Electric reactor load adjusting method and system based on electrical performance optimization

Abstract

The invention discloses an electric reactor load adjusting method and system based on electrical performance optimization, and relates to the technical field of power equipment control and optimization, and the method comprises the steps: carrying out the real-time monitoring of an electric reactor, and obtaining the monitoring data and load data; and obtaining an evaluation result through the three electrical performance evaluation channels, and judging whether an electrical performance evaluation constraint is met or not. If not, generating a first optimization strategy according to the load data and the evaluation constraint; and if yes, trend prediction is carried out based on the load data, and a second optimization strategy is generated in combination with the evaluation constraint. And finally, carrying out adaptive load adjustment on the reactor according to the first optimization strategy and the second optimization strategy. The technical problems that in the prior art, due to the fact that adjustment of an electric reactor cannot adapt to complex load changes, the adjustment precision of the electric reactor is low, and the electrical performance is poor are solved, and the technical effects of improving the adjustment precision of the electric reactor and optimizing the electrical performance are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of power equipment control and optimization, and in particular to a reactor load regulation method and system based on electrical performance optimization. Background Art

[0002] With the rapid development of power systems, reactors are important devices for regulating reactive power, limiting short-circuit current and improving grid stability. Their operating efficiency and regulation accuracy have an important impact on the overall performance of the system. However, when facing complex load changes and diverse operating conditions, the existing technology has problems such as insufficient regulation accuracy, limited electrical performance, poor dynamic adaptability and uncontrollable operating status. Traditional methods rely on fixed control strategies or simple feedback mechanisms, which are difficult to achieve fine-grained regulation. They lack the ability to predict load change trends and comprehensive evaluation of operating efficacy, energy efficiency and safety, resulting in delayed equipment response, high energy consumption, low operating efficiency, and even regulation failure or equipment failure under extreme conditions. Therefore, realizing real-time monitoring, comprehensive performance evaluation and dynamic adaptive regulation of reactors, and improving regulation accuracy and electrical performance have become technical problems that need to be solved in the field of power system equipment control.

[0003] In the current related technologies, there is a technical problem that the regulation of the reactor cannot adapt to complex load changes, resulting in low regulation accuracy and poor electrical performance of the reactor. Summary of the invention

[0004] The present application solves the technical problem in the prior art that the regulation of the reactor cannot adapt to complex load changes, resulting in low regulation accuracy and poor electrical performance of the reactor, by providing a reactor load regulation method and system based on electrical performance optimization.

[0005] The present application provides a reactor load regulation method based on electrical performance optimization, including: Monitor the reactor in real time to obtain reactor monitoring data and reactor load data; input the reactor monitoring data into the three channels of electrical performance evaluation to obtain electrical performance evaluation results; determine whether the electrical performance evaluation results meet the electrical performance evaluation constraints; if the electrical performance evaluation results do not meet the electrical performance evaluation constraints, optimize the reactor according to the reactor load data and the electrical performance evaluation constraints to generate a first optimization strategy for reactor adjustment; if the electrical performance evaluation results meet the electrical performance evaluation constraints, perform trend prediction according to the reactor load data to obtain load trend prediction results, and optimize the reactor in combination with the electrical performance evaluation constraints to generate a second optimization strategy for reactor adjustment; perform adaptive load adjustment on the reactor according to the first optimization strategy for reactor adjustment and the second optimization strategy for reactor adjustment.

[0006] The present application provides a reactor load regulation system based on electrical performance optimization, including: A real-time monitoring module, the real-time monitoring module is used to monitor the reactor in real time and obtain the reactor monitoring data and the reactor load data; an evaluation result acquisition module, the evaluation result acquisition module is used to input the reactor monitoring data into the three channels of electrical performance evaluation to obtain the electrical performance evaluation result; an evaluation result judgment module, the evaluation result judgment module is used to judge whether the electrical performance evaluation result meets the electrical performance evaluation constraint; an adjustment optimization module, the adjustment optimization module is used to adjust the electrical performance evaluation result according to the reactor load data and the electrical performance evaluation constraint if the electrical performance evaluation result does not meet the electrical performance evaluation constraint. The reactor is regulated and optimized to generate a first optimization strategy for reactor regulation; a load trend prediction result acquisition module is used to perform trend prediction according to the reactor load data if the electrical performance evaluation result meets the electrical performance evaluation constraint, obtain the load trend prediction result, and regulate the reactor in combination with the electrical performance evaluation constraint to generate a second optimization strategy for reactor regulation; an adaptive load regulation module is used to perform adaptive load regulation on the reactor according to the first optimization strategy for reactor regulation and the second optimization strategy for reactor regulation.

[0007] The reactor load regulation method and system based on electrical performance optimization proposed in this application is to first monitor the reactor in real time to obtain monitoring data and load data; obtain evaluation results through three channels of electrical performance evaluation, and determine whether the electrical performance evaluation constraints are met. If not, a first optimization strategy is generated based on the load data and the evaluation constraints; if satisfied, a trend prediction is performed based on the load data, and a second optimization strategy is generated in combination with the evaluation constraints. Finally, the reactor is adaptively load regulated according to the first and second optimization strategies, achieving the technical effect of improving the regulation accuracy of the reactor and optimizing electrical performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0009] Figure 1 A schematic flow chart of a reactor load regulation method based on electrical performance optimization provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a reactor load regulation system based on electrical performance optimization provided in an embodiment of the present application.

[0010] Explanation of the accompanying drawings: real-time monitoring module 10 , evaluation result acquisition module 20 , evaluation result judgment module 30 , adjustment optimization module 40 , load trend prediction result acquisition module 50 , adaptive load adjustment module 60 . DETAILED DESCRIPTION

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0012] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0013] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, 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 clearly listed, but may include other steps or modules that are not clearly 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 those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0014] The present application embodiment provides a reactor load adjustment method based on electrical performance optimization, such as Figure 1 As shown, the method includes: Step S100, monitor the reactor in real time to obtain reactor monitoring data and reactor load data. Specifically, according to the model, specification, rated parameters of the reactor and the complexity of the power system in which it is located, the appropriate monitoring equipment is automatically selected and configured. For electrical parameter monitoring, the system uses a voltage transformer with an accuracy of up to ±0.1%, which is automatically installed on the insulating bracket at the outlet end of the reactor winding to measure the voltage. At the same time, a current transformer with an accuracy of ±0.2% is selected, which is automatically mounted on the incoming wire and ensured by a fixing device to measure the current. For load data collection, the system uses an intelligent load sensor, which combines a pressure sensor based on the strain gauge principle and a current sensor based on the electromagnetic induction principle: the strain gauge sensor is automatically fixed at the cable connection clamp, and the electromagnetic induction current sensor is automatically wrapped around the cable and wrapped with electromagnetic shielding material to reduce external interference, thereby accurately collecting load data. After the equipment is installed and debugged, the monitoring system automatically collects data every 5 milliseconds, and the analog signal of the sensor is transmitted to the data acquisition terminal through a shielded cable or a wireless communication module (such as Bluetooth 5.0, ZigBee professional frequency band module). The data acquisition terminal automatically runs the finite impulse response filtering algorithm to denoise the voltage and current data, and corrects and filters the collected data in combination with the outlier discrimination algorithm.

[0015] Step S200, input the reactor monitoring data into the three channels of electrical performance evaluation to obtain the electrical performance evaluation results. Specifically, according to the reactor design standards, operating specifications and power system requirements, the system automatically builds a three-channel system for electrical performance evaluation and clarifies the pre-evaluation indicators of each channel function. The reactor efficacy evaluation channel automatically constructs an efficacy evaluation model library; the energy efficiency evaluation channel has a built-in energy loss algorithm; and the operation safety evaluation channel presets risk warning indicators. The collected reactor monitoring data is format-standardized in the data processing module, and an integrity check is automatically performed. After repairing a small amount of missing data, the sorted data is input into each channel in order. Then, each channel independently completes the evaluation calculation: the efficacy evaluation channel inputs the data into multiple models in the model library in turn, and obtains the reactor efficacy coefficient through weighted average calculation; the energy efficiency evaluation channel automatically calculates the copper loss and iron loss according to the energy loss algorithm, combined with voltage, current, temperature and other data, so as to obtain the energy efficiency coefficient; the operation safety evaluation channel compares the monitoring data with the preset risk warning indicators, and outputs the operation safety factor in combination with the abnormality detection algorithm. Finally, the system integrates the three coefficients and adds metadata in a structured format to output the electrical performance evaluation results, providing a basis for subsequent automated adjustment and optimization operations. In actual operation, the system dynamically adjusts and optimizes the entire process according to changes in system status to ensure stable operation of the power system and improve overall efficiency.

[0016] In a possible implementation, the reactor monitoring data is input into the three channels of electrical performance evaluation to obtain the electrical performance evaluation results. Step S200 further includes step S210, and the three channels of electrical performance evaluation include a reactor efficacy evaluation channel, a reactor energy efficiency evaluation channel, and a reactor operation safety evaluation channel. Specifically, based on the role of the reactor in the power system, the system automatically analyzes its design data, historical operation data, and industry standards, and constructs three channels of electrical performance evaluation from three key dimensions: efficacy, energy efficiency, and operation safety. Through the built-in data mining and feature extraction algorithm, the system automatically extracts key characteristic parameters such as reactive power output, power factor optimization, and harmonic suppression effect. Based on electromagnetic theory and power system analysis principles, it builds multiple efficacy evaluation models (such as models based on equivalent circuits and magnetic saturation characteristics) to collaboratively calculate and generate the reactor efficacy coefficient. At the same time, guided by the law of conservation of energy, the system uses high-precision current transformers to monitor winding currents and calculates copper losses in combination with known resistance values, uses hysteresis and eddy current loss formulas to calculate iron losses, and uses spectrum analysis algorithms to estimate stray losses. The three are added together and divided by the total input power to generate an energy efficiency coefficient. In terms of operational safety, the system extracts key safety monitoring indicators through fault tree analysis and failure mode analysis, collects data in real time with high-sensitivity sensors, and triggers alarms when compared with safety thresholds. At the same time, it uses machine learning anomaly detection algorithms to model based on historical data, detect potential anomalies in advance, and output operational safety factors. Finally, the system integrates the efficacy coefficient, energy efficiency coefficient, and operational safety factor to generate comprehensive electrical performance evaluation results, providing a scientific basis for the automatic adjustment and optimization of reactors.

[0017] Step S220, input the reactor monitoring data into the reactor efficacy evaluation channel to obtain the reactor efficacy coefficient. Specifically, after the reactor monitoring system collects data, the original data needs to be preprocessed. Use the data cleaning algorithm to eliminate erroneous data caused by sensor failure, electromagnetic interference, etc., and compare and verify with historical data and peripheral sensor data to determine and remove abnormal values; for a small amount of missing data, use linear interpolation or fill in the mean of adjacent time period data to ensure continuity and integrity, and then standardize the data format and unify it into the international unit standard format to prepare for entering the evaluation channel. Next, multiple sub-models based on different theoretical algorithms are preset in the reactor efficacy evaluation channel. According to the reactor type, operating conditions and past model performance, a group of models that are suitable for the current situation should be accurately selected. For example, for iron core reactors with frequent voltage fluctuations, a model that considers magnetic saturation characteristics and is sensitive to voltage should be selected. When the load is mostly resistive, a model that is good at handling resistive loads based on the equivalent circuit principle should be used. At the same time, some parameters are dynamically adjusted according to the characteristics of real-time monitoring data. Finally, the preprocessed and model-adapted data is input into the sub-model. Each sub-model calculates the local efficacy evaluation coefficient according to its own algorithm, such as the reactive power contribution coefficient based on the equivalent circuit model and the power factor improvement coefficient considering the magnetic saturation effect. The weighted average method is used to integrate them. The weights are determined based on the comprehensive consideration of the accuracy and scope of application of each sub-model, and the reactor efficacy coefficient is calculated to provide a basis for subsequent operations.

[0018] Step S230, input the reactor monitoring data into the reactor energy efficiency evaluation channel to obtain the reactor energy efficiency coefficient. Specifically, after obtaining the original monitoring data from the real-time monitoring system of the reactor, comprehensive data preparation and preprocessing must be performed first. Use professional data verification algorithms to check the integrity of data such as voltage, current, temperature, etc., mark possible missing or erroneous values, use scientific methods such as linear interpolation to supplement a small amount of missing data, and use filtering algorithms to remove high-frequency noise caused by electromagnetic interference, and convert all data into a unified standard format and unify the dimensions. Then, the preprocessed monitoring data enters the reactor energy efficiency evaluation channel. The analysis program in the channel first extracts key parameters closely related to energy loss, and accurately calculates copper loss according to Joule's law through high-precision current transformers combined with known winding resistance values. The iron loss is accurately estimated using hysteresis and eddy current loss formulas based on voltage, current data and core characteristic parameters, and the stray loss is evaluated with the help of spectrum analysis technology combined with the stray loss empirical estimation model. Finally, the copper loss, iron loss and stray loss are added together to obtain the total energy loss. The total input electrical energy is calculated by integrating the voltage and current in a specific time period. According to the energy efficiency calculation formula, the total energy loss is divided by the total electrical energy to obtain the reactor energy efficiency coefficient. This coefficient can be used as a key basis for judging the operating status of the reactor and formulating optimization strategies, helping to save energy and reduce emissions in the power system and promote industry development.

[0019] Step S240, input the reactor monitoring data into the reactor operation safety evaluation channel, and output the reactor operation safety factor. Specifically, after the real-time data from each monitoring point of the reactor, such as winding temperature, core temperature, vibration data, insulation resistance value, etc., are transmitted to the operation safety evaluation channel, the data integrity is first verified at the channel entrance using a cyclic redundancy check algorithm, combined with the normal operating parameter range, and the obvious abnormal data is screened by threshold judgment, and the fluctuating data is processed using a sliding average filter algorithm. Then enter the rule-based safety indicator comparison link, compare the temperature with the upper limit of the winding and core temperature, compare the vibration data with the standard vibration range, and compare the insulation resistance value with the minimum insulation resistance standard. Once the standard is exceeded, high temperature, vibration abnormality, and insulation warnings are triggered respectively. At the same time, historical operation and fault data are used to train machine learning models such as support vector machines and autoencoders. Even if the data does not touch the hard threshold, the model will give early warnings based on the dynamic changes of the data and the association of parameters to find hidden abnormalities. Finally, the two-stage detection results are integrated, and weights are set for various safety hazards according to factors such as the degree of threat. The weighted sum is used and a scoring conversion formula is used to output an operating safety factor between 0 and 1. Based on this, operation and maintenance personnel can understand the situation and take measures to ensure the reliable operation of the power system.

[0020] Step S250, output the reactor efficacy coefficient, the reactor energy efficiency coefficient and the reactor operation safety factor as the electrical performance evaluation result. Specifically, the system automatically collects basic data related to the reactor efficacy coefficient, energy efficiency coefficient and operation safety factor through high-precision power monitoring equipment, wherein the efficacy coefficient is obtained by monitoring reactive power through multiple groups of sensors and taking the average, the energy efficiency coefficient is obtained by automatically calibrating the mutual inductor and verifying the winding resistance measurement, and the operation safety factor is obtained by deploying multiple types of sensors to monitor key indicators in real time, and triggering the data re-collection mechanism when an abnormality occurs. After collecting the data, the system converts the efficacy coefficient according to the preset standard, converts it into the form of reactive power compensation ratio and is accurate to three decimal places; the energy efficiency coefficient is calculated as the ratio of unit power loss to input power and retains appropriate accuracy; the operation safety factor is normalized to a value range of 0-1, and a timestamp is added uniformly. Subsequently, the system generates a visualization interface suitable for power system monitoring, which is divided into three areas: the upper left corner uses green, yellow and red to dynamically mark the power efficiency coefficient value; the upper right corner uses a gradient from dark blue to dark red to display the energy efficiency coefficient; the bottom center uses a dynamic signal light icon to change color according to the operating safety factor and automatically mark the identification code and time information accurate to the minute. Finally, the system pushes the visualization interface data to the large screen in the main control room and the handheld intelligent inspection terminal of the operation and maintenance personnel in real time through a high-speed transmission network, making it convenient to grasp the status of the reactor in real time and quickly handle abnormalities, ensuring the efficient and stable operation of the power system.

[0021] In a possible implementation, the reactor monitoring data is input into the reactor efficacy evaluation channel to obtain the reactor efficacy coefficient. Step S220 further includes step S221, and the reactor efficacy evaluation channel includes S reactor efficacy evaluation models, where S is a positive integer greater than 1. Specifically, the system automatically determines that multiple evaluation models need to be constructed based on the characteristics of various application scenarios of reactors in industrial production, substations, residential communities, etc., and the emphasis on reactor efficacy under different working conditions (such as industrial scenarios focusing on rapid reactive power regulation and residential power supply focusing on harmonic suppression), and specifies that the number of models is a positive integer S greater than 1. The system collects massive operating data such as current and voltage under different working conditions and performance cases of the same type of reactors through a large-scale data acquisition module, and automatically parses professional literature and industry standards to extract key evaluation indicators such as reactive power response time and harmonic attenuation rate, providing theoretical and data support for model construction. Subsequently, the system combines machine learning algorithms with mathematical modeling techniques to automatically generate multiple efficacy evaluation models: for example, a reactive power dynamic compensation model is constructed through a neural network algorithm, and a harmonic control model is constructed through regression analysis methods. Each model relies on a unique structure, parameters, and algorithm to accurately measure the efficacy of the reactor from different dimensions, ultimately forming a reactor efficacy evaluation channel, providing a scientific basis for subsequent performance optimization and operation monitoring.

[0022] Step S222, input the reactor monitoring data into the S reactor efficacy evaluation models to obtain S reactor efficacy evaluation coefficients. Specifically, the system deploys high-precision power monitoring equipment to collect data such as current, voltage, reactive power, harmonic content and temperature during reactor operation in real time, and automatically performs preprocessing, such as filtering and denoising, standard correction and model calibration, to ensure that the data meets the model input requirements. Subsequently, the system extracts key data such as reactive power time series and harmonic spectrum according to the characteristics and input requirements of the S evaluation models, and inputs them into the corresponding model after packaging in a format. The model generates efficacy evaluation coefficients through built-in algorithm calculations to reflect reactive power compensation efficiency and harmonic governance capabilities. The generated coefficients are automatically compared and verified with historical data. If abnormal fluctuations are found, the system back-checks the input data or model status to ensure that the evaluation results are true and reliable.

[0023] Step S223, perform mean calculation based on the S reactor efficiency evaluation coefficients to generate the reactor efficiency coefficient. Specifically, after obtaining the S reactor efficiency evaluation coefficients, the system automatically reviews the S coefficients based on the standard coefficient interval of the same type of reactor under similar working conditions, focusing on checking whether the numerical range is reasonable. Once it is found that a coefficient deviates too much, the system automatically backtracks and checks the corresponding evaluation model and input monitoring data to ensure that the coefficient is reliable. Subsequently, according to the actual operating characteristics of the reactor, the accuracy of the evaluation model and the accumulation of historical data, the system automatically selects the mean calculation method. When the accuracy and reliability of each model are similar, a simple arithmetic mean method is used to add the S coefficients and divide them by S; when some models need to be given higher weights due to better technology or algorithms, the system reasonably allocates weights based on historical data and actual operation feedback, and uses weighted arithmetic mean method for calculation. Finally, the system uses the selected method to calculate the efficiency coefficient of the reactor, and automatically compares it with the recent average value under the same operating conditions to check whether the fluctuation is reasonable. At the same time, it performs reverse verification based on actual operating conditions such as reactive power output and harmonic suppression effect. If any discrepancy is found, the relevant links will be reviewed immediately to ensure that the efficiency coefficient is accurate and reliable.

[0024] Step S300, determine whether the electrical performance evaluation result meets the electrical performance evaluation constraint. Specifically, the system automatically formulates detailed quantitative index rules for the three constraints of the reactor's efficacy, energy efficiency, and operational safety based on national and industry standards, reactor design specifications, and a large amount of actual operation data, such as specifying the accuracy of reactive power compensation under different loads, clarifying the upper limit of unit power loss, and setting the thresholds of temperature and vibration amplitude. Subsequently, a high-precision electrical performance monitoring system is deployed to collect data related to the operation of the reactor in real time: for efficacy indicators, power data is collected and automatically converted into reactive power compensation rate; for energy efficiency indicators, electrical energy parameters are accurately measured and energy efficiency coefficients are calculated; for operational safety indicators, sensor data is collected in real time and organized into comparable values. Finally, the system compares the organized electrical performance evaluation result data with the quantitative index rules one by one according to the constraint type: efficacy checks the reactive compensation rate, energy efficiency compares the energy efficiency coefficient, and operational safety checks the temperature and vibration amplitude. If any indicator is not met, the system records the details; if all indicators are met, the electrical performance is determined to be up to standard and the evaluation process is completed.

[0025] Step S400, if the electrical performance evaluation result does not meet the electrical performance evaluation constraint, the reactor is adjusted and optimized according to the reactor load data and the electrical performance evaluation constraint, and the first optimization strategy for reactor adjustment is generated. Specifically, after the system generates the electrical performance evaluation result, it automatically compares it with the electrical performance evaluation constraint, accurately checks the reactive power compensation and power factor in the reactor efficacy, whether the unit power loss in energy efficiency exceeds the standard, and whether the temperature, vibration amplitude and other indicators in operation safety exceed the threshold, and automatically locates the specific problem that does not meet the constraint. At the same time, the system deeply analyzes the load data of the reactor through the built-in data mining algorithm, explores the load change law, including the time when the peak and valley values ​​appear, the fluctuation frequency and the duration under different load levels, etc., and comprehensively analyzes the impact mechanism of the load on electrical performance. Based on these analyses, the system combines the power system control rules and the electrical equipment optimization goals, and uses the intelligent optimization algorithm to automatically generate the first optimization strategy for reactor adjustment. For example, the switching timing is adjusted to address reactive power issues, operating parameters are optimized to address energy efficiency issues, and temperature is controlled or operating settings are adjusted to address operational safety issues. This generates a comprehensive and targeted regulation strategy and provides reliable guidance for the optimized operation of the reactor.

[0026] In a possible implementation, if the electrical performance evaluation result does not meet the electrical performance evaluation constraint, the reactor is adjusted and optimized according to the reactor load data and the electrical performance evaluation constraint to generate a first optimization strategy for reactor adjustment. Step S400 further includes step S410, making an adjustment decision on the reactor according to the reactor load data to obtain a first space for reactor adjustment decision. Specifically, the system deploys high-precision power monitoring equipment to automatically collect data of the power system where the reactor is located, focusing on monitoring the real-time current, voltage, power and other parameters of the load and their changing trends, and drawing fluctuation curves, recording the peak and valley values ​​of the system, and automatically identifying and classifying load types. Through the built-in data cleaning algorithm, the system processes the collected data to ensure the accuracy and completeness of the data. Subsequently, the system uses big data analysis technology and built-in professional models to automatically analyze the dynamic change characteristics of the load, identify the operating characteristics of different loads such as industrial motors and residential electricity, and divide the load conditions into stable high load, frequent fluctuations, low load standby and other categories. Based on the operating condition analysis, the system generates a preliminary adjustment plan, such as increasing the number of groups put into operation when the load is high, using intelligent switching when the load fluctuates, and lowering the voltage or reducing the number of groups when the load is low. This builds the first space for the reactor adjustment decision, providing a basis for subsequent optimization.

[0027] Step S420, based on the three channels of electrical performance evaluation, the first space of the reactor adjustment decision is optimized and analyzed according to the constraints of the electrical performance evaluation, and the second space of the reactor adjustment decision is established. Specifically, based on a large amount of experimental data, actual operation cases and authoritative standards, the system automatically sorts out the connotation of the three channels of electrical performance evaluation, clarifies the key indicators and constraints of each channel of reactor efficacy, energy efficiency, and operation safety, such as the reactive power compensation accuracy in the efficacy channel, the upper limit of unit power loss in the energy efficiency channel, and the winding temperature threshold in the operation safety channel. Subsequently, the system imports all the adjustment schemes of the first space of the reactor adjustment decision into the electrical performance evaluation platform of integrated simulation technology in a standard format. According to the three-channel indicators and constraints, the platform simulates and runs each scheme one by one, and automatically outputs preliminary performance evaluation results such as reactive power compensation deviation and unit power loss value. Finally, the system uses the built-in multi-objective optimization algorithm to conduct in-depth optimization based on the preliminary evaluation results, automatically weigh the pros and cons of each solution in terms of efficacy, energy efficiency, and operational safety, screen out solutions with better performance, and integrate them to generate the second space for reactor adjustment decisions, providing a reliable basis for further optimizing reactor operation.

[0028] Step S430, introduce the reactor adjustment fitness evaluation network, and optimize the adjustment fitness maximization of the second space of the reactor adjustment decision according to the reactor adjustment fitness evaluation network, and output the first optimization strategy for the reactor adjustment, wherein the reactor adjustment fitness evaluation network includes the reactor efficacy weight, the reactor energy efficiency weight and the reactor operation safety weight. Specifically, the system automatically analyzes the actual operating environment, application scenarios and user needs of the reactor in an all-round manner, combines historical operating data, and adopts hierarchical analysis method and fuzzy comprehensive evaluation method to assign reasonable weights to the reactor efficacy, energy efficiency and operation safety, and automatically construct the reactor adjustment fitness evaluation network, for example, setting the efficacy weight to 0.5, energy efficiency to 0.3, and operation safety to 0.2 for industrial intensive areas. Subsequently, the system imports the regulation schemes in the second space of the reactor regulation decision one by one according to the interface specifications preset by the network. The network uses machine learning algorithms and built-in mathematical models to calculate the fitness scores of each scheme according to the weights, simulates the operation to generate reactive power compensation, power loss and operation safety index data, and converts them into comprehensive fitness scores. Finally, the system starts the optimization algorithm, and through intelligent optimization methods such as genetic algorithms or particle swarm optimization algorithms, it iteratively searches with the goal of maximizing the regulation fitness, dynamically adjusts the scheme, and after multiple rounds of optimization, outputs the first optimization strategy for the reactor regulation corresponding to the maximum fitness, providing precise guidance for the regulation operation in actual operation.

[0029] In a possible implementation, the reactor is adjusted according to the reactor load data to obtain the first reactor adjustment decision space, and step S410 further includes step S411, the control scheme of the reactor is retrieved according to the reactor load data to obtain a retrieved control scheme set. Specifically, the system deploys high-precision, multi-parameter power monitoring equipment at the key nodes of the power system where the reactor is located, and accurately collects reactor load data such as current, voltage, power factor, active power, reactive power and its dynamic change curve in real time, and transmits the data to the data processing center. The system cleans the collected data through the built-in advanced data processing algorithm, automatically removes abnormal values ​​and noise, and completes the missing data to ensure that the data is complete and accurate, laying the foundation for subsequent retrieval. Subsequently, the system combines the electrical characteristics of the reactor, common working conditions and industry standards and specifications to build an intelligent retrieval system. The built-in semantic parsing module is used to parse the control requirements described in natural language, and locates the relevant control schemes according to the load data characteristics through machine learning classification and matching algorithms, and identifies the retrieval conditions under peak hours or low loads. The system inputs the sorted load data into the intelligent retrieval module, searches the massive database based on the preset rule algorithm, comprehensively analyzes the dynamic characteristics of the load such as real-time value, change trend and fluctuation frequency, and selects the control scheme that is suitable for the operating conditions. For the situation of periodic large fluctuations, the system gives priority to recommending adaptive adjustment schemes, and finally forms a set of retrieval control schemes.

[0030] Step S412, data cleaning is performed according to the retrieved control scheme set to obtain the reactor control space. Specifically, after generating the retrieved control scheme set, the system automatically conducts a comprehensive review of each scheme in the scheme set based on the actual operating parameters such as the rated current, voltage level, and load variation range of the reactor, and marks the schemes that are suitable for high voltage and large capacity but not suitable for the current medium and low voltage and small capacity reactors. Subsequently, the system cleans and optimizes the scheme set through the built-in data cleaning module, including correcting the numerical deviation of the voltage regulation range and the confusion of control parameter units caused by input errors; automatically verifies the accuracy of the data by comparing authoritative information with similar schemes; and estimates the data for the missing reactive power compensation strategy under low load conditions by combining the characteristics of the reactor, similar cases, and built-in theoretical models; at the same time, eliminates redundant and repeated information to ensure the simplicity and efficiency of the scheme set. Ultimately, the system retains the optimized and reliable control scheme after cleaning and builds a reactor control space, covering strategies such as precise switching under high load conditions to ensure reactive power balance and reducing operating losses under low load conditions to improve energy efficiency, laying a solid foundation for subsequent analysis of trigger conditions and making adjustment decisions.

[0031] Step S413, analyze the trigger range of each dimension according to the reactor control space, and establish the reactor control trigger domain. Specifically, the system focuses on the scheme in the reactor control space and the key dimensions it covers, such as reactive power compensation, voltage regulation, and current control, and automatically analyzes the operating rules and trigger conditions of each dimension. For the reactive power compensation dimension, the system uses a large amount of historical load data and reactor operation data, combined with the power system analysis model, to analyze the changes in reactive power demand under different load conditions, and determines the trigger range of switching to a large-capacity reactor when the load is light and the power factor is lower than 0.8, and switching to a large-capacity reactor when the load is heavy and the power factor is lower than 0.6. In the voltage regulation dimension, the system uses circuit analysis and simulation models according to the grid voltage standard and actual voltage fluctuation data to set the trigger conditions: when the system voltage is continuously higher than the rated voltage by 5% for a certain period of time, the step-down regulation is started, and when it is lower than the rated voltage by 3% and lasts for too long, the step-up regulation is started. In the current control dimension, the system combines the rated current of the reactor with the real-time load current data to set the trigger conditions: when the load current approaches or exceeds 120% of the rated current, the current limiting protection is triggered, and when it is lower than 50% of the rated current, the current control is optimized to improve energy efficiency. Finally, the system integrates the trigger ranges of each dimension and establishes the reactor control trigger domain, providing a scientific basis for the precise regulation of reactor operation.

[0032] Step S414, making an adjustment decision for the reactor according to the reactor control trigger domain, and generating the first space of the reactor adjustment decision. Specifically, the system automatically analyzes the key trigger conditions such as the reactive power compensation trigger point, the voltage adjustment threshold and the current control limit in the reactor control trigger domain. According to the reactive power demand node clearly defined in the trigger domain, the system monitors the actual load situation in real time. For example, when the load is medium and the power factor is close to the critical value of the small-capacity reactor, the system automatically formulates the investment decision, and comprehensively considers the system stability, subsequent load changes, and reactor performance parameters, estimates the reactive power improvement effect, the degree of power factor optimization and potential problems, and generates a response plan in advance. When the system voltage approaches the adjustment threshold, the system weighs according to the overall operating situation: when the voltage is slightly higher and the fluctuation is small, the reactance value is fine-tuned to steadily reduce the voltage; when the deviation is large and the duration is long, the reactor is decisively switched on to pull back the voltage, while taking into account reactive power distribution and subsequent adjustment requirements. In terms of current control, the system monitors the load current and current control limit in real time. When approaching the overload limit, it makes current limiting decisions based on the rated parameters and synergy characteristics of the reactor, automatically adjusts the connection method or optimizes the circuit parameters to ensure current compliance and maintain reactive power supply, while optimizing power loss. Finally, the system summarizes these fine-tuning strategies to generate the first space for reactor adjustment decisions, providing a basis for precise operation control.

[0033] In a possible implementation, based on the three channels of electrical performance evaluation, the first space of the reactor adjustment decision is optimized and analyzed according to the electrical performance evaluation constraints, and the second space of the reactor adjustment decision is established. Step S420 further includes step S421, according to the first space of the reactor adjustment decision, extracting the kth decision of the reactor adjustment, wherein k is a positive integer. Specifically, when extracting the kth decision of the reactor adjustment, the system automatically analyzes the adjustment decisions and their generation basis formulated in the first space for working conditions such as reactive power shortage, voltage fluctuation, and load change, and combines the real-time operation status to evaluate the reactive power demand, voltage stability and load change trend, and selects the decision that preferentially meets the current demand. Through the built-in evaluation system, weights are assigned according to dimensions such as reactive power, voltage regulation and load adaptability, and the overall value of all adjustment decisions is comprehensively evaluated. The intelligent algorithm is used to score the decisions one by one according to the weights, and the decisions that are balanced in all aspects and adapted to the working conditions are given higher scores. Finally, the kth decision is accurately extracted, the switching and voltage adjustment parameters of the reactor are refined, and reliable guidance is provided for the operation of the power system in specific scenarios.

[0034] Step S422, predict the state of the reactor according to the kth decision of the reactor adjustment, and obtain the kth state prediction result of the reactor. Specifically, when the system predicts the state of the reactor according to the kth decision of the reactor adjustment, it automatically analyzes the details of the kth decision, extracts key information such as the number of switching groups and voltage regulation parameters, and combines it with the core goals of responding to reactive power fluctuations, voltage stability and load adaptability requirements. The system builds a high-fidelity simulation model based on the electrical principles and historical operation data of the reactor, uses precise mathematical formulas to describe physical phenomena such as electromagnetic characteristics, heat conduction, and mechanical vibration, and integrates the operation data under different working conditions to determine key parameters, such as the influence of load change rate on temperature rise curve and vibration amplitude. Subsequently, the parameter value of the kth decision is input into the simulation model as the initial condition, and the simulation program is started to simulate the future operation process, and the key state indicators such as reactive power output, winding temperature change, core magnetic flux fluctuation, vibration amplitude and frequency are calculated in real time. The system processes the simulation data through trend extrapolation and time series analysis techniques. For example, it estimates the extreme value of a longer period based on the temperature rise trend in a short period of time, and analyzes the periodic changes in the vibration amplitude to determine the possible resonance risk. Finally, the system generates a detailed report on the prediction results of the reactor's kth state, clearly listing the reactive power output range, expected maximum winding temperature, possible maximum vibration amplitude and corresponding time nodes in different periods of time in the future, providing a scientific basis for formulating measures to ensure the stable operation of the reactor in advance.

[0035] Step S423, input the k-th state prediction result of the reactor into the three channels of electrical performance evaluation to obtain the k-th decision electrical performance evaluation result. Specifically, when the system inputs the k-th state prediction result of the reactor into the three channels of electrical performance evaluation to obtain the k-th decision electrical performance evaluation result, it first automatically analyzes the architecture and function of the three channels of electrical performance evaluation, focuses on the three key aspects of reactive power compensation accuracy, voltage regulation stability and power loss, and clarifies the evaluation points, indicator calculation methods and data processing procedures of each channel. Subsequently, the system preprocesses the k-th state prediction result, screens the data directly related to the electrical performance, and formats and organizes it according to the requirements of each channel, such as extracting the reactive power prediction value to generate time series data to adapt to the corresponding channel. Then, the system inputs the preprocessed data into the three channels respectively: in the reactive power compensation accuracy channel, compares the predicted value with the ideal target value to calculate the deviation rate; in the voltage regulation stability channel, judges whether it meets the standard according to the voltage fluctuation value combined with the preset standard; in the power loss channel, evaluates whether it exceeds the reasonable range according to the active power loss prediction curve. Finally, the system integrates the preliminary evaluation results of each channel, assigns different weights according to the importance of their impact on the overall electrical performance of the reactor, and calculates the comprehensive electrical performance evaluation result of the kth decision through weighted summation, providing a scientific basis for optimizing the regulation strategy and ensuring the efficient operation of the power system.

[0036] Step S424, determine whether the electrical performance evaluation result of the kth decision meets the electrical performance evaluation constraint. Specifically, when the system determines whether the electrical performance evaluation result of the kth decision meets the electrical performance evaluation constraint, it automatically parses the electrical performance evaluation constraint details. These details are based on the power industry specifications, the experience threshold of the reactor operation and the current power system requirements, covering key indicators such as the reactive power compensation accuracy deviation range, the voltage regulation fluctuation amplitude and recovery time limit, and the maximum value of the electric energy loss. The system checks the quantitative data such as reactive power, voltage and electric energy loss in the electrical performance evaluation result of the kth decision one by one, compares the reactive power compensation deviation rate with the allowable range, analyzes whether the voltage regulation fluctuation amplitude and recovery time are compliant, and calculates whether the measured value of the electric energy loss exceeds the limit. Through rigorous logical comprehensive judgment, only when all indicators meet the constraint conditions, the overall satisfaction is determined; if any indicator does not meet the requirements (such as the reactive power deviation rate exceeds the limit), even if other aspects are good, it is determined to be unsatisfactory. Based on the judgment results, the system automatically provides feedback: if the constraints are met, it is marked as a feasible solution and enters subsequent operation and control; if not, an alarm is triggered, the cause is analyzed and the decision is adjusted in a targeted manner to ensure the safety and stability of the power system operation.

[0037] Step S425, if the electrical performance evaluation result of the kth decision meets the electrical performance evaluation constraint, the kth decision of the reactor adjustment is added to the second space of the reactor adjustment decision. Specifically, when the electrical performance evaluation result of the kth decision meets the electrical performance evaluation constraint and the kth decision of the reactor adjustment needs to be added to the second space of the reactor adjustment decision, the system first automatically verifies whether the electrical performance evaluation result of the kth decision does meet all the constraints, and reviews whether indicators such as reactive power compensation accuracy, voltage regulation stability, and electric energy loss values ​​meet the standards to ensure accuracy and avoid power system problems caused by incorrect additions. Next, the system checks the storage architecture, storage space, and data index of the second space of the reactor adjustment decision to ensure that it has a good storage state and can accept new data. Subsequently, the system converts the key parameters of the kth decision of the reactor adjustment (such as reactor switching scheme, voltage adjustment value, etc.) into a unified format according to the preset format specification. Finally, the system accurately adds the processed kth decision to the second space through the database entry module, and automatically verifies the integrity and accuracy of the data after the addition is completed to ensure that the decision can be used for subsequent power system control.

[0038] Step S426, if the electrical performance evaluation result of the kth decision does not meet the electrical performance evaluation constraint, the kth decision of the reactor adjustment is eliminated. Specifically, when the system determines that the electrical performance evaluation result of the kth decision does not meet the electrical performance evaluation constraint, the system first comprehensively analyzes the specific reasons for not meeting the constraint, deeply analyzes the root causes of the reactive power compensation accuracy not meeting the standard, the voltage regulation stability is insufficient, or the power loss exceeds the standard, and accumulates data and experience for subsequent optimization decisions. Then, the system automatically locates the storage location of the decision in the entire decision-making process and accurately locks its location. Subsequently, the system uses built-in tools to carefully remove the decision from the storage location based on the preset process and data management specifications, and follows data security rules to prevent accidental deletion of other data or affect other records. Finally, the system automatically records the entire elimination process, including key information such as reasons, time nodes, and execution operations, and generates detailed archives to facilitate retrospective analysis, provide reliable references for the formulation of new decisions, and ensure efficient and safe operation of the power system.

[0039] Step S500, if the electrical performance evaluation result meets the electrical performance evaluation constraint, a trend prediction is performed according to the reactor load data to obtain a load trend prediction result, and the reactor is adjusted and optimized in combination with the electrical performance evaluation constraint to generate a second optimization strategy for reactor adjustment. Specifically, when the electrical performance evaluation result meets the electrical performance evaluation constraint, the system automatically cleans and preprocesses the reactor load data, uses the ARIMA model or LSTM algorithm to remove outliers, and restores the true load change trend. Subsequently, based on the processed data and the power system operation rules and historical data, the system selects an adaptive prediction model to predict the load trend: for simple systems, linear regression combined with the moving average method is used, and for complex systems, a neural network model is used to accurately predict the load trend in the next few hours or days. Based on the prediction, the system combines the electrical performance evaluation constraint and uses a genetic algorithm or a particle swarm optimization algorithm to perform multiple rounds of optimization on the reactor adjustment parameters to ensure that the requirements of reactive power compensation accuracy, voltage regulation stability, and power loss upper limit are met. During the optimization process, the system tentatively adjusts the number and time of reactive power compensation switching groups, optimizes key operations such as voltage regulator parameters, and finally generates an accurate second optimization strategy for reactor regulation, clarifies the regulation operations in each time period under different load trends, and provides a reliable basis for the efficient and stable operation of the power system.

[0040] In a possible implementation, if the electrical performance evaluation result satisfies the electrical performance evaluation constraint, a trend prediction is performed according to the reactor load data to obtain a load trend prediction result, and the reactor is adjusted and optimized in combination with the electrical performance evaluation constraint to generate a second optimization strategy for reactor adjustment, and step S500 further includes step S510, performing a change degree evaluation according to the load trend prediction result to obtain a load change degree. Specifically, when the change degree evaluation is performed according to the load trend prediction result to obtain the load change degree, the system automatically analyzes the data details of the load trend prediction result in depth, including the reactor load size, change rate and fluctuation cycle in different time periods in the future. The system uses a built-in data cleaning algorithm to automatically remove abnormal values ​​caused by sensor failure or electromagnetic interference to ensure that the data is true and reliable. Subsequently, the system selects an adaptive measurement indicator according to the operating characteristics of the power system and the law of load change. For example, when the power system runs smoothly and fluctuates slowly, the standard deviation is used to measure the dispersion of the load deviation from the mean; when the system is complex and the nonlinear changes are significant, the slope change rate is used to capture the speed of the increase and decrease trend. The system calculates through formulas, such as substituting the standard deviation formula to calculate the load fluctuation amplitude in each period, or calculating the rate of change of the load difference and time interval between adjacent periods through the slope change rate. Finally, the system organizes and normalizes the calculation results to obtain the load change degree, providing a scientific basis for subsequent reactor adjustment decisions.

[0041] Step S520, determine whether the load variation is greater than or equal to the predetermined load variation. Specifically, when determining whether the load variation is greater than or equal to the predetermined load variation, the system automatically analyzes the setting basis of the predetermined load variation, combines the historical operation data of the power system, power consumption patterns, seasonal characteristics and system fluctuation tolerance parameters, and comprehensively analyzes the sudden change in summer peak load and the carrying capacity after the system transformation to ensure that the predetermined value is scientific and reasonable. Subsequently, the system accurately obtains the load variation value, and ensures that the data is true and reliable and accurately reflects the load fluctuation by checking the data acquisition source, sensor accuracy, acquisition frequency and algorithm processing flow. The system makes a judgment based on the preset logic: if the load variation is greater than or equal to the predetermined value, it indicates that the load fluctuation is severe, and the system will generate adjustment suggestions, such as optimizing the switching of the reactor or voltage regulation; if it is less than the predetermined value, it indicates that the load fluctuation is gentle, and the system records the current situation for continuous monitoring. Finally, the system forms corresponding decision suggestions based on the judgment results. If adjustment is required, a detailed adjustment plan is automatically generated and submitted to the control department. If no adjustment is required, the data is recorded for subsequent analysis and traceability.

[0042] Step S530, if the load variation is greater than or equal to the predetermined load variation, the reactor is adjusted according to the load trend prediction result to obtain the third space of reactor adjustment decision. Specifically, when the reactor is adaptively load adjusted according to the first optimization strategy and the second optimization strategy for reactor adjustment, the system automatically analyzes the core objectives of the two optimization strategies. The first optimization strategy focuses on conventional load fluctuations, formulates a periodic reactor switching plan based on the historical load curve, and maintains reactive power balance; the second optimization strategy targets complex extreme working conditions, handles emergencies such as centralized equipment startup, dynamically optimizes voltage regulation parameters, and quickly adjusts the number of switching groups. The system receives data such as load size, change rate, power factor, etc. collected by high-precision sensors at each node of the power grid in real time, and judges the matching of the current load state and strategy through fast data transmission and processing technology. If the load is stable, the adjustment is carried out in an orderly manner according to the first strategy; if the load changes drastically, the second strategy is dominant, and the emergency plan is activated, such as quickly increasing or decreasing the number of switching groups and finely adjusting the voltage regulation tap position. During the execution process, the system converts the strategy into automated instructions and inputs them into the intelligent control system. It uses precise algorithms to ensure that the adjustment action is correct, and monitors the adjustment effect in real time. If deviations are found, the system will immediately make backtracking adjustments to ensure the stable operation of the power system.

[0043] Step S540, based on the three channels of electrical performance evaluation, the third space of the reactor adjustment decision is optimized and analyzed according to the electrical performance evaluation constraints, and the fourth space of the reactor adjustment decision is established. Specifically, when the third space of the reactor adjustment decision is optimized and analyzed according to the electrical performance evaluation constraints based on the three channels of electrical performance evaluation to establish the fourth space, the system first analyzes the functions and evaluation indicators of the three channels, clarifies the indicator calculation methods, data sources and weight distribution in the three major areas of reactive power compensation accuracy, voltage regulation stability and electric energy loss, and analyzes the constraints of electrical performance evaluation at the same time, clarifies the specific constraints such as the minimum accuracy of reactive power compensation, voltage fluctuation limit and recovery time, and electric energy loss upper limit, so as to provide boundaries for optimization. Subsequently, the system sequentially inputs the adjustment decision schemes of the third space into the three channels for analysis: in the reactive power compensation accuracy channel, calculates the deviation rate and sorts it; in the voltage regulation stability channel, determines whether the standard is met and marks the non-compliant schemes; in the electric energy loss channel, compares the loss upper limit to filter out the schemes that exceed the standard. Next, the system conducts a comprehensive weighted evaluation of the solutions based on the weight distribution. The reactive power compensation accuracy and voltage regulation stability have higher weights due to their greater impact, while the power loss has a lower weight. The comprehensive score is calculated by weighted summation, and the high-quality solutions with higher scores are selected to form the fourth space. Finally, the fourth space generated by the system provides a reliable basis for the precise control and efficient operation of the power system, ensuring that the performance of the reactor reaches the optimal state.

[0044] Step S550, according to the reactor adjustment fitness evaluation network, the fourth space of the reactor adjustment decision is optimized to maximize the adjustment fitness, and the second optimization strategy of the reactor adjustment is output. Specifically, when the fourth space is optimized to maximize the adjustment fitness according to the reactor adjustment fitness evaluation network to output the second optimization strategy of the reactor adjustment, the system first analyzes the network architecture and operation logic, and calculates the fitness score through multi-dimensional information such as reactive power compensation timeliness, voltage regulation smoothness and rationality of power loss. Subsequently, the system accurately inputs each scheme of the fourth space according to the network requirement format, and sorts out key data such as switching time series and voltage regulation parameters. The network calculates the adjustment fitness score of each scheme through a built-in algorithm, evaluates the advantages and disadvantages of the scheme in terms of system stability, load matching and operating efficiency, and sorts the scores to select high-quality schemes. The system comprehensively considers the real-time operating status, recent operating condition changes and potential risks, and gives priority to schemes with strong emergency adaptability. After multiple rounds of optimization, the second optimization strategy of the reactor adjustment is finally generated, and the precise adjustment method under different working conditions is clarified, providing a scientific basis for the stable and efficient operation of the power system.

[0045] Step S600, adaptive load regulation is performed on the reactor according to the first optimization strategy for reactor regulation and the second optimization strategy for reactor regulation. Specifically, when adaptive load regulation is performed on the reactor according to the first optimization strategy and the second optimization strategy for reactor regulation, the system first analyzes the core content and applicable scope of the two strategies. The first optimization strategy responds to conventional load fluctuations, formulates a time-based switching plan based on the historical load curve, and maintains reactive power balance; the second optimization strategy dynamically optimizes voltage regulation parameters and quickly adjusts the number of switching groups for complex extreme working conditions, such as sudden load changes or centralized equipment startup. The system collects data such as current, voltage, power factor and load change rate of the grid node in real time, and quickly analyzes and matches the strategy. When the load is stable, the operating state of the reactor is gradually adjusted according to the first strategy; when the load changes drastically, the second strategy is used to respond quickly, and fluctuations are responded to by increasing or decreasing the number of switching groups and optimizing the voltage regulation tap position. During the execution process, the system converts the strategy into an automated instruction, operates accurately through an intelligent control system, and monitors the regulation effect in real time. If deviations are found, the system automatically performs back-analysis and adjusts strategies to ensure that reactive power compensation, voltage stability and load adaptability meet standards, thereby achieving efficient and stable operation of the power system.

[0046] In the above, refer to Figure 1 The reactor load regulation method based on electrical performance optimization according to an embodiment of the present invention is described in detail. Figure 2 A reactor load regulation system based on electrical performance optimization according to an embodiment of the present invention is described.

[0047] The reactor load regulation system based on electrical performance optimization according to the embodiment of the present invention is used to solve the technical problem that the regulation of the reactor in the prior art cannot adapt to complex load changes, resulting in low reactor regulation accuracy and poor electrical performance, and achieves the technical effect of improving the reactor regulation accuracy and optimizing electrical performance. The reactor load regulation system based on electrical performance optimization includes: a real-time monitoring module 10, an evaluation result acquisition module 20, an evaluation result judgment module 30, an adjustment optimization module 40, a load trend prediction result acquisition module 50, and an adaptive load regulation module 60.

[0048] The real-time monitoring module 10 is used to perform real-time monitoring on the reactor and obtain reactor monitoring data and reactor load data.

[0049] The evaluation result acquisition module 20 is used to input the reactor monitoring data into the three channels of electrical performance evaluation to obtain electrical performance evaluation results.

[0050] The evaluation result judgment module 30 is used to judge whether the electrical performance evaluation result meets the electrical performance evaluation constraint.

[0051] The adjustment optimization module 40 is used to adjust and optimize the reactor according to the reactor load data and the electrical performance evaluation constraint if the electrical performance evaluation result does not meet the electrical performance evaluation constraint, and generate a first optimization strategy for reactor adjustment.

[0052] The load trend prediction result acquisition module 50 is used to perform trend prediction based on the reactor load data if the electrical performance evaluation result meets the electrical performance evaluation constraint, obtain the load trend prediction result, and adjust the reactor in combination with the electrical performance evaluation constraint to generate a second optimization strategy for reactor adjustment.

[0053] The adaptive load regulation module 60 is used to perform adaptive load regulation on the reactor according to the first optimization strategy for reactor regulation and the second optimization strategy for reactor regulation.

[0054] The specific configuration of the evaluation result acquisition module 20 will be described in detail below. As described above, the reactor monitoring data is input into the three channels of electrical performance evaluation to obtain the electrical performance evaluation results. The evaluation result acquisition module 20 further includes: an evaluation three-channel composition unit, the evaluation three-channel composition unit is used for the three channels of electrical performance evaluation including the reactor efficacy evaluation channel, the reactor energy efficiency evaluation channel and the reactor operation safety evaluation channel; a reactor efficacy coefficient acquisition unit, the reactor efficacy coefficient acquisition unit is used to input the reactor monitoring data into the reactor efficacy evaluation channel to obtain the reactor efficacy coefficient; a reactor energy efficiency coefficient acquisition unit The reactor energy efficiency coefficient acquisition unit is used to input the reactor monitoring data into the reactor energy efficiency evaluation channel to obtain the reactor energy efficiency coefficient; the reactor operation safety factor output unit is used to input the reactor monitoring data into the reactor operation safety evaluation channel to output the reactor operation safety factor; the electrical performance evaluation result output unit is used to output the reactor effectiveness coefficient, the reactor energy efficiency coefficient and the reactor operation safety factor as the electrical performance evaluation result.

[0055] Among them, the reactor monitoring data is input into the reactor efficacy evaluation channel to obtain the reactor efficacy coefficient, and the reactor efficacy coefficient acquisition unit further includes: a reactor efficacy evaluation channel composition subunit, the reactor efficacy evaluation channel composition subunit is used for the reactor efficacy evaluation channel to include S reactor efficacy evaluation models, wherein S is a positive integer greater than 1; an efficacy evaluation coefficient acquisition subunit, the efficacy evaluation coefficient acquisition subunit is used to input the reactor monitoring data into the S reactor efficacy evaluation models to obtain S reactor efficacy evaluation coefficients; a mean calculation subunit, the mean calculation subunit is used to perform mean calculation according to the S reactor efficacy evaluation coefficients to generate the reactor efficacy coefficient.

[0056] Next, the specific configuration of the adjustment optimization module 40 will be described in detail. As described above, if the electrical performance evaluation result does not meet the electrical performance evaluation constraint, the reactor is adjusted and optimized according to the reactor load data and the electrical performance evaluation constraint to generate a first optimization strategy for reactor adjustment. The adjustment optimization module 40 further includes: an adjustment decision unit, the adjustment decision unit is used to make an adjustment decision on the reactor according to the reactor load data to obtain a first space for reactor adjustment decision; an optimization analysis unit, the optimization analysis unit is used to perform optimization analysis on the first space for reactor adjustment decision based on the electrical performance evaluation three channels and according to the electrical performance evaluation constraint to establish a second space for reactor adjustment decision; a first optimization strategy output unit, the first optimization strategy output unit is used to introduce a reactor adjustment fitness evaluation network, and perform adjustment fitness maximization optimization on the second space for reactor adjustment decision according to the reactor adjustment fitness evaluation network, and output the first optimization strategy for reactor adjustment, wherein the reactor adjustment fitness evaluation network includes a reactor efficacy weight, a reactor energy efficiency weight and a reactor operation safety weight.

[0057] Among them, an adjustment decision is made on the reactor according to the reactor load data to obtain a first space for the reactor adjustment decision, and the adjustment decision unit further includes: a control scheme retrieval subunit, the control scheme retrieval subunit is used to retrieve a control scheme for the reactor according to the reactor load data to obtain a retrieved control scheme set; a data cleaning subunit, the data cleaning subunit is used to perform data cleaning according to the retrieved control scheme set to obtain a reactor control space; a trigger range parsing subunit, the trigger range parsing subunit is used to perform trigger range parsing in each dimension according to the reactor control space to establish a reactor control trigger domain; an adjustment decision subunit, the adjustment decision subunit is used to make an adjustment decision on the reactor according to the reactor control trigger domain to generate the first space for the reactor adjustment decision.

[0058] Among them, based on the three channels of electrical performance evaluation, the first space of the reactor adjustment decision is optimized and analyzed according to the electrical performance evaluation constraints, and the second space of the reactor adjustment decision is established. The optimization analysis unit further includes: a decision extraction subunit, the decision extraction subunit is used to extract the kth decision of the reactor adjustment according to the first space of the reactor adjustment decision, wherein k is a positive integer; a state prediction subunit, the state prediction subunit is used to predict the state of the reactor according to the kth decision of the reactor adjustment, and obtain the kth state prediction result of the reactor; a decision electrical performance evaluation result acquisition subunit, the decision electrical performance evaluation result acquisition subunit is used to predict the kth state of the reactor The result is input into the three channels of electrical performance evaluation to obtain the electrical performance evaluation result of the kth decision; an electrical performance evaluation constraint judgment subunit is used to judge whether the electrical performance evaluation result of the kth decision satisfies the electrical performance evaluation constraint; a decision second space adding subunit is used to add the kth decision of reactor adjustment to the second space of the reactor adjustment decision if the electrical performance evaluation result of the kth decision satisfies the electrical performance evaluation constraint; a decision elimination subunit is used to eliminate the kth decision of reactor adjustment if the electrical performance evaluation result of the kth decision does not satisfy the electrical performance evaluation constraint.

[0059] The specific configuration of the load trend prediction result acquisition module 50 will be described in detail below. As described above, if the electrical performance evaluation result satisfies the electrical performance evaluation constraint, a trend prediction is performed based on the reactor load data to obtain a load trend prediction result, and the reactor is adjusted and optimized in combination with the electrical performance evaluation constraint to generate a second optimization strategy for reactor adjustment. The load trend prediction result acquisition module 50 further includes: a variation evaluation unit, the variation evaluation unit is used to perform a variation evaluation based on the load trend prediction result to obtain a load variation; a predetermined load variation judgment unit, the predetermined load variation judgment unit is used to judge whether the load variation is greater than or equal to a predetermined load variation; a decision third space acquisition unit, the decision third space The acquisition unit is used for making an adjustment decision on the reactor according to the load trend prediction result if the load variation degree is greater than or equal to the predetermined load variation degree, so as to obtain a third space for the reactor adjustment decision; a fourth space construction unit for making a decision, the fourth space construction unit for making an optimization analysis on the third space for the reactor adjustment decision based on the electrical performance evaluation constraints based on the three channels for electrical performance evaluation, and establishing a fourth space for the reactor adjustment decision; a second optimization strategy output unit, the second optimization strategy output unit for maximizing the adjustment fitness of the fourth space for the reactor adjustment decision according to the reactor adjustment fitness evaluation network, and outputting the second optimization strategy for the reactor adjustment.

[0060] The reactor load regulation system based on electrical performance optimization provided in the embodiment of the present invention can execute the reactor load regulation method based on electrical performance optimization provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0061] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0062] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art 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 in the protection scope of this application.

Claims

1. A reactor load regulation method based on electrical performance optimization, characterized in that: The method comprises: Monitor the reactor in real time to obtain the reactor monitoring data and reactor load data; Inputting the reactor monitoring data into the three channels of electrical performance evaluation to obtain electrical performance evaluation results; Determining whether the electrical performance evaluation result satisfies the electrical performance evaluation constraint; If the electrical performance evaluation result does not satisfy the electrical performance evaluation constraint, the reactor is optimized according to the reactor load data and the electrical performance evaluation constraint to generate a first optimization strategy for reactor adjustment; If the electrical performance evaluation result satisfies the electrical performance evaluation constraint, a trend prediction is performed according to the reactor load data to obtain a load trend prediction result, and the reactor is adjusted and optimized in combination with the electrical performance evaluation constraint to generate a second optimization strategy for reactor adjustment; The reactor is adaptively load regulated according to the first reactor regulation optimization strategy and the second reactor regulation optimization strategy.

2. The method according to claim 1, characterized in that The reactor monitoring data is input into the three channels of electrical performance evaluation to obtain electrical performance evaluation results, including: The three electrical performance evaluation channels include a reactor efficacy evaluation channel, a reactor energy efficiency evaluation channel, and a reactor operation safety evaluation channel; Inputting the reactor monitoring data into the reactor efficacy evaluation channel to obtain a reactor efficacy coefficient; Inputting the reactor monitoring data into the reactor energy efficiency evaluation channel to obtain the reactor energy efficiency coefficient; Inputting the reactor monitoring data into the reactor operation safety evaluation channel, and outputting the reactor operation safety factor; The reactor effectiveness coefficient, the reactor energy efficiency coefficient and the reactor operation safety factor are output as the electrical performance evaluation result.

3. The method according to claim 2, characterized in that Inputting the reactor monitoring data into the reactor efficacy evaluation channel to obtain the reactor efficacy coefficient includes: The reactor efficacy evaluation channel includes S reactor efficacy evaluation models, where S is a positive integer greater than 1; Inputting the reactor monitoring data into the S reactor efficacy evaluation models to obtain S reactor efficacy evaluation coefficients; The reactor efficiency coefficient is generated by performing mean calculation based on the S reactor efficiency evaluation coefficients.

4. The method according to claim 1, characterized in that If the electrical performance evaluation result does not meet the electrical performance evaluation constraint, the reactor is adjusted and optimized according to the reactor load data and the electrical performance evaluation constraint to generate a first optimization strategy for reactor adjustment, including: Making an adjustment decision on the reactor according to the reactor load data to obtain a first adjustment decision space for the reactor; Based on the three electrical performance evaluation channels, optimizing the first reactor adjustment decision space according to the electrical performance evaluation constraints to establish a second reactor adjustment decision space; A reactor adjustment fitness evaluation network is introduced, and the adjustment fitness maximization optimization is performed on the second space of the reactor adjustment decision according to the reactor adjustment fitness evaluation network, and the first optimization strategy for the reactor adjustment is output, wherein the reactor adjustment fitness evaluation network includes the reactor efficacy weight, the reactor energy efficiency weight and the reactor operation safety weight.

5. The method according to claim 4, characterized in that Making an adjustment decision on the reactor according to the reactor load data to obtain a first adjustment decision space for the reactor includes: Performing a control scheme search for the reactor according to the reactor load data to obtain a search control scheme set; Perform data cleaning according to the retrieved control scheme set to obtain a reactor control space; Analyze the trigger range of each dimension according to the reactor control space to establish a reactor control trigger domain; An adjustment decision is made for the reactor according to the reactor control trigger domain to generate a first adjustment decision space for the reactor.

6. The method according to claim 4, characterized in that Based on the three electrical performance evaluation channels, the first reactor adjustment decision space is optimized and analyzed according to the electrical performance evaluation constraints to establish a second reactor adjustment decision space, including: Extracting a kth reactor adjustment decision according to the first reactor adjustment decision space, wherein k is a positive integer; Performing state prediction on the reactor according to the kth decision of reactor adjustment to obtain a kth state prediction result of the reactor; Inputting the k-th state prediction result of the reactor into the three electrical performance evaluation channels to obtain the k-th decision electrical performance evaluation result; Determining whether the k-th decision electrical performance evaluation result satisfies the electrical performance evaluation constraint; If the electrical performance evaluation result of the k-th decision satisfies the electrical performance evaluation constraint, adding the k-th decision for reactor adjustment to the second space of reactor adjustment decisions; If the electrical performance evaluation result of the kth decision does not meet the electrical performance evaluation constraint, the kth decision for adjusting the reactor is eliminated.

7. The method according to claim 1, characterized in that If the electrical performance evaluation result satisfies the electrical performance evaluation constraint, a trend prediction is performed according to the reactor load data to obtain a load trend prediction result, and the reactor is adjusted and optimized in combination with the electrical performance evaluation constraint to generate a second optimization strategy for reactor adjustment, including: Performing a change degree evaluation according to the load trend prediction result to obtain a load change degree; Determining whether the load change degree is greater than or equal to a predetermined load change degree; If the load variation degree is greater than or equal to the predetermined load variation degree, an adjustment decision is made on the reactor according to the load trend prediction result to obtain a third reactor adjustment decision space; Based on the three electrical performance evaluation channels, optimizing the third reactor adjustment decision space according to the electrical performance evaluation constraints to establish a fourth reactor adjustment decision space; According to the reactor adjustment fitness evaluation network, the fourth space of the reactor adjustment decision is optimized to maximize the adjustment fitness, and the second optimization strategy for the reactor adjustment is output.

8. Reactor load regulation system based on electrical performance optimization, characterized in that: The system is used to implement the reactor load regulation method based on electrical performance optimization according to any one of claims 1 to 7, and the system comprises: A real-time monitoring module, which is used to monitor the reactor in real time and obtain reactor monitoring data and reactor load data; An evaluation result acquisition module, the evaluation result acquisition module is used to input the reactor monitoring data into three channels of electrical performance evaluation to obtain electrical performance evaluation results; An evaluation result judgment module, the evaluation result judgment module is used to judge whether the electrical performance evaluation result meets the electrical performance evaluation constraint; An adjustment optimization module, wherein if the electrical performance evaluation result does not satisfy the electrical performance evaluation constraint, the adjustment optimization module is used to adjust and optimize the reactor according to the reactor load data and the electrical performance evaluation constraint to generate a first optimization strategy for reactor adjustment; A load trend prediction result acquisition module, wherein if the electrical performance evaluation result satisfies the electrical performance evaluation constraint, the load trend prediction result is obtained by performing trend prediction according to the reactor load data, and the reactor is adjusted and optimized in combination with the electrical performance evaluation constraint to generate a second optimization strategy for reactor adjustment; An adaptive load regulation module is used to perform adaptive load regulation on the reactor according to the first optimization strategy for regulating the reactor and the second optimization strategy for regulating the reactor.

Images