A transformer load regulating method for intelligent power distribution system

Abstract

The application relates to the technical field of power distribution network operation control, and discloses a transformer load adjusting method for an intelligent power distribution system, which comprises the following steps: collecting power distribution transformer operation parameters and power distribution system load parameters, fusing the two types of parameters to obtain a transformer comprehensive state parameter set and a power distribution system load characteristic parameter set; performing multi-dimensional comprehensive evaluation on the current load state of the power distribution transformer according to the two parameter sets, and outputting a load state evaluation result; performing double-parameter trend prediction in combination with a real-time load rate and a total load change rate, judging voltage regulation demand, and generating a transformer operation state adjusting instruction; completing phase-by-phase reactive power shortage quantitative analysis by using phase-by-phase load unbalance degree to generate a phase-by-phase reactive power compensation adjusting instruction; and executing the two types of instructions according to a preset time sequence to respectively control the on-load tap changer and the reactive power compensation device to complete the adjustment; and the application can improve the efficiency of transformer load adjustment for the intelligent power distribution system.

Description

Technical Field

[0001] This invention relates to the field of power distribution network operation control technology, and in particular to a transformer load regulation method for intelligent power distribution systems. Background Technology

[0002] Currently, load regulation of transformers in power distribution systems mainly relies on a single electrical parameter (such as voltage or current) as the basis for control, resulting in relatively isolated regulation strategies. For example, traditional methods typically drive on-load tap changers based solely on the voltage deviation on the low-voltage side of the transformer, or switch compensation capacitors based solely on reactive power deficits. These methods fail to effectively integrate the transformer's own operating health parameters, such as winding temperature and insulation aging status, with the load fluctuation characteristics of the power distribution system. Consequently, regulation decisions lack a comprehensive consideration of equipment status and system operating conditions, limiting the effectiveness of regulation.

[0003] For example, when a power distribution system experiences a short-term load surge, traditional regulation methods often lag behind due to a lack of prediction of the load change rate trend. When the three-phase load is severely unbalanced, existing technologies typically only provide coarse regulation through overall reactive power compensation, failing to implement refined phase-by-phase compensation for individual phase deficiencies. Furthermore, the lack of timing coordination between the on-load tap changer's position adjustment and the switching action of the reactive power compensation device easily leads to mutual interference between voltage regulation and reactive power compensation, affecting the stability and efficiency of regulation.

[0004] Therefore, existing distribution transformer load regulation technologies are significantly lacking in terms of the comprehensiveness of parameter utilization, the multi-dimensionality of load status assessment, the accuracy of regulation command generation, and the synergy between the two types of regulation methods, making it difficult to meet the actual needs of intelligent distribution systems for efficient, timely, and precise transformer load regulation. Summary of the Invention

[0005] To address the aforementioned technical shortcomings, the present invention aims to propose a transformer load regulation method for intelligent power distribution systems. This method addresses the technical problems in the prior art, which lacks effective integration and coordinated control of transformer operating status and power distribution system load characteristics, resulting in incomplete load regulation decisions, delayed regulation actions, and insufficient precision, making it difficult to simultaneously consider equipment health status and the three-phase balance requirements of the system.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a transformer load regulation method for intelligent power distribution systems. The transformer load regulation method for an intelligent power distribution system includes: S1. Collect the operating parameters of the distribution transformer and the load parameters of the distribution system; S2. Perform parameter fusion on the operating parameters and the load parameters to obtain the comprehensive state parameter set of the distribution transformer and the load characteristic parameter set of the distribution system; S3. Based on the comprehensive state parameter set and the load characteristic parameter set, perform a multi-dimensional comprehensive evaluation of the current load state of the distribution transformer and generate a load state evaluation result; S4. Based on the real-time load rate in the comprehensive state parameter set and the total load change rate in the load characteristic parameter set, perform dual-parameter trend prediction on the load state assessment result, generate voltage regulation demand determination result, and generate the operation state adjustment command of the distribution transformer based on the voltage regulation demand determination result. S5. Based on the load imbalance of each phase in the load characteristic parameter set, perform phase-by-phase reactive power deficit quantitative analysis on the load state assessment results, and generate corresponding reactive power compensation adjustment commands. S6. According to the preset execution sequence, execute the operating status adjustment command and the reactive power compensation adjustment command to control the on-load tap changer and reactive power compensation device of the distribution transformer to perform corresponding adjustment actions respectively.

[0007] Preferably, the operating parameters of the distribution transformer and the load parameters of the distribution system are collected, including: The three-phase voltage, three-phase current, winding temperature, and oil temperature of the distribution transformer are collected by voltage sensors, current sensors, and temperature sensors installed on the distribution transformer. The three-phase voltage, the three-phase current, the winding temperature, and the oil temperature are converted from analog to digital to generate a transformer digital signal set. By installing load monitoring terminals on the power distribution lines, the load current and load voltage of each feeder circuit of the power distribution system are collected, and the load current and load voltage are converted from analog to digital to generate a load digital signal set. The transformer digital signal set and the load digital signal set are normalized to generate the operating parameters and the load parameters.

[0008] Preferably, the operating parameters and the load parameters are fused to obtain a comprehensive state parameter set for the distribution transformer and a load characteristic parameter set for the distribution system, including: The effective values ​​of the high-voltage side line voltage, low-voltage side line voltage, high-voltage side line current, low-voltage side line current, and winding temperature state value of the distribution transformer are extracted from the operating parameters and used as the state characteristic components of the distribution transformer. The effective values ​​of phase current and phase voltage of each feeder circuit of the power distribution system are extracted from the load parameters and used as the load characteristic components of the power distribution system. The state feature components and the load feature components are timestamped and then normalized to obtain state normalized features and load normalized features. By using preset electrical correlation weights, the state normalization features and the load normalization features are weighted and combined to generate the comprehensive state parameter set; The effective values ​​of phase current and phase voltage of each feeder in the load normalization characteristics are grouped and integrated according to the feeder number to generate the load characteristic parameter set.

[0009] Preferably, the state normalization features and the load normalization features are weighted and combined using preset electrical association weights to generate the comprehensive state parameter set, including: The electrical association weight is read from a preset weight configuration table. The electrical association weight includes a first weight coefficient corresponding to the state feature component and a second weight coefficient corresponding to the load feature component. The state feature component is multiplied by the first weighting coefficient to obtain the weighted state component. The load characteristic component is multiplied by the second weighting coefficient to obtain the weighted load component; Based on the electrical structure hierarchy of the distribution transformer, the weighted state components and the weighted load components are accumulated and combined to generate the comprehensive state parameter set.

[0010] Preferably, based on the comprehensive state parameter set and the load characteristic parameter set, a multi-dimensional comprehensive evaluation of the current load state of the distribution transformer is performed to generate a load state evaluation result, including: Feature extraction is performed on the comprehensive state parameter set to obtain the real-time load rate, winding temperature state value, and insulation aging characteristic value of the distribution transformer; Feature extraction is performed on the load characteristic parameter set to obtain the total load power fluctuation characteristic value and the load current distribution characteristic value of each phase of the power distribution system; The real-time load rate is compared with a preset standard load rate range to generate a load rate deviation index for the distribution transformer. By combining the winding temperature state value and the insulation aging characteristic value, the thermal state and insulation state of the distribution transformer are analyzed in a coordinated manner to generate the health state coefficient of the distribution transformer. By using the total load power fluctuation characteristic value and the load current distribution characteristic value of each phase, the load fluctuation characteristics and three-phase load distribution characteristics of the power distribution system are correlated and analyzed to generate the load adaptability index of the power distribution system. The load rate deviation index, the health status coefficient, and the load adaptability index are comprehensively weighted and judged to generate the load status assessment result; The characteristic values ​​of the load current distribution of each phase are numerically converted to obtain the load rate of phase A, the load rate of phase B, and the load rate of phase C, and then incorporated into the load status evaluation results.

[0011] Preferably, based on the real-time load rate in the comprehensive state parameter set and the total load change rate in the load characteristic parameter set, a two-parameter trend prediction is performed on the load state assessment result to generate a voltage regulation demand determination result, including: Extract the real-time load rate at the current sampling moment from the comprehensive state parameter set, and extract the total load change rate at the current sampling moment from the load characteristic parameter set; The real-time load rate is compared with a preset load rate threshold to obtain the load rate deviation value; Obtain multiple historical real-time load rates and multiple historical total load change rates corresponding to multiple historical sampling times within a historical period, and arrange the multiple historical real-time load rates and multiple historical total load change rates in chronological order to generate a historical load rate sequence and a historical change rate sequence. The load rate change gradient is obtained by differential processing between the real-time load rate and the last historical real-time load rate in the historical load rate sequence, and the total load change rate is obtained by differential processing between the total load change rate and the last historical total load change rate in the historical change rate sequence. Based on the load rate deviation value, the load rate change gradient, and the change rate change gradient, the load status assessment result is trend-corrected to obtain the corrected load trend prediction value. The corrected load trend prediction value is matched with the preset multi-level voltage regulation demand range to generate the voltage regulation demand determination result.

[0012] Preferably, generating the operating status adjustment command for the distribution transformer based on the voltage regulation demand determination result includes: Extract the voltage regulation direction identifier and voltage regulation amplitude requirement value from the voltage regulation requirement determination result; Obtain the current tap position number of the on-load tap changer of the distribution transformer, and determine the current voltage adjustment step size corresponding to the current tap position based on the current tap position number; The voltage regulation direction indicator and the voltage regulation amplitude requirement value are analyzed to determine the number of voltage regulation steps required for the on-load tap changer to perform. The target gear number of the on-load tap changer is determined by combining the current gear number and the number of gear change steps with a gear offset. The operating status adjustment command is generated based on the target gear number.

[0013] Preferably, the load state assessment results are subjected to phase-by-phase reactive power deficit quantification analysis based on the phase load imbalance of the load characteristic parameter set, and corresponding reactive power compensation adjustment commands are generated, including: The unbalance degree is converted by the load current distribution characteristic values ​​of each phase in the load characteristic parameter set to obtain the load unbalance degree of phase A, phase B and phase C. Extract the A-phase load rate, B-phase load rate, and C-phase load rate of the distribution transformer from the load condition assessment results; The load imbalance of phase A is correlated with the load rate of phase A, the load imbalance of phase B is correlated with the load rate of phase B, and the load imbalance of phase C is correlated with the load rate of phase C to obtain the reactive power deficit characterization values ​​of phase A, phase B, and phase C. The reactive power deficit characterization values ​​of phase A, phase B, and phase C are matched with preset deficit level ranges to obtain the compensation level of phase A, phase B, and phase C. Based on the compensation levels of phase A, phase B, and phase C, corresponding reactive power compensation adjustment commands are generated.

[0014] Preferably, the operating state adjustment command and the reactive power compensation adjustment command are executed according to a preset execution sequence to control the on-load tap changer and reactive power compensation device of the distribution transformer to perform corresponding adjustment actions, including: Extract the target gear number from the operating status adjustment command, and extract the corresponding target compensation capacity identifier from the reactive power compensation adjustment command; Obtain the preset execution timing, which includes the first action time of the on-load tap changer and the second action time of the reactive power compensation device; When the first action time point is reached, the target gear number is sent to the drive controller of the on-load tap changer to drive the on-load tap changer to perform a gear switching action from the current gear to the target gear number. When the second action time point is reached, the corresponding target compensation capacity identifiers are sent to the switching controllers of the corresponding reactive power compensation branches to drive each reactive power compensation branch to perform the corresponding capacitor or reactor switching action according to the target compensation capacity identifiers. There is a preset time interval between the first action time point and the second action time point, so that the switching action of each reactive power compensation branch is executed after the on-load tap changer completes the gear switching action.

[0015] The beneficial effects of this invention are as follows: 1. This invention deeply integrates the operating parameters of distribution transformers with the load parameters of the distribution system to construct a comprehensive set of state parameters and load characteristic parameters covering both the transformer's physical state and the system's load characteristics, providing a comprehensive and accurate data foundation for load regulation. Simultaneously, by employing a multi-dimensional comprehensive evaluation strategy comprised of load rate deviation, health status coefficient, and load adaptability indicators, the invention can accurately determine the transformer's load status from multiple perspectives, significantly improving the comprehensiveness and accuracy of the evaluation.

[0016] 2. This invention utilizes real-time load rate and total load change rate for dual-parameter trend prediction and introduces a historical time-series differential correction mechanism, enabling the determination of voltage regulation demand to be forward-looking and adaptive, effectively enhancing the timeliness and reliability of regulation decisions. Regarding three-phase load imbalance, by performing a quantitative correlation analysis between the imbalance degree of each phase and the phase load rate, precise matching and differentiated compensation of phase reactive power deficits are achieved, greatly improving the precision of reactive power compensation and the three-phase load balancing capability.

[0017] 3. By pre-setting the execution sequence, this invention scientifically staggers and orderly arranges the switching of on-load tap changer positions and reactive power compensation switching actions in time, ensuring the synergy of the two adjustment methods, effectively avoiding mutual interference during the adjustment process, and comprehensively improving the overall efficiency and operational stability of transformer load regulation in intelligent power distribution systems. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a transformer load regulation method for an intelligent power distribution system according to an embodiment of the present invention. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1: As Figure 1 The diagram shown is a flowchart illustrating a first embodiment of a transformer load regulation method for an intelligent power distribution system according to the present invention. The present invention proposes a transformer load regulation method for an intelligent power distribution system, comprising: S1. Collect the operating parameters of the distribution transformer and the load parameters of the distribution system; In this embodiment of the invention, the collection of operating parameters of the distribution transformer and load parameters of the distribution system includes: The three-phase voltage, three-phase current, winding temperature, and oil temperature of the distribution transformer are collected by voltage sensors, current sensors, and temperature sensors installed on the distribution transformer. The three-phase voltage, the three-phase current, the winding temperature, and the oil temperature are converted from analog to digital to generate a transformer digital signal set. By installing load monitoring terminals on the power distribution lines, the load current and load voltage of each feeder circuit of the power distribution system are collected, and the load current and load voltage are converted from analog to digital to generate a load digital signal set. The transformer digital signal set and the load digital signal set are normalized to generate the operating parameters and the load parameters.

[0022] Specifically, voltage sensors and current sensors are installed at the high-voltage side lead-out and low-voltage side lead-out of the distribution transformer body, respectively. At the same time, temperature sensors are embedded between the winding coils of the distribution transformer and temperature sensors are installed on the side wall of the oil tank. These three types of sensors are used to sense the three-phase voltage, three-phase current, winding temperature and oil temperature of the distribution transformer in real time during operation, and convert the sensed three-phase voltage, three-phase current, winding temperature and oil temperature into analog voltage signals and analog current signals, respectively. The analog signals output by the voltage sensor, current sensor, and temperature sensor are filtered and amplified by the signal conditioning circuit and then sent to the input channels of the analog-to-digital converter. The analog-to-digital converter samples the analog signals of each input channel at equal intervals according to a fixed sampling period, and converts the instantaneous amplitude acquired at each sampling moment into the corresponding digital code through the internal comparator and encoding logic. The digital codes of each channel are combined and arranged in the order of sampling time to form a transformer digital signal set containing the instantaneous sampled values ​​of three-phase voltage, three-phase current, winding temperature, and oil temperature. A load monitoring terminal is installed at the beginning of each feeder circuit in the power distribution system, near the power supply side. The voltage sampling resistor divider network inside the load monitoring terminal obtains the load voltage signal between the phase line and the neutral line of the feeder circuit, and the current transformer sampling circuit obtains the load current signal from the phase line of the feeder circuit. The load monitoring terminal outputs the obtained load voltage and load current signals in analog voltage form to its internal integrated analog-to-digital converter module. After sampling and quantizing the analog signals of each feeder circuit, the analog-to-digital converter module classifies and arranges the digital values ​​of each feeder circuit according to the feeder circuit number and sampling time, generating a load digital signal set containing the digital values ​​of the load voltage and load current of each feeder circuit at each sampling time. The instantaneous sampled values ​​of three-phase voltage and three-phase current at each sampling moment are extracted from the transformer digital signal set. At the same time, the digital values ​​of load voltage and load current of each feeder circuit at each sampling moment are extracted from the load digital signal set. Data entries with the same time identifier in the two sets of digital signals are associated and classified into the same data record entry. The associated data entries are processed to unify the dimensions, converting the digital codes into values ​​expressed in standard units. The three-phase voltage, three-phase current, winding temperature and oil temperature values ​​at the same sampling moment after unification of dimensions are summarized and arranged to form operating parameters. At the same time, the load voltage and load current values ​​of each feeder circuit at the same sampling moment are classified and arranged according to the feeder number to form load parameters.

[0023] The beneficial effects are that by synchronously collecting electrical and thermal states of the transformer body and load electrical quantities of each feeder circuit of the power distribution system through the sensor array, a complete data source covering the internal operating status of the equipment and the load characteristics of the external system is provided for subsequent parameter fusion. This lays the data foundation for multi-dimensional comprehensive evaluation and ensures that subsequent load status judgment and adjustment command generation can be based on comprehensive and accurate real-time information on site, thereby ensuring the reliable implementation and effective execution of the overall adjustment method.

[0024] S2. Perform parameter fusion on the operating parameters and the load parameters to obtain the comprehensive state parameter set of the distribution transformer and the load characteristic parameter set of the distribution system; In this embodiment of the invention, the operating parameters and the load parameters are fused to obtain a comprehensive state parameter set for the distribution transformer and a load characteristic parameter set for the distribution system, including: The effective values ​​of the high-voltage side line voltage, low-voltage side line voltage, high-voltage side line current, low-voltage side line current, and winding temperature state value of the distribution transformer are extracted from the operating parameters and used as the state characteristic components of the distribution transformer. The effective values ​​of phase current and phase voltage of each feeder circuit of the power distribution system are extracted from the load parameters and used as the load characteristic components of the power distribution system. The state feature components and the load feature components are timestamped and then normalized to obtain state normalized features and load normalized features. By using preset electrical correlation weights, the state normalization features and the load normalization features are weighted and combined to generate the comprehensive state parameter set; The effective values ​​of phase current and phase voltage of each feeder in the load normalization characteristics are grouped and integrated according to the feeder number to generate the load characteristic parameter set.

[0025] By using preset electrical correlation weights, the state normalization features and the load normalization features are weighted and combined to generate the comprehensive state parameter set, including: The electrical association weight is read from a preset weight configuration table. The electrical association weight includes a first weight coefficient corresponding to the state feature component and a second weight coefficient corresponding to the load feature component. The state feature component is multiplied by the first weighting coefficient to obtain the weighted state component. The load characteristic component is multiplied by the second weighting coefficient to obtain the weighted load component; Based on the electrical structure hierarchy of the distribution transformer, the weighted state components and the weighted load components are accumulated and combined to generate the comprehensive state parameter set.

[0026] Specifically, the high-voltage side line voltage, low-voltage side line voltage, high-voltage side line current, low-voltage side line current, and winding temperature values ​​at each sampling time are extracted from the operating parameters. The extracted high-voltage side line voltage, low-voltage side line voltage, high-voltage side line current, and low-voltage side line current are converted into effective values ​​to obtain the effective values ​​of high-voltage side line voltage, low-voltage side line voltage, high-voltage side line current, and low-voltage side line current at each sampling time. The above four effective values, together with the winding temperature value, are used as the state characteristic components of the distribution transformer. The load voltage and load current values ​​of each feeder circuit at each sampling time are extracted from the load parameters. The effective value of the load voltage value of each feeder circuit is converted to the effective value to obtain the effective value of the phase voltage of that feeder circuit. The effective value of the load current value of each feeder circuit is converted to the effective value to obtain the effective value of the phase current of that feeder circuit. The effective values ​​of the phase voltage and phase current of all feeder circuits in the power distribution system at each sampling time are used together as the load characteristic components of the power distribution system. State characteristic components and load characteristic components are categorized according to their respective sampling times. The effective values ​​of high-voltage line voltage, low-voltage line voltage, high-voltage line current, low-voltage line current, winding temperature, phase voltage of each feeder circuit, and phase current of each feeder circuit belonging to the same sampling time are grouped into the same data record entry. Normalization is then performed on each component of the state characteristic components and each component of the load characteristic components belonging to the same data record entry. The current value of each component is divided by its maximum absolute value across all sampling times to obtain the normalized value of each component in the range of zero to one. The normalized state characteristic components constitute the state normalized feature, and the normalized load characteristic components constitute the load normalized feature. The first weighting coefficient corresponding to the state feature component and the second weighting coefficient corresponding to the load feature component are read from the preset weighting configuration table. The normalized values ​​of the high-voltage side line voltage RMS, low-voltage side line voltage RMS, high-voltage side line current RMS, low-voltage side line current RMS, and winding temperature RMS are multiplied by their corresponding first weighting coefficients from the state normalization features to obtain the weighted result of each state feature component. Simultaneously, the load normalization features are also multiplied by their corresponding first weighting coefficients. The normalized values ​​of the effective phase voltages of each feeder circuit are multiplied by the corresponding second weighting coefficients, and the normalized values ​​of the effective phase currents of each feeder circuit are multiplied by the corresponding second weighting coefficients to obtain the weighted results of each load characteristic component. The weighted results of all state characteristic components and all load characteristic components at the same sampling time are accumulated and combined according to the electrical structure level of the distribution transformer. That is, the weighted results of the high-voltage side electrical quantities are first grouped into one group, the weighted results of the low-voltage side electrical quantities are grouped into one group, and the weighted results of the winding temperature are grouped into a separate group. Then, the weighted results of the components belonging to the same electrical structure level in each group are added together to generate a comprehensive state parameter set. The normalized values ​​of the effective phase voltage and the effective phase current of each feeder circuit in the load normalization characteristics are classified according to the feeder number. The normalized values ​​of the effective phase voltage and the effective phase current belonging to the same feeder number are assigned to the data group corresponding to that feeder number. The data groups corresponding to each feeder number are arranged and summarized in the order of the feeder numbers to form a load characteristic parameter set.

[0027] Specifically, the first weight coefficient corresponding to the state characteristic component and the second weight coefficient corresponding to the load characteristic component are read from the preset weight configuration table. The weight configuration table stores multiple sets of weight coefficients in advance according to the rated capacity level and connection group of the distribution transformer.

[0028] It should be noted that the first weight coefficient and the second weight coefficient stored in the weight configuration table are determined as follows: based on the rated capacity level and connection group of the distribution transformer, a set of weight coefficients matching the current transformer model is selected from multiple pre-stored sets of weight coefficients; wherein, the sum of the first weight coefficient and the second weight coefficient is 1, and the allocation ratio of the high-voltage side electrical quantity weight, the low-voltage side electrical quantity weight, and the temperature status quantity weight in the first weight coefficient is set according to the priority of the operating conditions of the distribution transformer. When operating below the rated load, the temperature status quantity weight is 0.1 to 0.2 and the electrical quantity weight is 0.8 to 0.9; when operating above the rated load, the temperature status quantity weight is 0.3 to 0.5 and the electrical quantity weight is 0.5 to 0.7.

[0029] The weight coefficients in the weight configuration table are predetermined by the following method: obtaining the historical operating parameters and historical load parameters of the distribution transformer in multiple historical operating cycles, taking the optimal actual operating efficiency of the distribution transformer as the objective function, and using a genetic algorithm or particle swarm optimization algorithm to optimize the first weight coefficient and the second weight coefficient, and storing the optimization results in the weight configuration table.

[0030] The normalized values ​​of the effective values ​​of the high-voltage side line voltage, low-voltage side line voltage, high-voltage side line current, low-voltage side line current, and winding temperature in the state normalization features are multiplied by their respective first weighting coefficients to obtain the weighted state components of each state feature component. The normalized values ​​of the effective phase voltage and the effective phase current of each feeder circuit in the load normalization feature are multiplied by their respective second weighting coefficients to obtain the weighted load components of each feeder circuit. Based on the three electrical structure levels of the distribution transformer—the high-voltage side electrical layer, the low-voltage side electrical layer, and the temperature state layer—the weighted state components and weighted load components belonging to the high-voltage side electrical layer are summed together, the weighted state components and weighted load components belonging to the low-voltage side electrical layer are summed together, and the weighted state components belonging to the temperature state layer are collected separately. After the three levels are summed and combined, they together form a comprehensive state parameter set.

[0031] The beneficial effect is that by aligning and normalizing the operating parameters and load parameters with timestamps, the transformer state characteristics and the power distribution system load characteristics are weighted and combined under a unified dimension. This allows the electrical and thermal state quantities of the transformer body to be correlated with the load quantities of each feeder on the system side at the electrical structure level, thus constructing a comprehensive set of state parameters. This provides a structured data foundation that simultaneously covers the internal operating conditions of the equipment and the external load distribution of the system for subsequent multi-dimensional load state assessment and dual-parameter trend prediction.

[0032] The weight coefficients corresponding to the transformer state characteristics and system load characteristics are read from the weight configuration table. After weighting the two types of characteristics after normalization, they are accumulated and combined according to the electrical structure level. This allows the transformer body state and the power distribution system load to reflect their respective influence levels differently according to different electrical levels during the fusion process. This avoids the feature imbalance caused by equal weight superposition and provides a comprehensive state parameter set that is consistent with the actual physical structure for subsequent multi-dimensional load state assessment.

[0033] S3. Based on the comprehensive state parameter set and the load characteristic parameter set, perform a multi-dimensional comprehensive evaluation of the current load state of the distribution transformer and generate a load state evaluation result; In this embodiment of the invention, based on the comprehensive state parameter set and the load characteristic parameter set, a multi-dimensional comprehensive evaluation of the current load state of the distribution transformer is performed to generate a load state evaluation result, including: Feature extraction is performed on the comprehensive state parameter set to obtain the real-time load rate, winding temperature state value, and insulation aging characteristic value of the distribution transformer; Feature extraction is performed on the load characteristic parameter set to obtain the total load power fluctuation characteristic value and the load current distribution characteristic value of each phase of the power distribution system; The real-time load rate is compared with a preset standard load rate range to generate a load rate deviation index for the distribution transformer. By combining the winding temperature state value and the insulation aging characteristic value, the thermal state and insulation state of the distribution transformer are analyzed in a coordinated manner to generate the health state coefficient of the distribution transformer. By using the total load power fluctuation characteristic value and the load current distribution characteristic value of each phase, the load fluctuation characteristics and three-phase load distribution characteristics of the power distribution system are correlated and analyzed to generate the load adaptability index of the power distribution system. The load rate deviation index, the health status coefficient, and the load adaptability index are comprehensively weighted and judged to generate the load status assessment result; The characteristic values ​​of the load current distribution of each phase are numerically converted to obtain the load rate of phase A, the load rate of phase B, and the load rate of phase C, and then incorporated into the load status evaluation results.

[0034] Specifically, the cumulative combination results belonging to the high-voltage side electrical layer, low-voltage side electrical layer, and temperature state layer in the comprehensive state parameter set are analyzed item by item. The load rate value at the current sampling time is extracted from the cumulative values ​​of the high-voltage side electrical layer and the low-voltage side electrical layer as the real-time load rate. The current value of the winding temperature is extracted from the cumulative values ​​of the temperature state layer as the winding temperature state value. At the same time, the insulation-related cumulative values ​​at each sampling time in the comprehensive state parameter set are trend-tracked in chronological order to obtain the current characterization value of the change in insulation performance with the accumulation of operating time as the insulation aging characteristic value.

[0035] The data sets corresponding to each feeder number in the load characteristic parameter set are analyzed item by item. The power-related values ​​of each feeder circuit at the current sampling time are extracted from the normalized values ​​of the effective phase voltage and the normalized values ​​of the effective phase current of each feeder circuit. The power values ​​of all feeder circuits are added together to obtain the total load power of the power distribution system. Then, the variation amplitude of the total load power between multiple consecutive sampling times is extracted to obtain the total load power fluctuation characteristic value. At the same time, the three-phase current distribution values ​​of each feeder circuit at each sampling time are extracted from the load characteristic parameter set. The current values ​​of all feeder circuits are collected by phase A, phase B, and phase C to obtain the load current distribution characteristic value of each phase.

[0036] The real-time load rate is compared numerically with the upper and lower limits of the preset standard load rate range. When the real-time load rate is higher than the upper limit, the deviation is calculated; when the real-time load rate is lower than the lower limit, the deviation is calculated; when the real-time load rate is within the range, the deviation is recorded as zero. The calculated deviation is used as the load rate deviation index of the distribution transformer. The winding temperature status value and insulation aging characteristic value are input into the collaborative analysis process. In this collaborative analysis process, it is first determined whether the winding temperature status value is within the safe operating temperature range, and then it is determined whether the insulation aging characteristic value exceeds the insulation life. The permissible aging limit is determined, and a health status coefficient characterizing the degree of coordination between the current thermal state and insulation state of the distribution transformer is output after combining the results of the two judgments. The total load power fluctuation characteristic value and the load current distribution characteristic value of each phase are input into the correlation analysis process. In this correlation analysis process, it is first determined whether the total load power fluctuation characteristic value exceeds the permissible power fluctuation range, and then it is determined whether the difference between the load current distribution characteristic values ​​of each phase exceeds the permissible three-phase imbalance range. After combining the results of the two judgments, a load fit index characterizing the degree of fit between the current load fluctuation characteristics and the three-phase load distribution characteristics of the distribution system is output.

[0037] It should be noted that the insulation aging characteristic value is determined based on the Arrhenius thermal aging model, that is, based on the winding temperature-time data accumulated during the historical operation of the distribution transformer, the equivalent aging time converted to the rated temperature is calculated, and the ratio of the equivalent aging time to the insulation design life is used as the insulation aging characteristic value. The closer the ratio is to 1, the more severe the insulation aging.

[0038] The load rate deviation index, health status coefficient, and load fit index are comprehensively weighted according to a preset evaluation ratio. That is, the load rate deviation index, health status coefficient, and load fit index are multiplied by their respective evaluation weights and then summed to obtain the load status evaluation result. At the same time, the load rate conversion is performed on the A-phase current collection value, B-phase current collection value, and C-phase current collection value in the load current distribution characteristic value of each phase. That is, the A-phase load rate, B-phase load rate, and C-phase load rate are obtained by dividing the A-phase current collection value by the rated phase current value of the distribution transformer. The above three phase load rates are included as additional items in the load status evaluation result and output together.

[0039] The beneficial effects are that the transformer load rate, winding temperature and insulation aging characteristics are extracted from the comprehensive state parameter set, and the total load power fluctuation and load current distribution characteristics of each phase are extracted from the load characteristic parameter set. Load rate deviation index, health state coefficient and load adaptability index are generated respectively. Then, the load state assessment result is generated by comprehensive weighting judgment. This realizes the multi-dimensional collaborative assessment of the thermal state and insulation state of the transformer body, the load fluctuation and three-phase distribution on the system side, and provides a comprehensive judgment basis that fully reflects the current load state for subsequent dual-parameter trend prediction and phase-by-phase reactive power compensation.

[0040] S4. Based on the real-time load rate in the comprehensive state parameter set and the total load change rate in the load characteristic parameter set, perform dual-parameter trend prediction on the load state assessment result, generate voltage regulation demand determination result, and generate the operation state adjustment command of the distribution transformer based on the voltage regulation demand determination result. In this embodiment of the invention, based on the real-time load rate in the comprehensive state parameter set and the total load change rate in the load characteristic parameter set, a two-parameter trend prediction is performed on the load state assessment result to generate a voltage regulation demand determination result, including: Extract the real-time load rate at the current sampling moment from the comprehensive state parameter set, and extract the total load change rate at the current sampling moment from the load characteristic parameter set; The real-time load rate is compared with a preset load rate threshold to obtain the load rate deviation value; Obtain multiple historical real-time load rates and multiple historical total load change rates corresponding to multiple historical sampling times within a historical period, and arrange the multiple historical real-time load rates and multiple historical total load change rates in chronological order to generate a historical load rate sequence and a historical change rate sequence. The load rate change gradient is obtained by differential processing between the real-time load rate and the last historical real-time load rate in the historical load rate sequence, and the total load change rate is obtained by differential processing between the total load change rate and the last historical total load change rate in the historical change rate sequence. Based on the load rate deviation value, the load rate change gradient, and the change rate change gradient, the load status assessment result is trend-corrected to obtain the corrected load trend prediction value. The corrected load trend prediction value is matched with the preset multi-level voltage regulation demand range to generate the voltage regulation demand determination result.

[0041] Based on the voltage regulation demand determination result, an operation status adjustment command for the distribution transformer is generated, including: Extract the voltage regulation direction identifier and voltage regulation amplitude requirement value from the voltage regulation requirement determination result; Obtain the current tap position number of the on-load tap changer of the distribution transformer, and determine the current voltage adjustment step size corresponding to the current tap position based on the current tap position number; The voltage regulation direction indicator and the voltage regulation amplitude requirement value are analyzed to determine the number of voltage regulation steps required for the on-load tap changer to perform. The target gear number of the on-load tap changer is determined by combining the current gear number and the number of gear change steps with a gear offset. The operating status adjustment command is generated based on the target gear number.

[0042] Specifically, the cumulative combination results belonging to the high-voltage side electrical layer, low-voltage side electrical layer, and temperature state layer in the comprehensive state parameter set are analyzed item by item. The load rate value at the current sampling time is extracted from the cumulative values ​​of the high-voltage side electrical layer and the low-voltage side electrical layer as the real-time load rate. The current value of the winding temperature is extracted from the cumulative values ​​of the temperature state layer as the winding temperature state value. At the same time, the insulation-related cumulative values ​​at each sampling time in the comprehensive state parameter set are trend-tracked in chronological order to obtain the current characterization value of the change in insulation performance with the accumulation of operating time as the insulation aging characteristic value. The data sets corresponding to each feeder number in the load characteristic parameter set are analyzed item by item. The power-related values ​​of each feeder circuit at the current sampling time are extracted from the normalized values ​​of the effective phase voltage and the normalized values ​​of the effective phase current of each feeder circuit. The power values ​​of all feeder circuits are added together to obtain the total load power of the power distribution system. Then, the variation amplitude of the total load power between multiple consecutive sampling times is extracted to obtain the total load power fluctuation characteristic value. At the same time, the three-phase current distribution values ​​of each feeder circuit at each sampling time are extracted from the load characteristic parameter set. The current values ​​of all feeder circuits are collected by phase A, phase B, and phase C to obtain the load current distribution characteristic value of each phase. The real-time load rate is compared numerically with the upper and lower limits of the preset standard load rate range. When the real-time load rate is higher than the upper limit, the deviation is calculated; when the real-time load rate is lower than the lower limit, the deviation is calculated; when the real-time load rate is within the range, the deviation is recorded as zero. The calculated deviation is used as the load rate deviation index of the distribution transformer. The winding temperature status value and insulation aging characteristic value are input into the collaborative analysis process. In this collaborative analysis process, the temperature safety margin of the winding temperature status value relative to the preset maximum allowable operating temperature and the insulation life remaining rate of the insulation aging characteristic value relative to the preset insulation life end threshold are calculated respectively. Then, the temperature safety margin and the insulation life remaining rate are normalized and weighted to obtain the health status coefficient. Its expression is: health status coefficient = λ × temperature safety margin + μ × insulation life remaining rate, where λ + μ = 1, and λ and μ are dynamically set according to the transformer operation stage (for example, λ is 0.4 and μ is 0.6 in the early stage of operation, and λ is 0.7 and μ is 0.3 in the later stage of operation).

[0043] The total load power fluctuation characteristic value and the load current distribution characteristic value of each phase are input into the correlation analysis process. In this correlation analysis process, the power fluctuation over-limit rate of the total load power fluctuation characteristic value relative to the preset maximum allowable power fluctuation amount, and the three-phase unbalance over-limit rate of the maximum difference between the load current distribution characteristic values ​​of each phase relative to the three-phase average current are first calculated. Then, the power fluctuation over-limit rate and the three-phase unbalance over-limit rate are weighted and combined to obtain the load adaptability index. Its expression is: Load adaptability index = 1 - [γ × power fluctuation over-limit rate + δ × three-phase unbalance over-limit rate], where γ + δ = 1, and γ and δ are preset according to the sensitivity of the power distribution system to power fluctuation and three-phase balance (for example, γ is 0.6 and δ is 0.4 in industrial power distribution scenarios, and γ is 0.4 and δ is 0.6 in civil power distribution scenarios).

[0044] It should be noted that the preset standard load rate range is determined based on the rated capacity and insulation heat resistance class of the distribution transformer: for Class A insulation transformers, the lower limit of the standard load rate range is 20% and the upper limit is 80%; for Class B insulation transformers, the lower limit of the standard load rate range is 25% and the upper limit is 85%; for Class F insulation transformers, the lower limit of the standard load rate range is 30% and the upper limit is 90%; and for Class H insulation transformers, the lower limit of the standard load rate range is 35% and the upper limit is 95%.

[0045] The upper and lower limits of the standard load rate range are also corrected according to the ambient temperature of the environment where the distribution transformer is located: when the ambient temperature is higher than 40℃, the upper limit is corrected downward by 1% for every 1℃ increase; when the ambient temperature is lower than -5℃, the upper limit is corrected upward by 0.5% for every 1℃ decrease, but the corrected upper limit does not exceed 98%.

[0046] The load rate deviation index, health status coefficient, and load fit index are comprehensively weighted according to a preset evaluation ratio. That is, the load rate deviation index, health status coefficient, and load fit index are multiplied by their respective evaluation weights and then summed to obtain the load status evaluation result. At the same time, the load rate conversion is performed on the A-phase current collection value, B-phase current collection value, and C-phase current collection value in the load current distribution characteristic value of each phase. That is, the A-phase load rate, B-phase load rate, and C-phase load rate are obtained by dividing the A-phase current collection value by the rated phase current value of the distribution transformer. The above three phase load rates are included as additional items in the load status evaluation result and output together.

[0047] Specifically, the load rate value at the current sampling moment is extracted from the comprehensive state parameter set as the real-time load rate, and the change value of the total load power of the power distribution system at the current sampling moment per unit time is extracted from the load characteristic parameter set as the total load change rate. The real-time load rate is compared with the preset load rate threshold, and the difference between the real-time load rate and the preset load rate threshold is calculated. This difference is used as the load rate deviation value. It should be noted that the preset load rate threshold is the economic operating load rate of the distribution transformer, which is taken as 60% to 75% of the rated load rate of the distribution transformer. More preferably, for distribution transformers with a capacity of 1000kVA and below, the load rate threshold is 65%; for those with a capacity of 1000kVA to 5000kVA, the load rate threshold is 70%; and for those with a capacity of 5000kVA and above, the load rate threshold is 75%.

[0048] The load rate threshold is also dynamically adjusted based on the daily load curve of the distribution transformer: the average load rate of the distribution transformer at the same time over the past 7 natural days is obtained. If the average value is higher than the currently set load rate threshold, the load rate threshold is adjusted upward to the midpoint between the average value and the current threshold; if the average value is lower than the currently set load rate threshold, the current threshold is maintained unchanged.

[0049] The historical real-time load rate and historical total load change rate corresponding to multiple historical sampling times within a historical period are obtained. The obtained historical real-time load rates are arranged into a data sequence according to the sampling time from first to last to form a historical load rate sequence. At the same time, the obtained historical total load change rates are arranged into a data sequence according to the same sampling time order to form a historical change rate sequence. The difference between the real-time load rate and the historical real-time load rate at the last sampling moment in the historical load rate sequence is calculated, i.e., the real-time load rate is subtracted from the last historical real-time load rate to obtain the load rate change gradient. At the same time, the difference between the total load change rate and the historical total load change rate at the last sampling moment in the historical change rate sequence is calculated, i.e., the total load change rate is subtracted from the last historical total load change rate to obtain the change rate change gradient. The load rate deviation, load rate change gradient, and change rate change gradient are input into the trend correction process. In this process, the load rate deviation and load rate change gradient are added to obtain the first intermediate value. Then, the first intermediate value is added to the change rate change gradient to obtain the second intermediate value. Finally, the second intermediate value is superimposed on the load status assessment result to obtain the corrected load trend prediction value. The corrected load trend prediction value is compared with each of the preset multi-level voltage regulation demand ranges in turn to determine which voltage regulation demand range the corrected load trend prediction value falls into. The voltage regulation direction indicator and voltage regulation amplitude level corresponding to that range are output as the voltage regulation demand determination result.

[0050] It should be noted that the preset multi-level voltage regulation demand range is divided into five levels according to the magnitude of the corrected load trend prediction value: Level 1: Predicted value ≤ -10%, corresponding to the voltage regulation direction marked "reduction", and the voltage regulation amplitude demand value is "large" (corresponding to 3 to 5 steps of on-load tap changer position change); Level 2: -10% < predicted value ≤ -3%, corresponding to the voltage regulation direction marked "reduction", and the voltage regulation amplitude demand value is "small" (corresponding to 1 to 2 steps of position change); Level 3: -3% < predicted value ≤ 3%, corresponding to the voltage regulation direction marked "maintain", and the voltage regulation amplitude demand value is "zero" (corresponding to 0 steps of position change); Level 4: 3% < predicted value ≤ 10%, corresponding to the voltage regulation direction marked "increase", and the voltage regulation amplitude demand value is "small" (corresponding to 1 to 2 steps of position change); Level 5: Predicted value > 10%, corresponding to the voltage regulation direction marked "increase", and the voltage regulation amplitude demand value is "large" (corresponding to 3 to 5 steps of position change).

[0051] The corrected load trend prediction value is expressed as a percentage value normalized based on the rated load rate of the distribution transformer.

[0052] The beneficial effects are that by introducing dual-parameter trend prediction of real-time load rate and total load change rate, and combining the differential processing of historical sequences to obtain the load rate change gradient and change rate change gradient, the load status assessment results are trend-corrected and matched with multi-level voltage regulation demand ranges. This enables the determination of voltage regulation demand to have a forward-looking perspective based on the current deviation and historical change trends, overcomes the lag problem caused by static judgment at a single moment, and improves the timeliness and adaptability of voltage regulation decision-making.

[0053] The voltage regulation direction indicator and voltage regulation amplitude requirement value are extracted from the voltage regulation demand determination result. Combined with the current tap position number of the on-load tap changer and its corresponding voltage regulation step size, the voltage regulation step number is analyzed and the tap position offset is synthesized to generate an operating status adjustment command containing the target tap position number. This enables the voltage regulation demand to be accurately converted into a target tap position command that the on-load tap changer can directly execute. This ensures a precise correspondence between the adjustment command and the actual execution capability of the equipment, and provides a clear control basis for the subsequent orderly execution of tap position switching actions.

[0054] S5. Based on the load imbalance of each phase in the load characteristic parameter set, perform phase-by-phase reactive power deficit quantitative analysis on the load state assessment results, and generate corresponding reactive power compensation adjustment commands. In this embodiment of the invention, the load state assessment results are subjected to phase-by-phase reactive power deficit quantification analysis based on the load imbalance of each phase in the load characteristic parameter set, and corresponding reactive power compensation adjustment commands are generated, including: The unbalance degree is converted by the load current distribution characteristic values ​​of each phase in the load characteristic parameter set to obtain the load unbalance degree of phase A, phase B and phase C. Extract the A-phase load rate, B-phase load rate, and C-phase load rate of the distribution transformer from the load condition assessment results; The load imbalance of phase A is correlated with the load rate of phase A, the load imbalance of phase B is correlated with the load rate of phase B, and the load imbalance of phase C is correlated with the load rate of phase C to obtain the reactive power deficit characterization values ​​of phase A, phase B, and phase C. The reactive power deficit characterization values ​​of phase A, phase B, and phase C are matched with preset deficit level ranges to obtain the compensation level of phase A, phase B, and phase C. Based on the compensation levels of phase A, phase B, and phase C, corresponding reactive power compensation adjustment commands are generated.

[0055] Specifically, the load current distribution characteristic values ​​of each phase of each feeder circuit in the load characteristic parameter set are collected phase by phase. The average value of the current values ​​of all feeder circuits in phase A, phase B, and phase C are calculated separately. The three-phase average current values ​​are added together and divided by three to obtain the total average value of the three-phase average current. The percentage of the difference between the average current value of phase A and the total average value of the three-phase average current is calculated as the load unbalance of phase A, the load unbalance of phase B, and the load unbalance of phase C. Extract the three additional values ​​of phase A load rate, phase B load rate, and phase C load rate from the load status assessment results. The correlation analysis between the load imbalance of phase A and the load rate of phase A is carried out. Specifically, the value of the load imbalance of phase A is multiplied by the value of the load rate of phase A, and the product is used as the reactive power deficit value of phase A. At the same time, the load imbalance of phase B is multiplied by the load rate of phase B to obtain the reactive power deficit value of phase B. The load imbalance of phase C is multiplied by the load rate of phase C to obtain the reactive power deficit value of phase C. The reactive power deficit value of phase A is matched with a preset deficit level range. This preset deficit level range is divided into multiple consecutive deficit level segments according to the value from smallest to largest, and each segment corresponds to a compensation level. The compensation level of phase A is obtained by determining which segment the reactive power deficit value of phase A falls into. At the same time, the reactive power deficit values ​​of phase B and phase C are matched with the preset deficit level ranges respectively to obtain the compensation level of phase B and phase C respectively.

[0056] It should be noted that, preferably, the preset reactive power deficit level range is divided into four levels according to the magnitude of the reactive power deficit value: Level 1: Reactive power deficit value ≤ 5%, corresponding to compensation level "Level 0", and corresponding reactive power compensation capacity of 0 kvar (no switching); Level 2: 5% < reactive power deficit value ≤ 20%, corresponding to compensation level "Level 1", and corresponding reactive power compensation capacity of 25% of the rated reactive power compensation capacity of the phase; Level 3: 20% < reactive power deficit value ≤ 50%, corresponding to compensation level "Level 2", and corresponding reactive power compensation capacity of 50% of the rated reactive power compensation capacity of the phase; Level 4: Reactive power deficit value > 50%, corresponding to compensation level "Level 3", and corresponding reactive power compensation capacity of 100% of the rated reactive power compensation capacity of the phase.

[0057] The reactive power deficit value is a percentage value normalized based on the rated reactive power of the distribution transformer; the corresponding rated reactive power compensation capacity is predetermined based on the rated capacity of the distribution transformer and the rated voltage of the phase.

[0058] Based on the preset reactive power compensation capacity values ​​corresponding to the compensation levels of phase A, phase B, and phase C, reactive power compensation adjustment instructions containing target compensation capacity identifiers are generated for phases A, B, and C respectively. The instruction content of each phase in the reactive power compensation adjustment instruction uniquely corresponds to the compensation level of that phase.

[0059] The beneficial effects are that by performing a correlation analysis between the load imbalance of each phase and the load rate of each phase, the reactive power deficit characterization value of each phase is obtained. Then, by matching the preset deficit level range, an independent compensation level and corresponding reactive power compensation adjustment command are generated for each phase, realizing a differentiated reactive power compensation scheme for the three phases A, B, and C. This effectively overcomes the defect of the traditional unified compensation method that cannot accurately allocate reactive power by phase, significantly improves the pertinence and precision of three-phase reactive power compensation, and enhances the balance and stability of the three-phase operation of the power distribution system.

[0060] S6. According to the preset execution sequence, execute the operating status adjustment command and the reactive power compensation adjustment command to control the on-load tap changer and reactive power compensation device of the distribution transformer to perform corresponding adjustment actions respectively.

[0061] In this embodiment of the invention, the operating state adjustment command and the reactive power compensation adjustment command are executed according to a preset execution sequence to control the on-load tap changer and the reactive power compensation device of the distribution transformer to perform corresponding adjustment actions, including: Extract the target gear number from the operating status adjustment command, and extract the corresponding target compensation capacity identifier from the reactive power compensation adjustment command; Obtain the preset execution timing, which includes the first action time of the on-load tap changer and the second action time of the reactive power compensation device; When the first action time point is reached, the target gear number is sent to the drive controller of the on-load tap changer to drive the on-load tap changer to perform a gear switching action from the current gear to the target gear number. When the second action time point is reached, the corresponding target compensation capacity identifiers are sent to the switching controllers of the corresponding reactive power compensation branches to drive each reactive power compensation branch to perform the corresponding capacitor or reactor switching action according to the target compensation capacity identifiers. There is a preset time interval between the first action time point and the second action time point, so that the switching action of each reactive power compensation branch is executed after the on-load tap changer completes the gear switching action.

[0062] Specifically, the target gear number is parsed from the operating status adjustment command, and the target compensation capacity identifiers corresponding to A, B, and C are parsed from the reactive power compensation adjustment command, respectively. The first action time point value of the on-load tap changer and the second action time point value of the reactive power compensation device are read from the preset execution timing configuration. In this execution timing configuration, the first action time point is earlier than the second action time point and the time interval between the two has been preset. It should be noted that the preset time interval is determined based on the operating time required for the on-load tap changer to complete a single gear switching action, and the value of the time interval is not less than the cumulative operating time required for the on-load tap changer to complete the switching from the current gear to the target gear. The operating time required for the on-load tap changer to complete a single gear switching action is 3 to 8 seconds, and the value of the time interval is 5 to 30 seconds.

[0063] When the number of gear changes is 1 step, the time interval is 5 seconds; when the number of gear changes is 2 steps, the time interval is 10 seconds; when the number of gear changes is 3 steps, the time interval is 15 seconds; when the number of gear changes is 4 steps, the time interval is 20 seconds; and when the number of gear changes is 5 steps or more, the time interval is 30 seconds.

[0064] At the moment the first action time point arrives, the parsed target gear number is sent to the on-load tap changer drive controller through the communication interface. After receiving the target gear number, the drive controller drives the on-load tap changer operating mechanism to perform gear switching action, so that the on-load tap changer switches from the current gear to the gear indicated by the target gear number step by step. At the moment when the second action time point arrives, the target compensation capacity identifier of phase A is sent to the switching controller of phase A reactive power compensation branch, the target compensation capacity identifier of phase B is sent to the switching controller of phase B reactive power compensation branch, and the target compensation capacity identifier of phase C is sent to the switching controller of phase C reactive power compensation branch. Each phase switching controller drives its corresponding reactive power compensation branch to perform the switching action of capacitor or reactor according to the received target compensation capacity identifier. The pre-set time interval between the first and second action points ensures that after the on-load tap changer completes all the switching actions from the current position to the target position number, the switching controller of each phase reactive power compensation branch will start to execute the switching actions of its corresponding capacitor or reactor.

[0065] The beneficial effects are that by controlling the on-load tap changer and reactive power compensation device to operate in a time-sharing and orderly manner through preset execution timing, the switching of the tap position and the switching of reactive power are staggered in time, avoiding electrical disturbances and regulation conflicts that may occur when the two regulation methods are executed simultaneously, ensuring that voltage regulation and reactive power compensation are completed independently and stably, while ensuring the execution order and operation reliability of the entire load regulation process.

[0066] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A transformer load regulation method for an intelligent power distribution system, characterized in that, The method includes: S1. Collect the operating parameters of the distribution transformer and the load parameters of the distribution system; S2. Perform parameter fusion on the operating parameters and the load parameters to obtain the comprehensive state parameter set of the distribution transformer and the load characteristic parameter set of the distribution system; S3. Based on the comprehensive state parameter set and the load characteristic parameter set, perform a multi-dimensional comprehensive evaluation of the current load state of the distribution transformer and generate a load state evaluation result; S4. Based on the real-time load rate in the comprehensive state parameter set and the total load change rate in the load characteristic parameter set, perform dual-parameter trend prediction on the load state assessment result, generate voltage regulation demand determination result, and generate the operation state adjustment command of the distribution transformer based on the voltage regulation demand determination result. S5. Based on the load imbalance of each phase in the load characteristic parameter set, perform phase-by-phase reactive power deficit quantitative analysis on the load state assessment results, and generate corresponding reactive power compensation adjustment commands. S6. According to the preset execution sequence, execute the operating status adjustment command and the reactive power compensation adjustment command to control the on-load tap changer and reactive power compensation device of the distribution transformer to perform corresponding adjustment actions respectively.

2. The transformer load regulation method for an intelligent power distribution system as described in claim 1, characterized in that, Collect operating parameters of the distribution transformer and load parameters of the distribution system, including: The three-phase voltage, three-phase current, winding temperature, and oil temperature of the distribution transformer are collected by voltage sensors, current sensors, and temperature sensors installed on the distribution transformer. The three-phase voltage, the three-phase current, the winding temperature, and the oil temperature are converted from analog to digital to generate a transformer digital signal set. By installing load monitoring terminals on the power distribution lines, the load current and load voltage of each feeder circuit of the power distribution system are collected, and the load current and load voltage are converted from analog to digital to generate a load digital signal set. The transformer digital signal set and the load digital signal set are normalized to generate the operating parameters and the load parameters.

3. The transformer load regulation method for an intelligent power distribution system as described in claim 1, characterized in that, The operating parameters and the load parameters are fused to obtain a comprehensive state parameter set for the distribution transformer and a load characteristic parameter set for the distribution system, including: The effective values ​​of the high-voltage side line voltage, low-voltage side line voltage, high-voltage side line current, low-voltage side line current, and winding temperature state value of the distribution transformer are extracted from the operating parameters and used as the state characteristic components of the distribution transformer. The effective values ​​of phase current and phase voltage of each feeder circuit of the power distribution system are extracted from the load parameters and used as the load characteristic components of the power distribution system. The state feature components and the load feature components are timestamped and then normalized to obtain state normalized features and load normalized features. By using preset electrical correlation weights, the state normalization features and the load normalization features are weighted and combined to generate the comprehensive state parameter set; The effective values ​​of phase current and phase voltage of each feeder in the load normalization characteristics are grouped and integrated according to the feeder number to generate the load characteristic parameter set.

4. The transformer load regulation method for an intelligent power distribution system as described in claim 3, characterized in that, By using preset electrical correlation weights, the state normalization features and the load normalization features are weighted and combined to generate the comprehensive state parameter set, including: The electrical association weight is read from a preset weight configuration table. The electrical association weight includes a first weight coefficient corresponding to the state feature component and a second weight coefficient corresponding to the load feature component. The state feature component is multiplied by the first weighting coefficient to obtain the weighted state component. The load characteristic component is multiplied by the second weighting coefficient to obtain the weighted load component; Based on the electrical structure hierarchy of the distribution transformer, the weighted state components and the weighted load components are accumulated and combined to generate the comprehensive state parameter set.

5. The transformer load regulation method for an intelligent power distribution system as described in claim 1, characterized in that, Based on the comprehensive state parameter set and the load characteristic parameter set, a multi-dimensional comprehensive evaluation of the current load state of the distribution transformer is performed to generate a load state evaluation result, including: Feature extraction is performed on the comprehensive state parameter set to obtain the real-time load rate, winding temperature state value, and insulation aging characteristic value of the distribution transformer; Feature extraction is performed on the load characteristic parameter set to obtain the total load power fluctuation characteristic value and the load current distribution characteristic value of each phase of the power distribution system; The real-time load rate is compared with a preset standard load rate range to generate a load rate deviation index for the distribution transformer. By combining the winding temperature state value and the insulation aging characteristic value, the thermal state and insulation state of the distribution transformer are analyzed in a coordinated manner to generate the health state coefficient of the distribution transformer. By using the total load power fluctuation characteristic value and the load current distribution characteristic value of each phase, the load fluctuation characteristics and three-phase load distribution characteristics of the power distribution system are correlated and analyzed to generate the load adaptability index of the power distribution system. The load rate deviation index, the health status coefficient, and the load adaptability index are comprehensively weighted and judged to generate the load status assessment result; The characteristic values ​​of the load current distribution of each phase are numerically converted to obtain the load rate of phase A, the load rate of phase B, and the load rate of phase C, and then incorporated into the load status evaluation results.

6. The transformer load regulation method for an intelligent power distribution system as described in claim 5, characterized in that, Based on the real-time load rate in the comprehensive state parameter set and the total load change rate in the load characteristic parameter set, a two-parameter trend prediction is performed on the load state assessment results to generate voltage regulation demand determination results, including: Extract the real-time load rate at the current sampling moment from the comprehensive state parameter set, and extract the total load change rate at the current sampling moment from the load characteristic parameter set; The real-time load rate is compared with a preset load rate threshold to obtain the load rate deviation value; Obtain multiple historical real-time load rates and multiple historical total load change rates corresponding to multiple historical sampling times within a historical period, and arrange the multiple historical real-time load rates and multiple historical total load change rates in chronological order to generate a historical load rate sequence and a historical change rate sequence. The load rate change gradient is obtained by differential processing between the real-time load rate and the last historical real-time load rate in the historical load rate sequence, and the total load change rate is obtained by differential processing between the total load change rate and the last historical total load change rate in the historical change rate sequence. Based on the load rate deviation value, the load rate change gradient, and the change rate change gradient, the load status assessment result is trend-corrected to obtain the corrected load trend prediction value. The corrected load trend prediction value is matched with the preset multi-level voltage regulation demand range to generate the voltage regulation demand determination result.

7. The transformer load regulation method for an intelligent power distribution system as described in claim 6, characterized in that, Based on the voltage regulation demand determination result, an operation status adjustment command for the distribution transformer is generated, including: Extract the voltage regulation direction identifier and voltage regulation amplitude requirement value from the voltage regulation requirement determination result; Obtain the current tap position number of the on-load tap changer of the distribution transformer, and determine the current voltage adjustment step size corresponding to the current tap position based on the current tap position number; The voltage regulation direction indicator and the voltage regulation amplitude requirement value are analyzed to determine the number of voltage regulation steps required for the on-load tap changer to perform. The target gear number of the on-load tap changer is determined by combining the current gear number and the number of gear change steps with a gear offset. The operating status adjustment command is generated based on the target gear number.

8. The transformer load regulation method for an intelligent power distribution system as described in claim 1, characterized in that, By analyzing the phase load imbalance of the load characteristic parameter set, the load state assessment results are subjected to phase-by-phase reactive power deficit quantification analysis, and corresponding reactive power compensation adjustment commands are generated, including: The unbalance degree is converted by the load current distribution characteristic values ​​of each phase in the load characteristic parameter set to obtain the load unbalance degree of phase A, phase B and phase C. Extract the A-phase load rate, B-phase load rate, and C-phase load rate of the distribution transformer from the load condition assessment results; The load imbalance of phase A is correlated with the load rate of phase A, the load imbalance of phase B is correlated with the load rate of phase B, and the load imbalance of phase C is correlated with the load rate of phase C to obtain the reactive power deficit characterization values ​​of phase A, phase B, and phase C. The reactive power deficit characterization values ​​of phase A, phase B, and phase C are matched with preset deficit level ranges to obtain the compensation level of phase A, phase B, and phase C. Based on the compensation levels of phase A, phase B, and phase C, corresponding reactive power compensation adjustment commands are generated.

9. The transformer load regulation method for an intelligent power distribution system as described in claim 1, characterized in that, According to the preset execution sequence, the operating state adjustment command and the reactive power compensation adjustment command are executed to control the on-load tap changer and reactive power compensation device of the distribution transformer to perform corresponding adjustment actions, including: Extract the target gear number from the operating status adjustment command, and extract the corresponding target compensation capacity identifier from the reactive power compensation adjustment command; Obtain the preset execution timing, which includes the first action time of the on-load tap changer and the second action time of the reactive power compensation device; When the first action time point is reached, the target gear number is sent to the drive controller of the on-load tap changer to drive the on-load tap changer to perform a gear switching action from the current gear to the target gear number. When the second action time point is reached, the corresponding target compensation capacity identifiers are sent to the switching controllers of the corresponding reactive power compensation branches to drive each reactive power compensation branch to perform the corresponding capacitor or reactor switching action according to the target compensation capacity identifiers. There is a preset time interval between the first action time point and the second action time point, so that the switching action of each reactive power compensation branch is executed after the on-load tap changer completes the gear switching action.

Images