An intelligent monitoring system for operation state of distribution transformer based on data acquisition

Abstract

The application discloses an intelligent monitoring system for the operation state of a power distribution transformer based on data collection, and relates to the technical field of power distribution network monitoring.The system comprises a sensor sensing module, a data transmission module, a coupling evaluation module, a decision module and a monitoring center module.The system collects transformer operation state parameters through multiple sensing units, inputs the parameters into the coupling evaluation module after preprocessing, further calculates the independent health index of components, introduces the influence of thermal coupling, electromagnetic coupling and insulation coupling to complete index correction, realizes health index trend deduction and residual life assessment based on an improved exponential smoothing model and dynamic coupling coefficient.The system sets a basic monitoring mode and a coupling monitoring mode, automatically switches according to the health index threshold, optimizes the distribution of computing power while ensuring monitoring accuracy.The decision module completes fault positioning according to the health index and outputs operation and maintenance strategies and power distribution network control schemes, and the monitoring center module realizes data visualization, hierarchical alarm and instruction issuing.

Description

Technical Field

[0001] This invention relates to the field of power distribution network monitoring technology, specifically to an intelligent monitoring system for the operating status of power distribution transformers based on data acquisition. Background Technology

[0002] Distribution transformers are core equipment in power distribution networks, responsible for power conversion and distribution. Their operational status directly affects the reliability of power supply, power quality, and user safety. With the continuous advancement of smart grid construction, online monitoring technology based on multi-source data acquisition has become the mainstream method for transformer condition management. By sensing operational information such as temperature, electrical quantities, oil quality, and partial discharge in real time, it can effectively support equipment condition assessment and fault early warning, which is of significant engineering importance for ensuring the stable operation of the power distribution network.

[0003] Existing data acquisition-based transformer operation status monitoring systems mostly adopt a single-component independent analysis mode, failing to consider the continuous effects of thermal coupling, electromagnetic coupling, and insulation coupling between components, which can easily lead to distorted health assessment results. At the same time, they mostly adopt a fixed monitoring mode, resulting in unreasonable allocation of computing resources, and traditional prediction algorithms do not incorporate coupling effects, leading to insufficient early warning accuracy and short lead time. Therefore, it is necessary to provide a data acquisition-based intelligent monitoring system for the operation status of distribution transformers to solve the above-mentioned problems. Summary of the Invention

[0004] To address the aforementioned technical issues, this paper provides an intelligent monitoring system for the operating status of distribution transformers based on data acquisition. This technical solution solves the problems mentioned in the background section regarding existing transformer condition monitoring systems, such as the lack of consideration for the effects of multi-physics coupling, low accuracy of health assessment, single monitoring mode, waste of computing power, and poor fault early warning accuracy and insufficient early warning lead time due to the lack of integration of coupling mechanisms in traditional prediction.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A data acquisition-based intelligent monitoring system for the operating status of distribution transformers includes a sensor sensing module, a data transmission module, a coupled evaluation module, a decision-making module, and a monitoring center module.

[0007] The sensor sensing module is used to collect the operating status parameters of the transformer;

[0008] The data transmission module is connected to the sensor sensing module and is used to preprocess the operating status parameters and upload them to the coupling evaluation module.

[0009] The coupling evaluation module is connected to the data transmission module and is used to calculate the independent health index of each component based on the operating status parameters. It also introduces the effects of thermal coupling, electromagnetic coupling and insulation coupling between components to obtain the coupling-corrected health index. The overall health index is calculated based on the coupling-corrected health index to perform operating status evaluation, early warning of faults and deduction of health index deterioration trajectory.

[0010] The decision-making module is connected to the coupled evaluation module and is used to locate faults based on health indices and output operation and maintenance strategies and power grid control schemes.

[0011] The monitoring center module is connected to the coupling evaluation module and the decision module respectively, and is used for data visualization, alarm and command issuance.

[0012] In an optional embodiment, the system is configured with two monitoring modes: a basic monitoring mode and a coupled monitoring mode. The mode is automatically switched by triggering a monitoring threshold. The specific switching logic is as follows:

[0013] The system presets overall health index thresholds and independent health index thresholds. When the independent health index of any component is less than or equal to the corresponding component's independent health index threshold, the coupled monitoring mode is activated.

[0014] When the overall health index is greater than the overall health index threshold and the health indices of all components are greater than the corresponding component health index thresholds, the basic monitoring mode is activated.

[0015] The formula for calculating the overall health index is as follows:

[0016]

[0017] In the formula, For the first Component weights The health index after component coupling correction.

[0018] In an optional embodiment, the coupling evaluation module includes a component health calculation unit, a coupling correlation analysis unit, and a trend prediction and inference unit;

[0019] The component health calculation unit is used to calculate the independent health index of the core, winding, cooling system, insulating oil and bushing respectively based on the corresponding sensor data, and to obtain the data set of the health index change pattern of each component and upload it synchronously to the coupling correlation analysis unit and the trend prediction and inference unit.

[0020] The coupling correlation analysis unit is used to identify the coupling influence relationship between the health indices of each component, and to perform coupling correction on the health indices to obtain the coupled corrected health indices.

[0021] The trend prediction and deduction unit is used to combine the change pattern of the health index of a single component with the coupled correlation effects to perform pre-failure trend deduction of the health index of the target component.

[0022] The formula for calculating the independent health index is as follows:

[0023]

[0024] In the formula, For the first Item parameter weights, These are normalization parameters;

[0025] The coupling correction formula for the health index after coupling correction is as follows:

[0026]

[0027] In the formula, For the first Independent health index of each component, For the first The component is related to the first The coupling influence coefficient of each component, The health index is adjusted for coupling.

[0028] The pre-failure trend extrapolation of the target component's health index employs an exponential smoothing prediction model:

[0029]

[0030] In the formula, This is a predicted value for the health index at the next moment. The health index is the coupled and corrected index at the current moment. This is the predicted value of the health index at the current moment. For components The comprehensive coupling influence coefficient at future moments, This is the smoothing coefficient.

[0031] In an optional embodiment, the sensor sensing module includes a temperature sensor unit, an electrical quantity acquisition unit, an oil quality detection unit, a partial discharge sensing unit, and a vibration sensing unit.

[0032] The temperature sensing unit is used to collect the transformer top oil temperature, winding hot spot temperature and ambient temperature.

[0033] The electrical quantity acquisition unit is used to acquire voltage, current, load rate, and unbalance on the high-voltage and low-voltage sides.

[0034] The oil quality detection unit is used to collect the concentration of dissolved gases, trace water content, and dielectric loss parameters of insulating oil in the oil.

[0035] The partial discharge sensing unit and the vibration sensing unit are used to monitor the internal discharge signal and the mechanical vibration signal of the transformer body, respectively.

[0036] The load factor is calculated using the formula: η = I / I N ×100%, where I is the measured current, I N This is the rated current.

[0037] In an optional embodiment, the data transmission module supports wired and wireless modes and has functions such as data interruption resumption, encrypted transmission, and protocol adaptation.

[0038] In an optional embodiment, the health index value range of the independent health index is positively correlated with the health status of the component. The closer the value is to 1, the better the health status of the component and the stronger its support for the stability of the power distribution network. The closer the value is to 0, the worse the health status of the component and the greater the potential risk to the power distribution network.

[0039] In an optional embodiment, the trend prediction and extrapolation unit combines historical operating data, load change patterns, and ambient temperature data when performing health index trend extrapolation.

[0040] In an optional embodiment, the coupling evaluation module is further configured to generate an assessment value for the remaining service life of the transformer core components based on the trend prediction and extrapolation results. The remaining service life is calculated using a linear extrapolation formula.

[0041]

[0042] In the formula, The fault threshold, The actual health index at the current moment. The rate of deterioration of the health index.

[0043] In an optional embodiment, the decision module has a built-in fault case library, operation and maintenance strategy library and distribution network control scheme library. When the independent health index or the coupled modified health index of any component reaches the preset warning threshold or fault threshold, it automatically distinguishes between primary faults and secondary effects based on the coupling relationship, outputs accurate fault location results, corresponding maintenance schemes and distribution network control instructions, and stores and archives all evaluation results, warning information and decision schemes.

[0044] Furthermore, a data acquisition-based intelligent monitoring method for the operating status of distribution transformers is proposed, applicable to any of the monitoring systems described above, comprising the following steps:

[0045] S1. Collect multi-dimensional operating status parameters of the distribution transformer;

[0046] S2. The collected raw operating status parameters are filtered, denoised, and standardized to obtain preprocessed data;

[0047] S3. Based on the preprocessed data, calculate the independent health index of each core component of the transformer core, windings, cooling system, insulating oil and bushings respectively.

[0048] S4. Automatically switch monitoring modes according to preset independent health index thresholds: When the independent health index of any component is less than or equal to the corresponding threshold, the coupled monitoring mode is activated.

[0049] When the independent health index of all components is greater than the corresponding threshold, the basic monitoring mode is activated.

[0050] S5. In the coupled monitoring mode, identify the influence relationship between thermal coupling, electromagnetic coupling and insulation coupling between components, perform coupling correction on the independent health index to generate the coupled corrected health index, and obtain the overall health index of the transformer based on the coupled corrected health index of each component.

[0051] S6. An exponential smoothing prediction model incorporating dynamic coupling coefficients is adopted. By combining historical operating data, load change patterns and ambient temperature, the pre-fault trend of the health index after coupling correction is extrapolated, and the health index deterioration trajectory is generated.

[0052] S7. Calculate the remaining service life of the transformer's core components based on the health index deterioration trajectory and linear extrapolation formula.

[0053] S8. Based on the independent health index, the coupled modified health index and the coupling correlation, locate the root cause of the fault and distinguish between the primary fault and the secondary impact, match the corresponding operation and maintenance handling strategy, and generate the distribution network control scheme.

[0054] S9. Real-time visualization of operating status parameters, health index of each component, deterioration trajectory, remaining service life, fault location results and control scheme, and trigger graded audible and visual alarms according to the health index level, while storing and archiving all data and processing results.

[0055] Compared with the prior art, the beneficial effects of the present invention are:

[0056] This solution proposes an intelligent monitoring system for the operating status of distribution transformers based on data acquisition. By introducing the effects of thermal coupling, electromagnetic coupling, and insulation coupling to correct the health index, it achieves an accurate reflection of the actual operating status under the interaction of multiple components, solving the problem of evaluation distortion caused by single analysis. By incorporating the dynamic coupling coefficient into the improved exponential smoothing model for trend extrapolation, it realizes early warning of faults and accurate calculation of remaining life, improving the accuracy of early warning and the lead time for prediction. Attached Figure Description

[0057] Figure 1 This is a system framework diagram of an intelligent monitoring system for the operating status of distribution transformers based on data acquisition, as proposed in this invention.

[0058] Figure 2 This is an overall flowchart of an intelligent monitoring method for the operating status of distribution transformers based on data acquisition, as proposed in this invention. Detailed Implementation

[0059] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0060] Reference Figures 1-2 As shown, an intelligent monitoring system for the operating status of distribution transformers based on data acquisition includes:

[0061] Sensor modules are deployed in the core, windings, bushings, cooling system, and insulating oil area of ​​the distribution transformer. By deploying corresponding sensing units, the transformer's operating status parameters are collected in real time. The collected parameters include oil temperature, winding temperature, load electrical quantities, characteristic gases in the oil, partial discharge signals, and vibration signals.

[0062] The sensor sensing module specifically includes a temperature sensing unit, an electrical quantity acquisition unit, an oil quality detection unit, a partial discharge sensing unit, and a vibration sensing unit. The temperature sensing unit collects the transformer top oil temperature, winding hot spot temperature, and ambient temperature, providing a temperature reference for power distribution network load regulation. The electrical quantity acquisition unit collects voltage, current, load factor, and unbalance on both the high-voltage and low-voltage sides, where the load factor is calculated using the formula η=I / I0. N The result is obtained by multiplying by 100%, where I is the measured current. N The rated current is used; the oil quality detection unit is used to collect the dissolved gas concentration, trace water content and dielectric loss parameters of the insulating oil; the partial discharge sensing unit and the vibration sensing unit are used to monitor the internal discharge signal and the mechanical vibration signal of the transformer body, respectively, to provide early warning of transformer faults and avoid affecting the reliability of power supply in the distribution network.

[0063] The data transmission module establishes a data connection with the sensor sensing module, filters, reduces noise, and standardizes the collected raw state parameters, and uploads the processed state data to the backend analysis unit in real time. The data transmission module supports both wired and wireless transmission modes, and features data interruption resumption, encrypted transmission, and protocol adaptation capabilities. This ensures the integrity, security, and real-time performance of state data and power distribution network control commands, enabling data interoperability with the power distribution network dispatching system.

[0064] The coupling assessment module establishes a data connection with the data transmission module, calculates the health index of each component of the transformer based on the received status data, and introduces the effects of thermal coupling, electromagnetic coupling and insulation coupling between components when predicting and extrapolating the trend of the health index of each component. This enables the assessment of transformer operating status, early warning of faults and extrapolation of the trajectory of health index deterioration, and provides data basis for load adjustment and voltage regulation of the distribution network.

[0065] The coupled assessment module specifically includes a component health calculation unit, a coupled correlation analysis unit, and a trend prediction and deduction unit. The component health calculation unit calculates the independent health index of the core, windings, cooling system, insulating oil, and bushing based on the corresponding sensor data. The independent health index is obtained by normalizing and weighting the monitoring data; the calculation formula is as follows: ,in For the first Item parameter weights, As a normalized parameter, the health index ranges from 0 to 1, and the value is positively correlated with the component's health level. The coupling correlation analysis unit identifies the coupling effects between the health indices of various components, quantifies the transmission effect of different component degradation on the target component's health status, and corrects this through a coupling correction formula. The independent health index was adjusted, among which For the first Independent health index of each component, For the first The component is related to the first The coupling influence coefficient of each component, This is the health index after coupling correction.

[0066] In this embodiment of the invention, the component health calculation unit is used to calculate the independent health indices of the core, winding, cooling system, insulating oil, and bushing based on the corresponding sensor data collected by the sensor sensing module. The specific calculation process is as follows: First, the monitoring parameters of each component collected by the sensor sensing module, such as the winding temperature, load current, partial discharge intensity, characteristic gas concentration and trace water content of the insulating oil, and the heat dissipation temperature difference of the cooling system, are processed. Then, the monitoring parameters are normalized to eliminate the dimensional differences between different parameters. The normalization calculation formula is as follows: ,That This is the raw monitoring data. , These are the lower and upper limits of the parameter, respectively, and the normalized data. The value range is [0,1]; then the weights corresponding to each monitoring parameter are determined. Weight The analytic hierarchy process (AHP) combined with expert scoring is used to determine the relative importance of each parameter. First, a judgment matrix is ​​constructed to assess the relative importance of each parameter. Then, the largest eigenvalue and corresponding eigenvector of the judgment matrix are calculated, and a consistency test is performed (calculating the consistency index). With consistency ratio ,in The largest eigenvalue, Number of parameters As the average random consistency index, when (After determining the appropriateness of the weight allocation), initial weights are obtained. These initial weights are then corrected based on fault sensitivity analysis. Finally, the weights are normalized to ensure the sum of all parameter weights equals 1. The magnitude of the weight represents the degree of influence of the corresponding monitored parameter on the component's health status; the higher the degree of influence, the greater the weight. Then, a weighted scoring formula is used. The independent health index of each component was calculated. The independent health index ranges from 0 to 1, and its value is positively correlated with the health level of the component. The closer the value is to 1, the better the component's health and the stronger its support for the stability of the power distribution network. Conversely, the closer the value is to 0, the worse the component's health and the greater the potential risk to the power distribution network. After obtaining the independent health index of each component, the coupling correlation analysis unit in the coupling assessment module introduces the effects of persistent thermal coupling, electromagnetic coupling, and insulation coupling between components. The coupling influence coefficient is first calculated according to the coupling type. The thermal coupling coefficient is calculated based on the difference between the oil temperature and the rated oil temperature; the electromagnetic coupling coefficient is calculated based on the current imbalance; and the insulation coupling coefficient is calculated based on the concentration of characteristic gases in the oil. These are then calculated using the formula... ( The total coupling influence coefficient is obtained by fusing the weights of thermal, electromagnetic, and insulating couplings (respectively, respectively), constructing a coupling coefficient matrix, and then applying the formula... The independent health index is coupled and corrected to obtain the coupled and corrected health index that reflects the actual operating state of the component. .

[0067] Specifically, the The component is related to the first Coupling influence coefficient of individual components The range of values ​​is , This indicates no impact. The smaller the value, the greater the drag on other components caused by component degradation. The coupling effect coefficient is calculated separately according to the coupling type; for thermal coupling coefficient... Taking the effect of the cooling system (j) on the winding (i) as an example , To measure the oil temperature, Rated oil temperature This refers to the maximum permissible limit for the top layer oil temperature of the transformer. This is the thermal coupling weighting coefficient, with a value ranging from 0 to 1. For the electromagnetic coupling coefficient... Taking the influence of the iron core (j) on the winding (i) as an example , For current imbalance, The upper limit threshold for current imbalance specified by the procedure / manufacturer. This represents the electromagnetic coupling weight. For the insulation coupling coefficient... Taking the effect of insulating oil (j) on bushing (i) as an example , Characteristic gas concentration, For insulating coupling weights, This refers to the maximum permissible concentration specified by national standards / transformer manufacturers.

[0068] The trend prediction and extrapolation unit combines the changing patterns of the health index of a single component with its coupled influences to perform pre-fault trend extrapolation of the target component's health index. During the extrapolation process, historical operating data, load variation patterns, and the impact of ambient temperature are simultaneously incorporated to achieve high-precision real-time prediction. This allows for early assessment of the transformer's condition and its impact on the distribution network, providing advance notice for distribution network scheduling. Simultaneously, based on the trend prediction and extrapolation results, the remaining life calculation unit employs a linear extrapolation formula. Generate the remaining service life assessment values ​​for the core components of the transformer, among which The fault threshold, The actual health index at the current moment. This provides data support for planned maintenance and load optimization of the distribution network, thereby further improving the reliability of power supply. Furthermore, predicting trends and anticipating faults in the health indices of various transformer components is a crucial function of the intelligent monitoring system for the operating status of distribution transformers, enabling an upgrade from "post-event diagnosis" to "pre-event early warning."

[0069] Specifically, the health index of a single component can only reflect its own degradation trend and cannot characterize the thermal, electromagnetic, and insulation coupling effects between multiple components within a transformer. For example, an abnormal cooling system health index only indicates a fault in the heat dissipation process, but cannot distinguish whether the fault stems from a decrease in heat dissipation capacity or from abnormal heating of other components. Similarly, an abnormal winding or insulation system health index cannot rule out a cascading degradation caused by cooling system failure. Therefore, a joint assessment mechanism for the health indices of multiple components must be established, and reasoning through coupling relationships is necessary to accurately locate the root cause of the fault.

[0070] In this embodiment of the invention, the trend prediction and extrapolation unit introduces coupling effects to perform exponential smoothing trend prediction. The principle is as follows: the coupling effects between various components of the transformer are continuous. Traditional exponential smoothing prediction only relies on the historical data of the components themselves. Ignoring the coupling effects can easily lead to overly optimistic prediction results. This system uses the comprehensive coupling effect coefficient at future moments to... The improved prediction formula, dynamically incorporating an exponential smoothing model, is as follows: ,in This is a predicted value for the health index at the next moment. The health index after coupling correction at the current moment , For smoothing coefficients, For components The comprehensive coupling impact coefficient at future moments is determined by load forecasting, temperature trends, and insulation aging patterns. The specific prediction process involves first extracting the coupling-corrected health index sequence from historical data at multiple moments to predict the future coupling coefficient change trend. Then, the improved formula is used for further prediction. Subsequently, the health index degradation trajectory over a future period is iteratively derived moment by moment. Finally, the predicted health index sequence is compared with a preset fault threshold. In comparison, through linear extrapolation formula (in The actual health index at the current moment. The health index deterioration rate is used to calculate the time it takes for a component's health index to reach a fault threshold, enabling transformer operating status assessment, early fault warning, and health index deterioration trajectory prediction. This provides accurate data support for distribution network regulation, load optimization, and planned maintenance, ensuring stable distribution network operation. The health index deterioration rate is calculated from the slope of the historical health index sequence, i.e. .

[0071] The decision-making module establishes a data connection with the coupled evaluation module, locates the root cause of the fault based on the health index thresholds of each component and the coupling relationship, matches the corresponding operation and maintenance handling strategies and distribution network control schemes, and stores and archives the evaluation results, early warning information, decision schemes and power grid control instructions to support the stable operation of the distribution network.

[0072] The monitoring center module establishes data connections with both the coupled evaluation module and the decision-making module, providing real-time visual displays of transformer operating status, monitoring data curves, health index trends of various components, and fault matching results. It outputs distribution network control commands and features tiered audible and visual alarms, data querying, parameter configuration, and control command issuance, enabling coordinated operation with the distribution network dispatching system. The monitoring center module supports PC webpage display, mobile visualization, and integration with the distribution network dispatching system. Alarm information can be simultaneously issued via system pop-ups, SMS, and APP push notifications. It also supports manual input and modification of threshold parameters and control strategy parameters.

[0073] The system features two monitoring modes: a basic monitoring mode and a coupled monitoring mode, automatically switching between them based on monitoring thresholds. Preset thresholds for the overall health index and individual health indices are provided. When the individual health index of any transformer component is less than or equal to its corresponding component health index threshold, the coupled evaluation module automatically activates to perform monitoring and simulation, simultaneously outputting auxiliary commands for power grid regulation. Conversely, when the overall health index exceeds the overall health threshold and the health indices of all components are greater than their corresponding component health thresholds, the coupled evaluation module is not activated, and only the basic monitoring mode is activated. The overall health index is determined by the formula... The calculation yielded, where For the first Component weights This is a component coupling-corrected health index used to reflect the degree of impact of transformer operating status on the power distribution network.

[0074] Understandably, when the system is in basic monitoring mode, automatic switching is triggered only by the independent health index. The specific switching logic is as follows: a preset independent health index threshold is set. When the independent health index of any component is less than or equal to the threshold, the coupled monitoring mode is activated; when the independent health indices of all components are greater than the threshold, the basic monitoring mode is activated. The overall health index is calculated only in the coupled monitoring mode.

[0075] Furthermore, the decision-making module has a built-in fault case library, operation and maintenance strategy library, and distribution network control scheme library. When the independent health index or coupled correction health index of any component reaches the preset warning threshold or fault threshold, it automatically distinguishes between primary faults and secondary effects based on the coupling relationship, outputs accurate fault location results, corresponding maintenance plans and distribution network control instructions, and stores and archives all evaluation results, warning information and decision-making schemes.

[0076] In this embodiment, the fault case library stores historical transformer fault data. Each case includes the faulty component, fault type, pre-fault health index change curve, coupling coefficient change characteristics, fault cause, and handling result. Faults are categorized into three levels based on severity: general faults, important faults, and major faults. The fault case library uses an incremental update method; after each fault handling is completed, the fault data and handling effect are automatically entered into the library. Clustering algorithms are used periodically to optimize the case feature matching accuracy. The operation and maintenance strategy library corresponds one-to-one with the fault case library, storing standardized operation and maintenance processes corresponding to different fault types, including maintenance priorities, maintenance personnel qualification requirements, required tools and spare parts, maintenance steps, and acceptance standards. For secondary faults caused by coupling relationships, the operation and maintenance strategy library prioritizes handling the primary fault first, then investigating the secondary impacts. The distribution network control scheme library stores power grid control measures corresponding to different health status levels, including load transfer schemes, voltage regulation ranges, reactive power compensation capacity, power supply mode switching strategies, and temporary power supply plans. Control schemes are categorized into three levels based on their impact range: single transformer control, distribution area control, and feeder control, prioritizing the scheme with the least impact on user power supply.

[0077] Understandably, when the independent health index or the health index after coupling correction of any component reaches the preset warning threshold or fault threshold, the decision module calls the coupling coefficient matrix generated by the coupling correlation analysis unit to trace the fault source: First, all components with health indices below the threshold are screened out to form a fault candidate set; the sum of the coupling influence coefficients of each candidate component on other faulty components is calculated, and the component with the largest sum of coupling influence coefficients and the first independent health index to decrease is determined as the primary faulty component; the remaining components with health indices below the threshold and coupling coefficients with the primary faulty component greater than 0.5 are determined as secondary deteriorated components affected by coupling; for multiple components with no obvious coupling correlation that are abnormal at the same time, they are determined to be multiple independent faults occurring concurrently, and corresponding operation and maintenance strategies are matched respectively. The decision-making module outputs differentiated decision results based on the health index level and fault type: In the early warning state (0.6 < health index ≤ 0.8), it outputs early warning information, component degradation trend analysis, and preventative maintenance suggestions, without triggering grid control; in the abnormal state (0.4 < health index ≤ 0.6), it outputs accurate fault location results, emergency repair plans, and light load control suggestions (such as reducing the load factor to below 80%); in the fault state (health index ≤ 0.4), it outputs the highest-level alarm, an immediate power outage repair plan, and a load transfer plan, simultaneously pushing control instructions to the distribution network dispatching system to prevent the fault from escalating into a large-scale power outage. The output distribution network control instructions include the target area for load transfer, the target voltage regulation value, the reactive power compensation switching capacity, and the execution time window. All instructions are auxiliary decision suggestions and must be confirmed by dispatch personnel before execution.

[0078] Specifically, the decision-making module automatically stores and archives all operational data and processing results. The archived content includes: raw sensor data, preprocessed data, independent health indices of each component and coupled-corrected health indices, coupling coefficient matrix, trend prediction results, remaining service life assessment values, fault location results, operation and maintenance strategies, control schemes, and alarm records. The archived data is stored encrypted and retained for at least the transformer's design service life. It supports multi-dimensional querying and exporting by time, equipment number, fault type, and other dimensions, providing data support for equipment lifecycle management and subsequent fault analysis.

[0079] Reference Figure 2 As shown, a smart monitoring method for the operating status of distribution transformers based on data acquisition is proposed and applied to the monitoring system described above. The method includes the following steps:

[0080] S1. Collect multi-dimensional operating status parameters of the distribution transformer;

[0081] S2. The collected raw operating status parameters are filtered, denoised, and standardized to obtain preprocessed data;

[0082] S3. Based on the preprocessed data, calculate the independent health index of each core component of the transformer core, windings, cooling system, insulating oil and bushings respectively.

[0083] S4. Automatically switch monitoring modes according to preset independent health index thresholds: When the independent health index of any component is less than or equal to the corresponding threshold, the coupled monitoring mode is activated.

[0084] When the independent health index of all components is greater than the corresponding threshold, the basic monitoring mode is activated.

[0085] S5. In the coupled monitoring mode, identify the influence relationship between thermal coupling, electromagnetic coupling and insulation coupling between components, perform coupling correction on the independent health index to generate the coupled corrected health index, and obtain the overall health index of the transformer based on the coupled corrected health index of each component.

[0086] S6. An exponential smoothing prediction model incorporating dynamic coupling coefficients is adopted. By combining historical operating data, load change patterns and ambient temperature, the pre-fault trend of the health index after coupling correction is extrapolated, and the health index deterioration trajectory is generated.

[0087] S7. Calculate the remaining service life of the transformer's core components based on the health index deterioration trajectory and linear extrapolation formula.

[0088] S8. Based on the independent health index, the coupled modified health index and the coupling correlation, locate the root cause of the fault and distinguish between the primary fault and the secondary impact, match the corresponding operation and maintenance handling strategy, and generate the distribution network control scheme.

[0089] S9. Real-time visualization of operating status parameters, health index of each component, deterioration trajectory, remaining service life, fault location results and control scheme, and trigger graded audible and visual alarms according to the health index level, while storing and archiving all data and processing results.

[0090] The advantages of this invention are as follows: Starting from the actual operating mechanism of distribution transformers and the management and control requirements of smart distribution networks, it possesses significant technical advantages and practical value. The system fully considers the continuous thermal, electromagnetic, and insulation coupling effects between internal transformer components. Through multi-source data acquisition and standardized processing, it achieves comprehensive perception of key components such as the core, windings, cooling system, insulating oil, and bushings. It quantifies independent health indices using the analytic hierarchy process (AHP) and weighted scoring, and obtains health indicators that accurately reflect the equipment status through coupling coefficient correction, effectively avoiding assessment biases caused by single-component analysis. The system innovatively sets up a dual-mode system of basic monitoring and coupled monitoring, automatically switching operating modes based on health index thresholds. While ensuring high-precision analysis of abnormal states, it rationally allocates computing resources, improving overall operating efficiency. Combined with an improved index smoothing model incorporating dynamic coupling coefficients, it can accurately predict the trajectory of health index deterioration, achieving early fault warning and remaining life assessment, significantly improving warning accuracy and prediction lead time. Meanwhile, the system combines transformer status monitoring with distribution network control, and can output operation and maintenance strategies and power grid dispatching schemes to achieve coordinated protection of equipment safety and stable power grid operation, thereby improving the intelligence level and engineering applicability of distribution transformer monitoring.

[0091] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A smart monitoring system for the operating status of distribution transformers based on data acquisition, characterized in that, It includes a sensor perception module, a data transmission module, a coupled evaluation module, a decision-making module, and a monitoring center module: The sensor sensing module is used to collect the operating status parameters of the transformer; The data transmission module is connected to the sensor sensing module and is used to preprocess the operating status parameters and upload them to the coupling evaluation module. The coupling evaluation module is connected to the data transmission module and is used to calculate the independent health index of each component based on the operating status parameters. It also introduces the effects of thermal coupling, electromagnetic coupling and insulation coupling between components to obtain the coupling-corrected health index. The overall health index is calculated based on the coupling-corrected health index to perform operating status evaluation, early warning of faults and deduction of health index deterioration trajectory. The decision-making module is connected to the coupled evaluation module and is used to locate faults based on health indices and output operation and maintenance strategies and power grid control schemes. The monitoring center module is connected to the coupling evaluation module and the decision module respectively, and is used for data visualization, alarm and command issuance.

2. The intelligent monitoring system for the operating status of distribution transformers based on data acquisition according to claim 1, characterized in that: The system is configured with two monitoring modes: a basic monitoring mode and a coupled monitoring mode. The mode automatically switches based on a monitoring threshold. The specific switching logic is as follows: The system presets overall health index thresholds and independent health index thresholds. When the independent health index of any component is less than or equal to the corresponding component's independent health index threshold, the coupled monitoring mode is activated. When the overall health index is greater than the overall health index threshold and the health indices of all components are greater than the corresponding component health index thresholds, the basic monitoring mode is activated. The formula for calculating the overall health index is as follows: In the formula, For the first Component weights The health index after component coupling correction.

3. The intelligent monitoring system for the operating status of distribution transformers based on data acquisition according to claim 1, characterized in that: The coupling assessment module includes a component health calculation unit, a coupling correlation analysis unit, and a trend prediction and inference unit. The component health calculation unit is used to calculate the independent health index of the core, winding, cooling system, insulating oil and bushing respectively based on the corresponding sensor data, and to obtain the data set of the health index change pattern of each component and upload it synchronously to the coupling correlation analysis unit and the trend prediction and inference unit. The coupling correlation analysis unit is used to identify the coupling influence relationship between the health indices of each component, and to perform coupling correction on the health indices to obtain the coupled corrected health indices. The trend prediction and deduction unit is used to combine the change pattern of the health index of a single component with the coupled correlation effects to perform pre-failure trend deduction of the health index of the target component. The formula for calculating the independent health index is as follows: In the formula, For the first Item parameter weights, These are normalization parameters; The coupling correction formula for the health index after coupling correction is as follows: In the formula, For the first Independent health index of each component, For the first The component is related to the first The coupling influence coefficient of each component, The health index is adjusted for coupling. The pre-failure trend extrapolation of the target component's health index employs an exponential smoothing prediction model: In the formula, This is a predicted value for the health index at the next moment. The health index is the coupled and corrected index at the current moment. This is the predicted value of the health index at the current moment. For components The comprehensive coupling influence coefficient at future moments, This is the smoothing coefficient.

4. The intelligent monitoring system for the operating status of distribution transformers based on data acquisition according to claim 1, characterized in that: The sensor sensing module includes a temperature sensor unit, an electrical quantity acquisition unit, an oil quality detection unit, a partial discharge sensing unit, and a vibration sensing unit. The temperature sensing unit is used to collect the transformer top oil temperature, winding hot spot temperature and ambient temperature. The electrical quantity acquisition unit is used to acquire voltage, current, load rate, and unbalance on the high-voltage and low-voltage sides. The oil quality detection unit is used to collect the concentration of dissolved gases, trace water content, and dielectric loss parameters of insulating oil in the oil. The partial discharge sensing unit and the vibration sensing unit are used to monitor the internal discharge signal and the mechanical vibration signal of the transformer body, respectively. The load factor is calculated using the formula: η = I / I N ×100%, where I is the measured current, I N This is the rated current.

5. The intelligent monitoring system for the operating status of distribution transformers based on data acquisition according to claim 1, characterized in that: The data transmission module supports both wired and wireless modes and features data interruption resume, encrypted transmission, and protocol adaptation.

6. The intelligent monitoring system for the operating status of distribution transformers based on data acquisition according to claim 3, characterized in that: The independent health index range is positively correlated with the health status of the component. The closer the value is to 1, the better the health status of the component and the stronger its support for the stability of the power distribution network. The closer the value is to 0, the worse the health status of the component and the greater the potential risk to the power distribution network.

7. The intelligent monitoring system for the operating status of distribution transformers based on data acquisition according to claim 3, characterized in that: When performing health index trend projection, the trend prediction and projection unit combines historical operating data, load change patterns, and ambient temperature data.

8. The intelligent monitoring system for the operating status of distribution transformers based on data acquisition according to claim 3, characterized in that: The coupling evaluation module is also used to generate an assessment value of the remaining service life of the transformer core components based on the trend prediction and extrapolation results. The remaining service life is calculated using a linear extrapolation formula. In the formula, The fault threshold, The actual health index at the current moment. The rate of deterioration of the health index.

9. The intelligent monitoring system for the operating status of distribution transformers based on data acquisition according to claim 1, characterized in that: The decision-making module has a built-in fault case library, operation and maintenance strategy library, and distribution network control scheme library. When the independent health index or the coupled modified health index of any component reaches the preset warning threshold or fault threshold, it automatically distinguishes between primary faults and secondary effects based on the coupling relationship, outputs accurate fault location results, corresponding maintenance schemes and distribution network control instructions, and stores and archives all evaluation results, warning information and decision-making schemes.

10. A method for intelligent monitoring of the operating status of distribution transformers based on data acquisition, applied to the monitoring system described in any one of claims 1-9, characterized in that, Includes the following steps: S1. Collect multi-dimensional operating status parameters of the distribution transformer; S2. The collected raw operating status parameters are filtered, denoised, and standardized to obtain preprocessed data; S3. Based on the preprocessed data, calculate the independent health index of each core component of the transformer core, windings, cooling system, insulating oil and bushings respectively. S4. Automatically switch monitoring modes according to preset independent health index thresholds: When the independent health index of any component is less than or equal to the corresponding threshold, the coupled monitoring mode is activated. When the independent health index of all components is greater than the corresponding threshold, the basic monitoring mode is activated. S5. In the coupled monitoring mode, identify the influence relationship between thermal coupling, electromagnetic coupling and insulation coupling between components, perform coupling correction on the independent health index to generate the coupled corrected health index, and obtain the overall health index of the transformer based on the coupled corrected health index of each component. S6. An exponential smoothing prediction model incorporating dynamic coupling coefficients is adopted. By combining historical operating data, load change patterns and ambient temperature, the pre-fault trend of the health index after coupling correction is extrapolated, and the health index deterioration trajectory is generated. S7. Calculate the remaining service life of the transformer's core components based on the health index deterioration trajectory and linear extrapolation formula. S8. Based on the independent health index, the coupled modified health index and the coupling correlation, locate the root cause of the fault and distinguish between the primary fault and the secondary impact, match the corresponding operation and maintenance handling strategy, and generate the distribution network control scheme. S9. Real-time visualization of operating status parameters, health index of each component, deterioration trajectory, remaining service life, fault location results and control scheme, and trigger graded audible and visual alarms according to the health index level, while storing and archiving all data and processing results.

Images