Power equipment control method and system based on real-time sensing and operation parameter adjustment

Abstract

The invention discloses a power equipment control method and system based on real-time perception and operation parameter adjustment, relates to the field of intelligent control, and determines operation state adjustment information by obtaining a plurality of power operation parameters of power equipment and combining operation scene parameters. Then, the power operation parameters are deeply analyzed, the optimal target operation state is identified, and it is ensured that the power grid operates on the premise of safety, stability and high efficiency. And formulating a detailed adjustment control strategy based on the optimal target operation state and the real-time operation parameters of the power equipment. And then, executing the strategy according to a plurality of control stages, and issuing a specific control instruction to dynamically adjust the operation parameters of the equipment. The process effectively improves the ability of the power grid to deal with complex working conditions, realizes accurate regulation and control, enhances the flexibility and stability of the system, and ensures the safety and reliability of power supply.

Description

Technical Field

[0001] The present application relates to the field of intelligent control, and more specifically, to a control method and system for power equipment based on real-time perception and operating parameter adjustment. Background Art

[0002] With the development of the complexity and intelligence of modern power systems, the operating environments of various power equipment in the power grid (such as transformers, circuit breakers, inverters, etc.) show dynamic and multi-scenario characteristics. Against the background of large-scale integration of new energy and increasing load fluctuations, power equipment needs to respond in real time to multi-dimensional operating state changes such as voltage surges / drops, frequency offsets, and power oscillations.

[0003] The prior art CN118763723A discloses a multi-operation mode switching control method for a flexible interconnected distribution network. It uses an operation perception unit to monitor the flexible interconnected distribution network in real time, collect power operation data and combine with scenario parameters, and uses data mining and prediction models to determine operation switching requirements. Introduce constraints such as voltage, frequency, and power limits for risk assessment, and accordingly judge and select the target operation mode. Based on the selected mode, formulate a detailed switching control strategy, and dynamically adjust operation scheduling data through linear and non-linear control stages to achieve intelligent operation scheduling and optimal configuration, ensuring the stable and efficient operation of the power grid.

[0004] In the above patent, a preset change critical value is used as a key indicator to evaluate the operating state of the power grid, used to identify whether the current operating parameters of the power grid are close to or reach an unsatisfactory or dangerous state. However, this patent method based on a static preset change critical value has several defects: First, it lacks dynamic adaptability and cannot cope with power grid state changes caused by factors such as real-time load fluctuations, easily leading to misjudgment or missed judgment. Second, the "one-size-fits-all" setting is difficult to meet the differentiated requirements of different regions, voltage levels, seasons, etc. Third, it has insufficient ability to judge complex working conditions and chain reactions, and it is difficult to identify the risks of the synergistic effect of multiple parameters. And the critical value set relying on experience and standards lags behind the development of power grid technology and cannot reflect new risk characteristics in time, restricting the effectiveness and accuracy of the control strategy.

[0005] Therefore, an optimized control scheme for power equipment based on real-time perception and operating parameter adjustment is expected. Summary of the Invention

[0006] To solve the above technical problems, the present application is proposed. According to one aspect of the present application, a control method for power equipment based on real-time perception and operating parameter adjustment is provided, which includes:

[0007] Obtain multiple power operation parameters of the power equipment through an operation perception unit, and the multiple power operation parameters include voltage, current, frequency, and power;

[0008] Determine the operation state adjustment information based on the multiple power operation parameters and operation scenario parameters;

[0009] Analyze the multiple power operation parameters to determine the optimal target operation state, including: determining a preset change critical value; determining the optimal target operation state based on the comparison between the multiple power operation parameters and the preset change critical value;

[0010] Formulate a corresponding adjustment control strategy based on the optimal target operation state and the real-time power operation parameters of the power equipment;

[0011] Combine multiple control stages, execute the adjustment control strategy and issue control instructions to adjust the equipment operation parameters;

[0012] Among them, determining the preset change critical value includes: obtaining the time queue of the power operation parameter data of the power equipment; obtaining the scenario parameter data of the power equipment; performing fine-grained response encoding and decoding of the operation parameters based on the time sequence characteristics on the time queue of the power operation parameter data and the scenario parameter data of the power equipment to obtain the preset change critical value.

[0013] According to another aspect of the present application, a power equipment control system based on real-time perception and operation parameter adjustment is provided, which includes:

[0014] A power operation parameter acquisition module for obtaining multiple power operation parameters of a power equipment through an operation perception unit, where the multiple power operation parameters include voltage, current, frequency, and power;

[0015] A determination module for determining operation state adjustment information based on the multiple power operation parameters and operation scenario parameters;

[0016] An optimal target operation state determination module for analyzing the multiple power operation parameters to determine the optimal target operation state, and the optimal target operation state determination module includes: a critical value determination unit for determining a preset change critical value; a comparison unit for determining the optimal target operation state based on the comparison between the multiple power operation parameters and the preset change critical value;

[0017] A generation module for formulating a corresponding adjustment control strategy based on the optimal target operation state and the real-time power operation parameters of the power equipment;

[0018] An adjustment module for combining multiple control stages, executing the adjustment control strategy and issuing control instructions to adjust the equipment operation parameters;

[0019] Among them, the critical value determination unit is configured to: obtain the time queue of the power operation parameter data of the power equipment; obtain the scenario parameter data of the power equipment; perform fine-grained response encoding and decoding of the operation parameters based on the time series characteristics on the time queue of the power operation parameter data and the scenario parameter data of the power equipment to obtain the preset change critical value.

[0020] Compared with the prior art, a power equipment control method and system based on real-time perception and operation parameter adjustment provided by the present application determine the operation state adjustment information by obtaining multiple power operation parameters of the power equipment and combining the operation scenario parameters. Then, these power operation parameters are deeply analyzed to identify the optimal target operation state, ensuring that the power grid operates under the premise of safety, stability, and high efficiency. Based on this optimal target operation state and the real-time operation parameters of the power equipment, a detailed adjustment control strategy is formulated. Subsequently, the strategy is executed according to multiple control stages, and specific control instructions are issued to dynamically adjust the operation parameters of the equipment. This process effectively improves the ability of the power grid to cope with complex working conditions, realizes precise control, enhances the flexibility and stability of the system, and ensures the safety and reliability of power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other objects, features, and advantages of the present application will become more obvious. The accompanying drawings are used to provide a further understanding of the embodiments of the present application, and constitute a part of the specification, and are used to explain the present application together with the embodiments of the present application, and do not constitute a limitation to the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0022] Figure 1 It is a flowchart of a power equipment control method based on real-time perception and operation parameter adjustment according to an embodiment of the present application.

[0023] Figure 2 It is a flowchart of step S3 in the power equipment control method based on real-time perception and operation parameter adjustment according to an embodiment of the present application.

[0024] Figure 3 It is a flowchart of step S31 in the power equipment control method based on real-time perception and operation parameter adjustment according to an embodiment of the present application.

[0025] Figure 4 It is a flowchart of step S313 in the power equipment control method based on real-time perception and operation parameter adjustment according to an embodiment of the present application

[0026] Figure 5 It is a block diagram of a power equipment control system based on real-time perception and operation parameter adjustment according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for exemplary purposes only and are not intended to limit the scope of protection of the present disclosure.

[0028] Based on this, the present application proposes a power equipment control method based on real-time perception and operating parameter adjustment. Figure 1 It is a flowchart of the power equipment control method based on real-time perception and operating parameter adjustment according to an embodiment of the present application. As Figure 1 shown, the power equipment control method based on real-time perception and operating parameter adjustment includes: S1, obtaining multiple power operating parameters of the power equipment through an operating perception unit, where the multiple power operating parameters include voltage, current, frequency, and power; S2, determining operating state adjustment information based on the multiple power operating parameters and operating scenario parameters; S3, analyzing the multiple power operating parameters to determine the optimal target operating state; S4, formulating a corresponding adjustment control strategy based on the optimal target operating state and the real-time power operating parameters of the power equipment; S5, combining multiple control stages, executing the adjustment control strategy and issuing a control instruction to adjust the equipment operating parameters.

[0029] In step S1, multiple power operating parameters of the power equipment are obtained through an operating perception unit, where the multiple power operating parameters include voltage, current, frequency, and power. It should be understood that the operation of the power equipment is a complex process, and a single parameter cannot comprehensively and accurately reflect its operating state. Parameters such as voltage, current, frequency, and power describe the operation of the equipment from different perspectives. By obtaining these parameters, a comprehensive understanding of the operating state of the equipment can be achieved. In particular, in the context of large-scale integration of new energy and increasing load fluctuations, the operating environment of power equipment has become dynamic and multi-scenario. Obtaining multiple power operating parameters can track the operating changes of the equipment in different environments in real time and provide a basis for subsequent adjustment of the control strategy.

[0030] In step S2, based on the multiple power operation parameters and operation scenario parameters, the operation state adjustment information is determined. In particular, the operation scenario parameters refer to a series of parameters for the current operation state and environmental conditions of the power equipment, such as environment-related parameters, new energy access ratio, distributed power supply operation state and other parameters. Accordingly, modern power systems are becoming increasingly complex, and the operation environment of various types of power equipment presents dynamic and multi-scenario characteristics. Large-scale grid connection of new energy and intensified load fluctuations have caused power equipment to face a variety of operating state changes such as voltage surge / drop, frequency offset, and power oscillation. Relying only on a single parameter or a fixed judgment method, it is impossible to accurately grasp the actual operating status of the equipment. For example, in different seasons, due to changes in electricity demand and energy supply structure, the operating state of power equipment varies greatly. By integrating multiple power operation parameters (such as voltage, current, frequency and power) and operation scenario parameters, the real-time operating state of power equipment in a complex environment can be fully and accurately reflected, providing a reliable basis for subsequent decision-making.

[0031] The implementation process is as follows: In actual operation, the operation perception unit will continuously monitor the power equipment in real time and collect multiple power operation parameters, including voltage, current, frequency and power. At the same time, the operation scenario parameters will also be obtained synchronously, such as the ambient temperature and humidity of the equipment, the season, time, and the characteristics of the power grid in the area. These rich data form the basis for subsequent analysis.

[0032] Next, the obtained multiple power operation parameters and operation scenario parameters are combined for data mining. Using data mining algorithms such as cluster analysis, association rule mining, and classification algorithms, these complex data are processed to mine the hidden patterns, relationships, and trends, and then generate operation data mining results. This result contains important information such as typical characteristics of power equipment under different operating conditions, future trend predictions, and abnormal situation detection. For example, through data mining, it may be found that during the high temperature and peak electricity consumption period in summer, the load and power of power equipment will show a specific change pattern, and there is a certain correlation between the voltage and current.

[0033] Then, the operation data mining results are deeply identified to extract multiple operation characteristics closely related to the operation of power equipment, such as voltage fluctuation amplitude, current change rate, power factor fluctuation, load change rate, etc. Based on these operation characteristics, multiple operation change curves are constructed. These curves show the change trend of the operation characteristics of power equipment at different time scales, such as hours, days, weeks, months, etc. Through these curves, operation and maintenance personnel can intuitively observe the changes in the equipment operation status over time and detect abnormal fluctuations in time.

[0034] Among numerous operation change curves, traverse and randomly or targeted select multiple operation change data points according to certain rules. These data points represent the operation conditions of power equipment under different operation conditions and scenarios. Reasonably divide the selected operation change data to construct an operation training data set and an operation test data set.

[0035] Use the operation training data set to construct a power grid operation prediction model. Common model construction algorithms include machine learning algorithms such as neural networks, support vector machines, and random forests. During the construction process, continuously adjust the model parameters and optimize the algorithm to improve the prediction accuracy of the model. After construction, use the operation test data set to strictly verify the model. By carefully comparing the prediction results of the model with the actual operation data, comprehensively evaluate the accuracy and reliability of the model. If the verification result of the model does not meet the expected performance indicators, make necessary adjustments and optimizations to the model according to the verification situation until the model meets the predetermined performance requirements, and then output and deploy the power grid operation prediction model to the management system of power equipment.

[0036] The deployed power grid operation prediction model will perform real-time or regular operation predictions on power equipment. Based on the prediction results, deeply analyze the performance, stability, and reliability of power equipment under different operation scenarios. For example, predict the operation state of power equipment under extreme weather conditions, or predict the load change of equipment during different power consumption demand periods. On this basis, comprehensively consider various factors to determine the operation state adjustment information of power equipment under different scenarios. These adjustment information cover a variety of operation suggestions. For example, when it is predicted that the equipment load is about to be too high, it is recommended to adjust the power output of the equipment or switch to the standby power supply; when it is found that the voltage is unstable, it is recommended to adjust the voltage regulating device of the relevant equipment, etc., so as to realize the intelligent operation management of power equipment, ensure that the power equipment is always in the best operation state, and improve the overall operation efficiency and reliability of the power system.

[0037] In step S3, analyze the multiple power operation parameters to determine the optimal target operation state. Figure 2 It is a flowchart of step S3 in the power equipment control method based on real-time perception and operation parameter adjustment according to an embodiment of the present application. Specifically, in the embodiment of the present application, as Figure 2 shown, in step S3, analyze the multiple power operation parameters to determine the optimal target operation state, including: S31, determine a preset change critical value; S32, based on the comparison between the multiple power operation parameters and the preset change critical value, determine the optimal target operation state.

[0038] Specifically, in step S31, a preset change critical value is determined. It should be understood that the safe and reliable operation of power equipment is the basis for the stable operation of the power system. The operation of power equipment is affected by various factors, such as load changes, environmental conditions, equipment aging, etc., resulting in the complexity and uncertainty of its operation parameters. The preset change critical value can be used as the boundary condition for the safe operation of the equipment. When the operation parameters exceed the critical value, it indicates that there may be safety risks, and measures need to be taken in a timely manner to adjust to avoid equipment failures and accidents.

[0039] Based on this, in view of the technical problems in the above-mentioned background art, in the process of determining the preset change critical value, the technical concept of this application is to first extract the time series correlation of power operation parameters (such as voltage, current, frequency, etc.) through time series feature encoding, and capture the parameter fluctuation trend and potential abnormal patterns; at the same time, perform structured embedding encoding on the scenario parameters (such as equipment type, environmental temperature, grid connection mode, etc.) to establish the spatial representation of multi-dimensional scenario features. Subsequently, use fine-grained transfer aggregation to cross-domain interact and fuse the time series encoding vector and the scenario embedding vector, and simulate the sensitivity difference of parameter changes in different scenarios through a dynamic weight allocation mechanism. Finally, convert the fused feature vector into a dynamically adjusted preset change critical value through non-linear mapping, so that it can be adaptively updated according to the real-time state of the power grid and scenario features. This concept solves the misjudgment problem caused by the "one-size-fits-all" of traditional static critical values. For example, when the new energy output fluctuates randomly, the time series encoding can identify the inertial characteristics of power changes and avoid invalid regulation triggered by instantaneous disturbances; while the scenario embedding encoding can distinguish the grid connection differences between photovoltaic power plants and wind farms and dynamically adjust the tolerance range of the critical value. In addition, the fine-grained mechanism can capture the non-linear correlation between multiple parameters (such as the voltage-frequency coordinated offset caused by a sudden increase in load), and avoid cascading misjudgments caused by the over-limit of a single parameter. By dynamically generating the critical value, it can not only meet the differentiated requirements of different regions and seasons, but also synchronize the new risk characteristics brought by the iteration of power grid technology (such as the transient oscillation caused by the grid connection of virtual synchronous machines), thereby improving the accuracy and robustness of the control strategy.

[0040] Figure 3 It is a flowchart of step S31 in the power equipment control method based on real-time perception and operation parameter adjustment according to an embodiment of the present application. Specifically, in the embodiment of the present application, as Figure 3 shown, in step S31, determining a preset change critical value includes: S311, obtaining a time queue of power operation parameter data of the power equipment; S312, obtaining scenario parameter data of the power equipment; S313, performing fine-grained response encoding and decoding of operation parameters based on time series features on the time queue of the power operation parameter data and the scenario parameter data of the power equipment to obtain the preset change critical value.

[0041] Specifically, in step S311, obtain the time queue of the power operation parameter data of the power equipment. It should be understood that the operation parameters of the power equipment (such as voltage, current, frequency) have significant time-varying characteristics. Factors such as load fluctuations, new energy access, and equipment aging will cause the parameters to change complexly over time. Analyzing only the data at a single moment cannot capture the change trend and fluctuation law of the parameters, such as the transient process of voltage sudden rise / sudden drop, and the periodic change of load. The time queue can record the continuous change of the parameters, reflect the dynamic behavior of the system, and provide better data support for the generation of the subsequent preset change critical value.

[0042] Specifically, in step S312, obtain the scenario parameter data of the power equipment. Correspondingly, the power equipment does not operate in a single and fixed environment, and the scenarios it is in cover many factors. The power grid structures and load characteristics in different regions are different, the equipment types and operating conditions are various, and the environmental conditions (such as temperature, humidity, altitude) and grid connection modes (new energy access ratio, distributed power source type) etc. will also change. These factors are intertwined and affect the operation parameters and performance of the power equipment. If only focusing on the power operation parameters themselves and ignoring the scenario parameters they are in, it is impossible to comprehensively and accurately grasp the actual operation state of the power equipment to formulate personalized critical values.

[0043] Specifically, in step S313, perform fine-grained response encoding and decoding of the operation parameters based on the time series characteristics on the time queue of the power operation parameter data and the scenario parameter data of the power equipment to obtain the preset change critical value. Figure 4 It is a flowchart of step S313 in the power equipment control method based on real-time perception and operation parameter adjustment according to the embodiment of the present application. Specifically, in the embodiment of the present application, as Figure 4 shown, in step S313, performing fine-grained response encoding and decoding of the operation parameters based on the time series characteristics on the time queue of the power operation parameter data and the scenario parameter data of the power equipment to obtain the preset change critical value includes: S3131, performing time series feature encoding on the time queue of the power operation parameter data to obtain a time series encoding vector of the power operation parameter data; S3132, performing structured embedding encoding on the scenario parameter data of the power equipment to obtain a structured embedding encoding vector of the scenario parameters; S3133, performing fine-grained response fusion processing of the operation parameter interaction on the time series encoding vector of the power operation parameter data and the structured embedding encoding vector of the scenario parameters to obtain a response fusion encoding vector of the power operation parameters; S3134, obtaining the preset change critical value based on the response fusion encoding vector of the power operation parameters.

[0044] More specifically, in the embodiments of the present application, step S3131, performing temporal feature encoding on the time queue of the power operation parameter data to obtain a temporal encoding vector of the power operation parameter data, includes: performing temporal feature encoding based on causal dilated convolution on the time queue of the power operation parameter data to obtain the temporal encoding vector of the power operation parameter data. It should be understood that considering that the temporal data of parameters such as voltage, current, and frequency often contains complex short-term fluctuations and long-term evolution characteristics. For example, the power oscillation caused by the intermittency of new energy output may present the superposition characteristics of high-frequency short-time disturbances and low-frequency trend offsets, while traditional temporal analysis methods (such as sliding window statistics or simple recurrent neural networks) are difficult to capture the correlation patterns of different time scales at the same time. In addition, the power equipment control has extremely high requirements for real-time performance. If data of future time points (such as standard convolution operations) are introduced during the feature encoding process, it will violate the causality constraint, resulting in delays or mistriggers of the control strategy. Therefore, a feature extraction method that can both model long-range dependence relationships and strictly follow temporal causality is needed to accurately represent the potential laws of parameter changes. Based on this, the present application performs temporal feature encoding based on causal dilated convolution on the time queue of the power operation parameter data to obtain the temporal encoding vector of the power operation parameter data. Specifically, by designing a convolutional kernel stacking structure with dilation coefficients, the temporal receptive field is gradually expanded, enabling the network to capture multi-scale dynamic features from the time queue of power operation parameters. The "unidirectional mask" feature of causal convolution ensures that the features at each time point are generated only by historical and current data, avoiding future information leakage; and the dilation mechanism expands the convolution coverage range through interval sampling. For example, when analyzing the voltage drop caused by a sudden increase in load, the low-level convolution captures millisecond-level transient fluctuations, and the high-level convolution identifies second-level recovery trends, forming a multi-granularity feature representation. At the same time, the parallel computing advantage of the convolution operation significantly reduces the delay of temporal modeling, meets the real-time requirements of power equipment control, and forms a comprehensive temporal feature representation.

[0045] More specifically, in step S3132, the scenario parameter data of the power equipment is subjected to structured embedding encoding to obtain a scenario parameter structured embedding encoding vector. Correspondingly, considering that scenario parameters such as equipment type, ambient temperature, and grid connection mode have multi-dimensional heterogeneous characteristics. For example, the response mechanisms of photovoltaic power plants and wind farms to grid frequency disturbances are essentially different, and the change in the insulation performance of the same equipment in high and low temperature environments will also affect the safety boundaries of its operating parameters. Traditional methods usually simplify such scenario parameters into discrete labels or independent input features, resulting in the inability to effectively model the potential correlations between different parameters (such as the combined effects of grid connection mode and load type). For example, when a distributed energy storage device operates in island mode, its voltage regulation threshold needs to be set differently from the standard in the main grid connection mode, but static critical values are difficult to dynamically distinguish the internal correlations of such scenario features, easily causing insufficient adaptability of control strategies. Therefore, in the technical solution of this application, the scenario parameter data of the power equipment is subjected to structured embedding encoding to obtain a scenario parameter structured embedding encoding vector. Specifically, for categorical parameters such as equipment type, an embedding layer is used to transform them into dense vectors to capture the functional similarities between different types of equipment (such as the potential correlation between photovoltaic inverters and wind power converters in reactive power compensation characteristics); for continuous parameters such as temperature and humidity, feature expressions related to the physical characteristics of the equipment are constructed through normalization and non-linear transformation. At the same time, an attention mechanism is introduced to dynamically weight the importance of different scenario parameters. For example, in the high-temperature scenario in summer, the constraint weight of the heat dissipation condition on the equipment load capacity is strengthened, while in the low-temperature scenario in winter, the attenuation effect of the insulation medium performance is highlighted. This encoding method not only retains the topological structure between parameters but also reflects the potential impact of scenario differences on the operating state of the equipment through the distance metric in the vector space.

[0046] More specifically, in step S3133, performing an operation parameter interaction fine-grained response fusion process on the power operation parameter data time-series coding vector and the scenario parameter structured embedding coding vector to obtain a power operation parameter response fusion coding vector, including: S3133-1, performing an equal-grained ordering process on the power operation parameter data time-series coding vector and the scenario parameter structured embedding coding vector to obtain a sequence of power operation parameter time-series local feature ordered coding vectors and a sequence of scenario parameter structured embedding ordered coding vectors; S3133-2, inputting each pair of corresponding power operation parameter time-series local feature ordered coding vectors and scenario parameter structured embedding ordered coding vectors in the sequence of power operation parameter time-series local feature ordered coding vectors and the sequence of scenario parameter structured embedding ordered coding vectors into a transfer response unit to obtain a sequence of power operation parameter response local transfer response coding matrices; S3133-3, performing a transfer response fusion on the sequence of power operation parameter response local transfer response coding matrices to obtain the power operation parameter response fusion coding vector. Particularly, considering that the interaction relationship between time-series operation parameters (such as the fluctuation trends of voltage and current) and scenario parameters (such as equipment type and environmental conditions) is highly complex. For example, the power output attenuation of a photovoltaic inverter in a high-temperature environment may have a non-linear coupling with the time-series characteristics of current harmonics. When traditional methods use simple weighting or splicing fusion, it is difficult to distinguish the causal relationship between parameters (such as how the environmental temperature affects the equipment heat dissipation and modulates the safety boundary of current fluctuations). In addition, multi-parameter dynamic interactions in new energy grid connection scenarios often exhibit cross-scale characteristics: a millisecond-level impact of load mutation may cause an instantaneous voltage drop, while a change in environmental humidity indirectly leads to the accumulation of leakage current after several hours through the aging of insulating media. If only a surface association is made between time-series and scenario characteristics, such cross-domain and cross-time implicit associations cannot be captured, easily resulting in the omission of key risk factors when generating critical values. Based on this, in this application, an operation parameter interaction fine-grained response fusion process is performed on the power operation parameter data time-series coding vector and the scenario parameter structured embedding coding vector to obtain a power operation parameter response fusion coding vector.

[0047] Particularly, in the embodiment of this application, step S3133-1, performing an equal-grained ordering process on the power operation parameter data time-series coding vector and the scenario parameter structured embedding coding vector to obtain a sequence of power operation parameter time-series local feature ordered coding vectors and a sequence of scenario parameter structured embedding ordered coding vectors, including:

[0048] Performing an ordered arrangement on the power operation parameter data time-series coding vector and the scenario parameter structured embedding coding vector based on the eigenvalue size to obtain a power operation parameter time-series feature ordered arrangement vector and a scenario parameter structured embedding ordered arrangement vector, which can be expressed by the formula:

[0049]

[0050]

[0051] Among them, is the time series coding vector of power operation parameter data, is the structured embedding coding vector of scenario parameters, represents sorting the vector elements, is the ordered arrangement vector of the time series characteristics of power operation parameters, is the ordered arrangement vector of the structured embedding of scenario parameters;

[0052] Equal-granularity feature segmentation is performed on the ordered arrangement vector of the time series characteristics of the power operation parameters and the ordered arrangement vector of the structured embedding of the scenario parameters to obtain the sequence of the ordered coding vectors of the local time series characteristics of the power operation parameters and the sequence of the ordered coding vectors of the structured embedding of the scenario parameters, which can be expressed by the formula:

[0053]

[0054]

[0055] Among them, represents the feature segmentation function, is the sequence of the ordered coding vectors of the local time series characteristics of the power operation parameters, is the sequence of the ordered coding vectors of the structured embedding of the scenario parameters, , , and are respectively the 1st, 2nd, th, and th ordered coding vectors of the local time series characteristics of the power operation parameters in the sequence of the ordered coding vectors of the local time series characteristics of the power operation parameters, , , and are respectively the 1st, 2nd, th, and th ordered coding vectors of the structured embedding of the scenario parameters in the sequence of the ordered coding vectors of the structured embedding of the scenario parameters.

[0056] It should be understood that the arrangement order of the characteristic elements of the time series coding vector of power operation parameter data (such as the fluctuation trend of voltage and current) and the structured embedding coding vector of scenario parameters (such as equipment type, environmental conditions) is often interfered by the sensor acquisition logic or manually defined rules. For example, the same device may output parameters in the order of current-voltage-frequency or voltage-frequency-current in different monitoring systems, resulting in the original arrangement of the feature vector being random. When directly using the feature vector with the original arrangement for fusion, the model may mistakenly regard the parameter order as valid information (such as mistakenly believing that the device with the current parameter at the head of the vector has a higher failure risk), thus interfering with the analysis of the actual feature intensity distribution. Therefore, by arranging the time series coding vector of the power operation parameter data and the structured embedding coding vector of the scenario parameters in an ordered manner based on the eigenvalue size, the elements in the time series coding vector and the scenario embedding vector are reordered according to the numerical intensity. For example, the voltage time series fluctuation characteristics are arranged in sequence from the highest amplitude to the lowest amplitude, and at the same time, the environmental temperature, equipment load rate, etc. in the scenario parameters are arranged in descending order according to the influence degree, obtaining an ordered arrangement vector of the time series characteristics of the power operation parameters and an ordered arrangement vector of the structured embedding of the scenario parameters, so that the structure of the feature vector reflects the intensity distribution characteristics of the parameters themselves.

[0057] Accordingly, although the ordered arrangement of temporal features (such as voltage fluctuation trends) and scenario features (such as ambient temperature) eliminates the randomness of the original arrangement, it is still difficult to accurately capture the dynamic correlations in local intervals at the global level. For example, the power temporal anomalies caused by new energy output fluctuations may be concentrated in a certain intensity interval (such as the medium-high power segment), and the adjustment requirements for this interval in the corresponding grid connection mode (such as island operation) in the scenario are specific. If direct global fusion is performed, it is easy to ignore the coupling characteristics between the local intensity interval and the scenario constraints. Therefore, in order to strengthen the interaction information between the two, in this application, equal-grained feature segmentation is performed on the ordered arrangement vector of the temporal features of the power operation parameters and the ordered arrangement vector of the structured embedding of the scenario parameters to obtain a sequence of ordered encoded vectors of the local temporal features of the power operation parameters and a sequence of ordered encoded vectors of the structured embedding of the scenario parameters. That is, the ordered temporal feature vector and the scenario embedding vector are evenly cut into multiple local sub-vector segments along the feature dimension. For example, the ordered voltage temporal vector is divided into several equal-length intervals according to the intensity level, and each interval corresponds to a feature pattern with different fluctuation amplitudes (such as high-voltage transient, steady-state operation, low-voltage recovery stage). At the same time, the scenario parameter vector is synchronously cut into local segments that match dimensions such as device attributes and environmental conditions. This aligned segmentation enables subsequent interaction reasoning to focus on the cross-domain correlations in the same intensity interval: when analyzing the grid connection scenario of a wind farm, the high-power temporal segment corresponding to high wind speed interacts locally with the scenario segment of the rated capacity of the wind turbine to quantify the dynamic impact of the power overlimit risk on the frequency threshold; while in a low-temperature environment, the weak insulation interval of the current ordered vector interacts with the device material parameter segment to identify early signs of leakage current accumulation.

[0058] Specifically, in step S3133-2, each group of corresponding ordered encoded vectors of the local temporal features of the power operation parameters and the ordered encoded vectors of the structured embedding of the scenario parameters in the sequence of ordered encoded vectors of the local temporal features of the power operation parameters and the sequence of ordered encoded vectors of the structured embedding of the scenario parameters are input into the transfer response unit to obtain a sequence of local transfer response coding matrices of the power operation parameter responses, which can be expressed by the formula:

[0059]

[0060] Wherein, is the th ordered encoded vector of the local temporal features of the power operation parameters in the sequence of ordered encoded vectors of the local temporal features of the power operation parameters, is the th ordered encoded vector of the structured embedding of the scenario parameters in the sequence of ordered encoded vectors of the structured embedding of the scenario parameters, is the activation function, is the temporal weight matrix of the power operation parameters, For and the power operation parameter response local transfer response coding matrix in between, that is, the th power operation parameter response local transfer response coding matrix in the sequence of power operation parameter response local transfer response coding matrices.

[0061] It should be understood that the local interaction relationship between the timing operation characteristics (such as voltage fluctuation trend) and the scenario constraints (such as device operation mode) often presents non-uniformity and conditional dependence. For example, during the charge and discharge process of the energy storage system, the local high-amplitude interval of the current timing (such as the fast charging stage) may have a strong correlation with the battery temperature interval in the scenario parameters, while the low-amplitude interval (such as trickle charging) has a higher correlation with the environmental humidity. The traditional global interaction model regards the entire feature vector as a single entity for correlation analysis, and cannot distinguish the differences in the effects of parameters in different intensity intervals, which easily leads to the dilution of key local correlations by global statistical features. For example, when analyzing the frequency disturbance caused by the grid connection of a wind turbine, if the power fluctuation characteristics corresponding to the high wind speed period are not locally correlated with specific grid connection protection parameters, the need to dynamically tighten the frequency threshold in this interval may be ignored. Based on this, the ordered coding vectors of the local characteristics of each group of corresponding power operation parameter timings and the structured embedding ordered coding vectors of the scenario parameters are input into the transfer response unit to obtain a sequence of power operation parameter response local transfer response coding matrices, so as to perform in-depth interaction modeling on each group of corresponding timing local characteristics and scenario local characteristics, realizing the refined coupling analysis of cross-domain local characteristics and significantly improving the scenario adaptability of critical value generation.

[0062] Specifically, in the embodiment of the present application, in step S3133-3, performing transfer response fusion on the sequence of the power operation parameter response local transfer response coding matrices to obtain the power operation parameter response fusion coding vector includes: flattening each power operation parameter response local transfer response coding matrix in the sequence of the power operation parameter response local transfer response coding matrices into a vector to obtain a sequence of power operation parameter response local transfer response coding vectors; respectively performing local transfer response regularization coding based on a Poisson-like constraint on each power operation parameter response local transfer response coding vector in the sequence of the power operation parameter response local transfer response coding vectors and the sequence of the power operation parameter response local transfer response coding matrices to obtain a sequence of power operation parameter response local transfer response constraint coding vectors; and inputting the sequence of the power operation parameter response local transfer response constraint coding vectors into a transfer response fusion device based on LSTM to obtain the power operation parameter response fusion coding vector. Correspondingly, although the power operation parameter response local transfer response coding matrix accurately depicts the interaction details between the time series parameters and the scenario constraints within a specific feature intensity range (such as the correlation intensity between the voltage dip period and the device heat dissipation condition), these scattered local interaction information often has spatial fragmentation and time series evolution. For example, when analyzing the cyclic charge and discharge process of an energy storage system, the local matrix of current harmonics in the charging stage and the local matrix of voltage recovery in the discharging stage respectively reflect the risk characteristics of different operation intervals, but the two may have a cross-interval cumulative effect through the battery aging effect. If only relying on independent analysis of each local matrix, such hidden correlations across intensity intervals cannot be captured, resulting in ignoring the long-term evolution law of parameter interaction when generating the critical value, such as underestimating the progressive impact of electrode corrosion on the voltage safety boundary in a frequent charge and discharge scenario. Based on this, the present application performs transfer response fusion on the sequence of the power operation parameter response local transfer response coding matrices to realize the sublimation of local interaction features to global semantics, obtain the power operation parameter response fusion coding vector, and provide feature support with both fine granularity and systematicness for critical value generation.

[0063] Specifically, flattening each power operation parameter response local transfer response coding matrix in the sequence of the power operation parameter response local transfer response coding matrices into a vector to obtain a sequence of power operation parameter response local transfer response coding vectors can be expressed by the formula:

[0064]

[0065] Wherein, represents flattening the matrix into a vector, is the flattened power operation parameter response local transfer response coding vector.

[0066] Specifically, local transfer response regularization encoding based on Poisson-like constraints is performed on each power operation parameter response local transfer response encoding vector and the sequence of power operation parameter response local transfer response encoding matrices in the sequence of power operation parameter response local transfer response encoding vectors to obtain a sequence of power operation parameter response local transfer response constrained encoding vectors, which can be expressed by the formula:

[0067]

[0068] Wherein, is the exponential function value with the natural constant as the base, is to calculate the F-norm of, is the number of matrices in the sequence of power operation parameter response local transfer response encoding matrices, is the activation function, is the th power operation parameter response local transfer response constrained encoding vector in the sequence of power operation parameter response local transfer response constrained encoding vectors.

[0069] Specifically, constraint iteration is performed based on the local region size to obtain . The specific processing is as follows:

[0070] It can be seen that when the segmentation granularity of the feature segmentation function affects the local region size of the power operation parameter response local transfer response encoding matrix, it will also directly affect the expression of the interaction mode between the sequence of corresponding power operation parameter time-series local feature ordered encoding vectors and the sequence of scene parameter structured embedding ordered encoding vectors.

[0071] Since the power operation parameter response local transfer response encoding matrix is the transfer response between two local regions, it essentially expresses the spatial measure of the transfer response space based on its row vectors, and the length of the row vector is also the representation of the above local region size. If the local region size, that is, the row vector length is introduced as the local region size intensity constraint, then the low-rank spatial measure expression of the power operation parameter response local transfer response encoding matrix, that is, the F-norm should follow the Poisson-like relationship:

[0072]

[0073] That is to say, the row vector length acts on the low-rank spatial measure of the local region transfer response space as the local region size intensity constraint for times.

[0074] From this, we can solve the parameter Thus, the transfer-response interaction is introduced by and the mean expectation is In the case of a Poisson-like process, and The edge connection representation describing the spatial measure further determines the transfer response connection probability between two local regions as:

[0075]

[0076] Then, the local area size is adjusted based on the transfer response connection probability. Perform iterative correction:

[0077]

[0078] That is, under strict guarantee of the expected degree In this case, the regularization of the overall spatial measure is determined by the regularity constraint of the mean expected connection probability fluctuation of each row, so that the structured interactive information within the local spatial measure can avoid local overfitting and improve the overall expression effect of the sequence of the power operation parameter response local transfer response encoding matrix.

[0079] In particular, the sequence of the power operation parameter response local transfer response constraint encoding vector is input into the LSTM-based transfer response fuser to obtain the power operation parameter response fusion encoding vector, which can be expressed as follows:

[0080]

[0081] in, , and They are the first, second and third in the sequence of power operation parameter response local transfer response constraint encoding vectors. and The power operation parameter response local transfer response constraint encoding vector, yes coding, It is the power operation parameter response fusion coding vector.

[0082] More specifically, in the embodiments of the present application, step S3134, obtaining the preset change critical value based on the power operation parameter response fusion coding vector, includes: inputting the power operation parameter response fusion coding vector into a decoder-based critical value generator to obtain the preset change critical value. That is to say, although the power operation parameter response fusion coding vector integrates the key features of power operation parameters and scenario parameters, these features exist in the form of vectors and are difficult to directly use to judge whether the operation state of the power equipment is normal. The decoder-based critical value generator can convert and process this abstract vector data, map it to a specific preset change critical value that can be used to measure the change range of the power equipment operation parameters, and it can comprehensively consider various factors according to the operation characteristics of the power equipment and the correlation relationship between parameters in different scenarios included in the vector, so as to determine a preset change critical value that more conforms to the actual operation situation. This conversion is a key step in realizing the effective evaluation and control of the operation state of power equipment, enabling complex data information to be converted into actual actionable indicators.

[0083] In summary, step S31 is clarified. In power equipment control, to accurately set the preset change critical value, it first uses time series feature coding to mine the time series correlation of power operation parameters such as voltage and current, capture the fluctuation trend and anomalies; at the same time, it performs structured embedding coding on scenario parameters such as equipment type and ambient temperature to construct a multi-dimensional feature representation. Then, through fine-grained transfer aggregation, the two types of coding vectors are cross-domain interactively fused, and dynamic weights are used to reflect the parameter change sensitivity in different scenarios. Finally, through non-linear mapping, the fused vector is converted into a dynamic critical value, which is updated in real time with the grid state and scenarios. This method effectively avoids the misjudgment problem of traditional static critical values and improves the accuracy and anti-interference ability of power equipment control strategies.

[0084] Specifically, in the embodiment of the present application, step S32, determining the optimal target operating state based on the comparison of the multiple power operation parameters and the preset change critical value includes: in response to the multiple power operation parameters reaching the preset change critical value, determining the adjustment time point based on the operating state adjustment information, and generating a to-be-adjusted instruction; obtaining the optimal target operating state based on the to-be-adjusted instruction and the adjustment constraint conditions; in response to the multiple power operation parameters not reaching the preset change critical value, verifying the multiple power operation parameters to determine the optimal target operating state. It should be understood that when the power equipment is operating, multiple power operation parameters (such as voltage, current, frequency, and power) will be comprehensively affected by various factors, including load changes, environmental conditions, equipment aging, etc., showing complex and variable characteristics. The preset change critical value, as an important reference for measuring the operating state of the equipment, is compared with multiple power operation parameters, which can comprehensively evaluate the operating condition of the equipment from multiple dimensions and more accurately grasp the actual operating situation of the equipment. By comparing the multiple power operation parameters with the preset change critical value, it is possible to timely detect whether the operating parameters of the equipment exceed the safe range. When the parameters are close to or exceed the critical value, it indicates that there may be safety risks in the equipment. Determining the optimal target operating state can provide a clear direction for the adjustment of the equipment, enabling the equipment to quickly return to the safe operating range, avoiding equipment failures or damages caused by abnormal parameters, thus ensuring the safe and reliable operation of the power equipment, reducing the occurrence of power outages, and improving the stability and reliability of the power system.

[0085] The implementation process is as follows: When multiple power operation parameters reach the preset change critical value, it indicates that the operating state of the power equipment has approached or is in an unsatisfactory or even dangerous range. At this time, it is particularly crucial to determine the adjustment time point based on the previously determined operating state adjustment information. The operating state adjustment information is obtained by comprehensively analyzing power operation parameters and operating scenario parameters, and includes judgments on the operating trends of the equipment and suggestions on adjustment directions. For example, if it is monitored that the current of the power equipment continues to rise and reaches the preset change critical value, by combining the analysis of the current load growth trend and the equipment heating situation in the operating state adjustment information, an adjustment time point can be accurately determined that can not only avoid damage to the equipment due to overload but also ensure the stability of power supply.

[0086] After determining the adjustment time point, an adjustment instruction to be generated. This instruction is like a detailed operation guide, specifying the specific operations required to restore the power equipment to its optimal target operating state. For example, it may instruct to reduce the power output of the equipment or adjust the relevant voltage regulating devices to stabilize the voltage. Then, based on the adjustment instruction and adjustment constraints, the optimal target operating state is determined. The adjustment constraints are a series of rules to ensure the safe and stable operation of the power equipment, covering aspects such as voltage range constraints, frequency range constraints, and power limit constraints. According to these constraints, multiple possible adjustment schemes are screened and matched. Taking the adjustment of power output as an example, it is necessary to meet the requirement of reducing power in the adjustment instruction, ensure that the adjusted power is within the power limit constraint range, and also consider the impact on voltage and frequency during the adjustment process to ensure that they are within the corresponding constraint ranges. Through such a matching process, the optimal target operating state that can meet the current operating requirements and various constraints is finally determined.

[0087] When multiple power operation parameters do not reach the preset change critical values, it does not mean that the operation of the power equipment is without hidden dangers. To determine the optimal target operating state, it is necessary to verify multiple power operation parameters. Usually, the previously constructed power grid operation prediction model is used to predict the future operating state of the power equipment. If the prediction result shows that at a certain future time point, the power operation parameters may reach or exceed the preset change critical values, it indicates that although the current operating state seems normal, there are potential risks. At this time, an adjustment instruction to be generated as well. For example, if it is predicted that as the load continues to increase, the power of the equipment may exceed the critical value within the next few hours, an adjustment instruction will be generated in advance. The subsequent process is similar to when the operation parameters reach the critical value. Based on the adjustment instruction and adjustment constraints, multiple operating modes are matched to determine the optimal target operating state, and measures are taken in advance to prevent the deterioration of the equipment operation parameters and ensure the continuous and stable operation of the power equipment.

[0088] Throughout the process, whether the operation parameters reach the critical value or there are predicted risks when they do not reach the critical value, after determining the optimal target operating state, it is necessary to continuously monitor the operating state of the power equipment. On the one hand, ensure that the equipment operates according to the expected optimal target operating state; on the other hand, according to the feedback of the actual operating situation, continuously optimize the preset change critical values, adjustment constraints, and operation prediction model to adapt to the operating requirements of the power equipment in different environments and working conditions, and further improve the operation management level of the power equipment and the stability of the power system.

[0089] In step S4, based on the optimal target operating state and the real-time power operation parameters of the power equipment, a corresponding adjustment and control strategy is formulated. That is, the operation of power equipment is vulnerable to various factors, and the operation parameters often deviate from the optimal target operating state. Load mutations, environmental changes, equipment aging, etc. can cause fluctuations in parameters such as voltage, current, and frequency. If not adjusted in time, it may lead to equipment failures. Taking a transformer as an example, when the load is overloaded, if not adjusted, the continuously rising temperature will accelerate the insulation aging and may even cause a short-circuit fault in severe cases.

[0090] The implementation process is as follows: First, deeply analyze the characteristics of the optimal target operating state. This includes a detailed study of key elements such as the voltage level of the power equipment, frequency stability, power distribution, and energy supply mode. For example, it is clear that in the optimal target operating state, the voltage of the equipment should be maintained within a specific stable range, the frequency should be kept within the standard range, the power distribution should reach an optimal configuration, and the energy supply mode should meet the requirements of high efficiency and reliability. By deeply understanding these characteristics, it is possible to predict the operating conditions of the power equipment after switching to this state and the possible problems it may face.

[0091] At the same time, comprehensively evaluate the current real-time operating state of the power equipment. Examine in detail the working states of each component in the equipment, such as whether transformers, lines, switches, etc. are operating normally; sort out the topological structure of the power grid to understand the power transmission paths and connection methods; master the energy reserve and dispatching capabilities, including the storage capacity of various types of energy and the flexibility of allocation. By accurately grasping the current state, it provides a realistic basis for formulating a reasonable adjustment and control strategy later.

[0092] Next, with the help of power system simulation software or models, simulate the operation process of the power equipment in the optimal target operating state. During the simulation process, focus on the change trends of key parameters such as voltage, frequency, and power flow. Combining the historical operating data and real-time collected data of the power equipment, use data analysis techniques to make a more accurate prediction of the operating state. Through simulation and prediction, evaluate the stability and reliability of the power equipment during the process of switching to the optimal target operating state, and discover possible risks and problems in advance, such as excessive voltage fluctuations, frequency deviation exceeding the standard, equipment overload, etc.

[0093] For the identified risks, conduct a comprehensive and in-depth assessment. Analyze the possibility of each risk occurring, and judge whether it is a high-probability event or a low-probability event; evaluate the impact degree of the risk on the power equipment and the entire power system once it occurs, whether it is a minor impact or will cause a serious failure. For example, if a certain line has a high probability of overload during the switching process and the overload may cause the line to trip, affecting a large area of power supply, then it needs to be focused on and corresponding countermeasures should be formulated. These assessment results will become an important basis for formulating the adjustment and control strategy.

[0094] After fully understanding the optimal target operating state, the current real-time operating state, the simulation prediction results, and the risk assessment, proceed to formulate a detailed adjustment and control strategy. First, determine the optimal adjustment time points and the adjustment order of equipment and lines. For example, according to the predicted voltage fluctuation trend, select the moment with the smallest voltage fluctuation to adjust the relevant equipment, and at the same time reasonably arrange the order of adjusting equipment to avoid chain reactions caused by improper adjustment. Secondly, accurately set the control ranges and thresholds of key parameters such as voltage, frequency, and power. Clearly define the allowable fluctuation ranges of these parameters under different operating conditions, and immediately take corresponding control measures once the threshold is exceeded. Then, formulate emergency measures for dealing with emergencies. For example, for equipment failures, plan in advance the process of putting into use standby equipment; for energy supply interruptions, formulate an emergency energy allocation plan to ensure the accuracy and efficiency of the adjustment process.

[0095] In step S5, combine multiple control stages, execute the adjustment and control strategy, and issue control instructions to adjust the operating parameters of the equipment. It should be understood that different control stages have different control characteristics and advantages. By combining multiple control stages to execute the adjustment and control strategy, the operating parameters of power equipment can be monitored and adjusted in real time, so that they always remain within a safe and stable operating range.

[0096] The implementation process is as follows: First, it is necessary to clarify that there are different control stages in the operation of power equipment, among which the linear control stage and the non-linear control stage are particularly crucial. The linear control stage is applicable to scenarios where the operating parameters of power equipment are relatively stable and change little, and the entire system is in a high stability state; while the non-linear control stage is used to deal with situations where the operating parameters of power equipment change violently and the dynamic characteristics of the system are complex. The cooperation of these two control stages can comprehensively cover the requirements of power equipment under different operating states.

[0097] When entering the link of executing the adjustment and control strategy, according to the pre-formulated adjustment and control strategy, corresponding operations are carried out in the linear control stage. The strategy execution focuses on the key operating parameters of power equipment, including fine adjustment of parameters such as voltage, frequency, and power, as well as optimization of the equipment operating state and reasonable allocation of the power grid topology structure. For example, when it is detected that the voltage of the power equipment shows a slight fluctuation, the voltage is fine-tuned in a timely manner according to the strategy to keep it within the stable range and ensure the normal operation of the equipment. After completing the operations in the linear control stage, a first operating adjustment data set that details the operation of the power equipment in this stage will be generated, covering information such as the equipment operating state, parameter change details, and equipment adjustment conditions, providing key basis for subsequent analysis.

[0098] As the operating conditions of power equipment change, if parameter changes intensify and the system dynamic characteristics become complex, it is necessary to switch to the nonlinear control stage. In this stage, control operations are also performed according to the regulation control strategy, but due to its complexity, more advanced control algorithms and strategies need to be used. For example, intelligent algorithms are used to handle complex power distribution problems to ensure that power equipment can operate efficiently under complex working conditions. After executing the strategy of the nonlinear control stage, the second operation regulation data set is generated to record the operating details of the power equipment in this stage, providing supplementary information for a comprehensive understanding of the equipment operation process.

[0099] Based on the generated first operation regulation data set and second operation regulation data set, the operation feedback analysis of the power equipment is carried out. The key indicators such as the operation stability, economy and safety of the power equipment are comprehensively evaluated, and possible problems and potential risks are carefully checked. Through analysis, it is found that a certain line has an overload risk, or the energy consumption of some equipment is too high, affecting the economy. Based on these feedback results, control instructions are generated in a targeted manner. The instruction content includes the adjustment of control parameters, the optimization of equipment scheduling, and the change of the power grid topology structure, etc., aiming to comprehensively improve the operation efficiency and stability of the power equipment.

[0100] Next, we enter the dynamic correction link for the operation adjustment data set. Taking the linear control stage as an example, according to the control instruction, a specific adjustment optimization step size is used to correct the first operation adjustment data set to obtain the first adjustment correction parameter. This parameter reflects the operating state that the power equipment may reach after preliminary adjustment. Then, the first adjustment correction parameter is used to predict the operation switching adjustment. By simulation or calculation, the operating state of the power equipment after correction is inferred, and the change trend of key parameters such as voltage, frequency, and power flow is obtained, and then the first adjustment parameter is obtained. This parameter represents the actual operating state of the power equipment after correction. At the same time, according to the specific adjustment objectives and evaluation criteria, the adjustment fitness corresponding to the first adjustment parameter is calculated. The adjustment fitness is used to measure the degree of fit between the operating state of the power equipment under the current adjustment parameters and the preset adjustment target. The preset adjustment target comprehensively considers multiple factors such as economy, stability and safety.

[0101] Judge the magnitude relationship between the adjustment fitness of the first adjustment parameter and the preset adjustment threshold. The preset adjustment threshold is set according to the actual situation and adjustment target of the power equipment and is an important criterion for measuring the adjustment effect. If the adjustment fitness is greater than the preset adjustment threshold, it indicates that the current adjustment parameter is relatively ideal and only fine-tuning is required. At this time, reduce the adjustment optimization compensation, that is, reduce the step size of the next correction to achieve a more refined adjustment. If the adjustment fitness is less than or equal to the preset adjustment threshold, it means that there is still room for optimization of the current adjustment parameter, and it is necessary to increase the adjustment optimization compensation and increase the step size of the next correction to accelerate the adjustment process. After adjustment, a new first adjustment correction parameter is obtained and incorporated into the correction result.

[0102] The configuration of the adjustment optimization compensation constructs an adjustment parameter adjustment space centered on the preset adjustment threshold and with the adjustment optimization step size as the span. This space shows all possible value ranges of the adjustment parameters and their corresponding adjustment fitness evaluation results, providing an intuitive basis for reasonably configuring the adjustment optimization compensation. According to the specific situation of the adjustment parameter adjustment space and the real-time operating state of the power equipment, determine an appropriate adjustment optimization compensation value so that it can not only ensure the stable operation of the power equipment but also optimize the adjustment target.

[0103] In summary, the power equipment control method based on real-time perception and operating parameter adjustment according to the embodiments of the present application is elucidated. It determines the operating state adjustment information by obtaining multiple power operating parameters of the power equipment and combining the operating scenario parameters. Then, these power operating parameters are deeply analyzed to identify the optimal target operating state to ensure the operation of the power grid under the premise of safety, stability, and high efficiency. Based on this optimal target operating state and the real-time operating parameters of the power equipment, a detailed adjustment control strategy is formulated. Subsequently, this strategy is executed according to multiple control stages, and specific control instructions are issued to dynamically adjust the operating parameters of the equipment. This process effectively improves the ability of the power grid to cope with complex working conditions, achieves precise control, enhances the flexibility and stability of the system, and ensures the safe and reliable power supply.

[0104] Figure 5 For the block diagram of the power equipment control system based on real-time perception and operating parameter adjustment according to the embodiments of the present application. As Figure 5As shown in the figure, the power equipment control system 100 based on real-time perception and operating parameter adjustment according to an embodiment of the present application includes: a power operation parameter acquisition module 110, configured to obtain multiple power operation parameters of the power equipment through an operation perception unit, where the multiple power operation parameters include voltage, current, frequency, and power; a determination module 120, configured to determine operation state adjustment information based on the multiple power operation parameters and operation scenario parameters; an optimal target operation state determination module 130, configured to analyze the multiple power operation parameters to determine an optimal target operation state, where the optimal target operation state determination module includes: a threshold determination unit, configured to determine a preset change threshold; a comparison unit, configured to determine the optimal target operation state based on a comparison between the multiple power operation parameters and the preset change threshold; a generation module 140, configured to formulate a corresponding adjustment control strategy based on the optimal target operation state and the real-time power operation parameters of the power equipment; an adjustment module 150, configured to execute the adjustment control strategy in combination with multiple control stages and issue a control instruction to adjust the equipment operation parameters. Wherein, the threshold determination unit is configured to: obtain a time queue of power operation parameter data of the power equipment; obtain scenario parameter data of the power equipment; perform fine-grained response encoding and decoding of operation parameters based on temporal features on the time queue of the power operation parameter data and the scenario parameter data of the power equipment to obtain the preset change threshold.

[0105] Here, those skilled in the art can understand that the specific operations of each step in the above power equipment control system based on real-time perception and operating parameter adjustment have been described in detail above with reference to Figures 1 to 4 the description of the power equipment control method based on real-time perception and operating parameter adjustment, and therefore, the repeated description thereof will be omitted.

[0106] As described above, the power equipment control system 100 based on real-time perception and operating parameter adjustment according to an embodiment of the present disclosure can be implemented in various wireless terminals, such as a server having a power equipment control algorithm based on real-time perception and operating parameter adjustment. In a possible implementation manner, the power equipment control system 100 based on real-time perception and operating parameter adjustment according to an embodiment of the present disclosure can be integrated into a wireless terminal as a software module and / or a hardware module. For example, the power equipment control system 100 based on real-time perception and operating parameter adjustment can be a software module in the operating system of the wireless terminal, or can be an application program developed for the wireless terminal; of course, the power equipment control system 100 based on real-time perception and operating parameter adjustment can also be one of the numerous hardware modules of the wireless terminal.

[0107] Alternatively, in another example, the power device control system 100 based on real-time perception and operating parameter adjustment and the wireless terminal may also be separate devices, and the power device control system 100 based on real-time perception and operating parameter adjustment can be connected to the wireless terminal through a wired and / or wireless network and transmit interactive information according to a predefined data format.

Claims

1. A power equipment control method based on real-time perception and operating parameter adjustment, characterized in that, Including: Obtaining multiple power operation parameters of the power equipment through the operation perception unit, where the multiple power operation parameters include voltage, current, frequency, and power; Determining operation state adjustment information based on the multiple power operation parameters and operation scenario parameters; Analyzing the multiple power operation parameters to determine the optimal target operation state, including: determining a preset change critical value; determining the optimal target operation state based on the comparison between the multiple power operation parameters and the preset change critical value; Formulating a corresponding adjustment control strategy based on the optimal target operation state and the real-time power operation parameters of the power equipment; Combining multiple control stages, executing the adjustment control strategy and issuing a control instruction to adjust the equipment operation parameters; Among them, determining the preset change critical value includes: obtaining the time queue of the power operation parameter data of the power equipment; obtaining the scenario parameter data of the power equipment; performing fine-grained response encoding and decoding of the operation parameters based on the time series characteristics on the time queue of the power operation parameter data and the scenario parameter data of the power equipment to obtain the preset change critical value.

2. The power equipment control method based on real-time perception and operating parameter adjustment according to claim 1, wherein, Performing fine-grained response encoding and decoding of the operation parameters based on the time series characteristics on the time queue of the power operation parameter data and the scenario parameter data of the power equipment to obtain the preset change critical value, including: Performing time series feature encoding on the time queue of the power operation parameter data to obtain a time series encoding vector of the power operation parameter data; Performing structured embedding encoding on the scenario parameter data of the power equipment to obtain a structured embedding encoding vector of the scenario parameters; Performing fine-grained response fusion processing of the operation parameters on the time series encoding vector of the power operation parameter data and the structured embedding encoding vector of the scenario parameters to obtain a response fusion encoding vector of the power operation parameters; Obtaining the preset change critical value based on the response fusion encoding vector of the power operation parameters.

3. The power equipment control method based on real-time perception and operating parameter adjustment according to claim 2, wherein, Performing time series feature encoding on the time queue of the power operation parameter data to obtain a time series encoding vector of the power operation parameter data, including: performing time series feature encoding based on causal dilated convolution on the time queue of the power operation parameter data to obtain the time series encoding vector of the power operation parameter data.

4. The power equipment control method based on real-time perception and operating parameter adjustment according to claim 2, characterized in that, Performing fine-grained response fusion processing of the operation parameters on the time series encoding vector of the power operation parameter data and the structured embedding encoding vector of the scenario parameters to obtain a response fusion encoding vector of the power operation parameters, including: Performing equal-grained ordering processing on the time series encoding vector of the power operation parameter data and the structured embedding encoding vector of the scenario parameters to obtain a sequence of locally ordered encoding vectors of the time series features of the power operation parameters and a sequence of structured embedding ordered encoding vectors of the scenario parameters; Inputting each group of corresponding locally ordered encoding vectors of the time series features of the power operation parameters and the structured embedding ordered encoding vectors of the scenario parameters in the sequence of locally ordered encoding vectors of the time series features of the power operation parameters and the sequence of structured embedding ordered encoding vectors of the scenario parameters into the transfer response unit to obtain a sequence of response local transfer response encoding matrices of the power operation parameters; Perform transfer response fusion on the sequence of the local transfer response coding matrices for the power operation parameters to obtain the power operation parameter response fusion coding vector.

5. The power equipment control method based on real-time perception and operating parameter adjustment according to claim 4, characterized in that Perform equal-granularity ordering processing on the power operation parameter data time-series coding vector and the scenario parameter structured embedding coding vector to obtain a sequence of power operation parameter time-series local feature ordered coding vectors and a sequence of scenario parameter structured embedding ordered coding vectors, including: Perform ordering arrangement based on the eigenvalue magnitudes on the power operation parameter data time-series coding vector and the scenario parameter structured embedding coding vector to obtain a power operation parameter time-series feature ordered arrangement vector and a scenario parameter structured embedding ordered arrangement vector; Perform equal-granularity feature segmentation on the power operation parameter time-series feature ordered arrangement vector and the scenario parameter structured embedding ordered arrangement vector to obtain the sequence of the power operation parameter time-series local feature ordered coding vectors and the sequence of the scenario parameter structured embedding ordered coding vectors.

6. The power equipment control method based on real-time perception and operating parameter adjustment according to claim 5, characterized in that, Perform transfer response fusion on the sequence of the local transfer response coding matrices for the power operation parameters to obtain the power operation parameter response fusion coding vector, including: Flatten each power operation parameter local transfer response coding matrix in the sequence of the power operation parameter local transfer response coding matrices into a vector to obtain a sequence of power operation parameter local transfer response coding vectors; Perform local transfer response regularization coding based on class Poisson constraints on each power operation parameter local transfer response coding vector in the sequence of the power operation parameter local transfer response coding vectors and the sequence of the power operation parameter local transfer response coding matrices respectively to obtain a sequence of power operation parameter local transfer response constraint coding vectors; Input the sequence of the power operation parameter local transfer response constraint coding vectors into an LSTM-based transfer response fusion unit to obtain the power operation parameter response fusion coding vector.

7. The power equipment control method based on real-time perception and operating parameter adjustment according to claim 6, wherein Based on the power operation parameter response fusion coding vector, obtain the preset change critical value, including: Input the power operation parameter response fusion coding vector into a decoder-based critical value generator to obtain the preset change critical value.

8. The power equipment control method based on real-time perception and operating parameter adjustment according to claim 7, wherein Based on the comparison between the multiple power operation parameters and the preset change critical value, determine the optimal target operation state, including: In response to the multiple power operation parameters reaching the preset change critical value, based on the operation state adjustment information, determine the adjustment time point and generate a to-be-adjusted instruction; Based on the to-be-adjusted instruction and the adjustment constraint conditions, obtain the optimal target operation state; In response to the multiple power operation parameters not reaching the preset change critical value, verify the multiple power operation parameters to determine the optimal target operation state.

9. A power equipment control system based on real-time perception and operating parameter adjustment, characterized in that, Including: A power operation parameter acquisition module, configured to obtain multiple power operation parameters of a power device through an operation sensing unit, where the multiple power operation parameters include voltage, current, frequency, and power; A determination module, configured to determine operation state adjustment information based on the multiple power operation parameters and operation scenario parameters; The optimal target operating state determination module is used to analyze the multiple power operating parameters and determine the optimal target operating state. The optimal target operating state determination module includes: a critical value determination unit for determining a preset change critical value; a comparison unit for determining the optimal target operating state based on the comparison of the multiple power operating parameters and the preset change critical value; A generation module for formulating a corresponding adjustment control strategy based on the optimal target operating state and the real-time power operating parameters of the power equipment; An adjustment module for combining multiple control stages, executing the adjustment control strategy and issuing a control instruction to adjust the equipment operating parameters; Among them, the critical value determination unit is used to: obtain the time queue of the power operating parameter data of the power equipment; obtain the scenario parameter data of the power equipment; perform fine-grained response encoding and decoding of the operating parameters based on the time sequence characteristics on the time queue of the power operating parameter data and the scenario parameter data of the power equipment to obtain the preset change critical value.

Images