to S550, and will not be elaborated here. Step S626: The intersection of the fourth frequency converter adjustment library and the sixth frequency converter adjustment library is identified to generate the second global frequency converter adjustment library. The second global frequency converter adjustment library includes multiple intersection schemes between the fourth and sixth frequency converter adjustment libraries.

[0046] Step S630: Based on the second global frequency converter regulation library, continue to perform global multiplication optimization on the fault suppression frequency converter regulation library and the stability frequency converter regulation library until multiple global frequency converter regulation libraries that meet the global multiplication optimization number are obtained. Step S640: According to the frequency converter optimality evaluation conditions, perform frequency converter optimality maximization optimization on the multiple global frequency converter regulation libraries to obtain the frequency converter regulation optimization result. The frequency converter optimality evaluation conditions include output stability weight, voltage fluctuation risk weight, current surge risk weight, and harmonic distortion risk weight.

[0047] The specific implementation process of step S630 is similar to steps S621 to S626. First, using the second global frequency converter regulation library as a benchmark, the difference parameters of the accident suppression frequency converter regulation library are identified to form a difference vector, and candidate schemes are generated through cross-mutation. Then, the frequency converter risk prediction model is used to optimize the risk of the candidate schemes, obtaining the risk-optimized result. Simultaneously, using the second global frequency converter regulation library as a reference, the stability frequency converter regulation library is mutated to generate new candidate schemes, and the new candidate schemes are screened through the output stability evaluation network to obtain the stability-optimized result. Finally, the intersection of the risk-optimized result and the stability-optimized result is identified to generate a new global frequency converter regulation library, and the above process is repeated until multiple global frequency converter regulation libraries satisfying the global multiplication optimization number are obtained. The global multiplication optimization number is a positive integer and can be adaptively set and determined.

[0048] Furthermore, based on the frequency conversion optimization evaluation criteria, a frequency conversion optimization evaluation function is constructed. This function comprises: Output stability weight × Output stability – Voltage fluctuation risk weight × Voltage fluctuation risk coefficient – ​​Current surge risk weight × Current surge risk coefficient – ​​Harmonic distortion risk weight × Harmonic distortion risk coefficient = Frequency conversion optimization coefficient. The evaluation criteria include pre-set weights for output stability, voltage fluctuation risk, current surge risk, and harmonic distortion risk. Subsequently, the frequency conversion optimization coefficients corresponding to each scheme within multiple global frequency conversion regulation libraries are calculated based on the evaluation function. The scheme with the highest optimization coefficient within each library is then selected as the frequency conversion regulation optimization result. Finally, the optimization result is applied to the inverter's operation control to achieve real-time control optimization. This ensures output stability while effectively suppressing power quality incidents such as voltage fluctuations, current surges, and harmonic distortion, achieving a globally optimized control effect.

[0049] In the above text, refer to Figure 1 A power quality detection method in a frequency converter according to an embodiment of the present invention is described in detail. Next, reference will be made to... Figure 2 An electronic device according to an embodiment of the present invention is described. Figure 2 The electronic device shown is merely an example and should not be construed as limiting the functionality or scope of the embodiments of the present invention. Figure 2 As shown, the electronic device includes a processor 31, a memory 32, an input device 33, and an output device 34; the number of processors 31 in the electronic device can be one or more. Figure 2 Taking a processor 31 as an example, the processor 31, memory 32, input device 33, and output device 34 in an electronic device can be connected via a bus or other means. Figure 2 Taking the example of a connection between China and Israel via a bus.

[0050] The memory 32, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the power quality detection method in the frequency converter in this embodiment of the invention. The processor 31 executes various functional applications and data processing of the computer device by running the software programs, instructions, and modules stored in the memory 32, thereby realizing the power quality detection method in the frequency converter described above.

[0051] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for detecting power quality in a frequency converter, characterized in that, The method includes: Real-time monitoring of the frequency converter of the target equipment is performed to obtain frequency converter monitoring data; Based on the inverter monitoring data, perform multi-dimensional power quality detection on the inverter and construct a power quality map. Based on the power quality spectrum, multi-dimensional fault prediction is performed on the frequency converter to obtain the frequency converter fault spectrum; Based on the inverter fault map, the inverter control decision of the inverter is adjusted and optimized according to the inverter risk prediction model to obtain the fault suppression inverter regulation library. Based on the load prediction data of the target equipment, the frequency converter control decision is optimized by output stability adjustment to obtain a stability frequency converter adjustment library. Based on the frequency conversion optimization evaluation conditions, the accident suppression frequency conversion control library and the stability frequency conversion control library are subjected to global breeding optimization to obtain the frequency conversion control optimization results, and frequency conversion optimization control is performed based on the frequency conversion control optimization results. Based on the power quality spectrum, multi-dimensional fault prediction is performed on the frequency converter to obtain a frequency converter fault spectrum, including: Based on the power quality map, voltage fluctuation accidents are predicted for the frequency converter to obtain voltage fluctuation accident paths. Based on the power quality map, the inverter is used to predict current surge accidents and obtain the current surge accident path. Based on the power quality spectrum, the frequency converter is used to predict harmonic distortion accidents and obtain the harmonic distortion accident path. The inverter fault map is generated based on the voltage fluctuation fault path, the current surge fault path, and the harmonic distortion fault path.

2. The power quality detection method in a frequency converter as described in claim 1, characterized in that, Based on the power quality map, voltage fluctuation fault prediction is performed on the frequency converter to obtain the voltage fluctuation fault path, including: Based on the frequency converters of the same family, a voltage fluctuation accident search is performed to obtain a historical voltage fluctuation accident set. Based on the historical voltage fluctuation incident set, an incident tree is trained to obtain a voltage fluctuation incident tree model; The voltage fluctuation fault tree model is subjected to adversarial training based on the historical voltage fluctuation fault set to obtain a voltage fluctuation fault inference model. Based on the power quality map, and according to the voltage fluctuation accident simulation model, the voltage fluctuation accident path is output.

3. The power quality detection method in a frequency converter as described in claim 1, characterized in that, Based on the inverter fault map, the inverter control decision is adjusted and optimized according to the inverter risk prediction model to obtain a fault suppression inverter regulation library, including: Based on the inverter fault map, the inverter control decision is analyzed for correlation features to obtain the fault inverter control correlation map; Based on the accident variable frequency control correlation diagram, the variable frequency control decision is adaptively adjusted to obtain the first variable frequency control library; Based on the inverter risk prediction model, risk prediction is performed on the first inverter regulation library to obtain the inverter regulation risk library. Based on the variable frequency control risk library, the first variable frequency control library is optimized for variable frequency control risk according to the variable frequency control risk constraints to obtain the accident suppression variable frequency control library.

4. The power quality detection method in a frequency converter as described in claim 1, characterized in that, Based on the load prediction data of the target equipment, the frequency converter control decision is optimized by adjusting the output stability to obtain a stability frequency converter adjustment library, including: The frequency converter control decision is adjusted based on the load prediction data to obtain a second frequency converter adjustment library; The Yth frequency conversion control scheme is extracted from the second frequency conversion control library, where Y is a positive integer; Based on the load prediction data, the inverter is simulated and adjusted according to the Yth inverter adjustment scheme to obtain the Yth inverter simulation data. Based on the simulated data of the Y-th frequency converter, the output stability of the Y-th frequency converter is obtained according to the output stability evaluation network. If the Y-th output stability is greater than or equal to the predetermined output stability, the Y-th frequency conversion adjustment scheme is added to the stability frequency conversion adjustment library.

5. The power quality detection method in a frequency converter as described in claim 1, characterized in that, Based on the frequency conversion optimality evaluation criteria, a global multiplication optimization is performed on the fault suppression frequency conversion control library and the stability frequency conversion control library to obtain the frequency conversion control optimization results, including: The intersection of the accident suppression frequency converter control library and the stability frequency converter control library is identified to obtain the first global frequency converter control library; Based on the first global frequency conversion control library, the accident suppression frequency conversion control library and the stability frequency conversion control library are globally multiplied and optimized to obtain the second global frequency conversion control library. Based on the second global frequency conversion control library, the fault suppression frequency conversion control library and the stability frequency conversion control library are further subjected to global breeding optimization until multiple global frequency conversion control libraries that meet the number of global breeding optimizations are obtained. Based on the frequency conversion optimality evaluation conditions, the multiple global frequency conversion adjustment libraries are optimized to maximize the frequency conversion optimality, and the frequency conversion adjustment optimization result is obtained.

6. The power quality detection method in a frequency converter as described in claim 5, characterized in that, Based on the first global frequency converter control library, a second global frequency converter control library is obtained by globally multiplying and optimizing the fault suppression frequency converter control library and the stability frequency converter control library, including: Based on the first global frequency conversion control library, the frequency conversion control parameter difference detection is performed on the accident suppression frequency conversion control library to obtain the first frequency conversion difference vector set; Based on the first frequency conversion difference vector set, the accident suppression frequency conversion adjustment library is cross-mutated to obtain the third frequency conversion adjustment library; Based on the inverter risk prediction model, the third inverter regulation library is optimized for risk prediction to obtain the fourth inverter regulation library. The stability frequency conversion control library is modified according to the first global frequency conversion control library to obtain the fifth frequency conversion control library. The output stability adjustment of the fifth frequency converter library is optimized based on the output stability evaluation network to obtain the sixth frequency converter library. The intersection of the fourth frequency converter control library and the sixth frequency converter control library is identified to generate the second global frequency converter control library.

7. The power quality detection method in a frequency converter as described in claim 1, characterized in that, The inverter risk prediction model includes inverter risk factors, which include voltage fluctuation risk, current surge risk, and harmonic distortion risk.

8. The power quality detection method in a frequency converter as described in claim 1, characterized in that, The evaluation criteria for frequency conversion optimization include output stability weight, voltage fluctuation risk weight, current surge risk weight, and harmonic distortion risk weight.

9. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the power quality detection method in the frequency converter according to any one of claims 1 to 8.

Images