Transformer energy efficiency test method and system based on intelligent compensation

Abstract

The invention discloses a transformer energy efficiency test method and system based on intelligent compensation. A plurality of low-voltage test periods are set based on a target transformer voltage range; the current I, the voltage V, the input power P1 and the output power P2 are monitored in real time in a test period, and I and V parameters are dynamically compensated through a reactive power compensation device; extracting I, V, P1 and P2 data items of each time node, and constructing a multi-cycle data set; mining strong association data items and time nodes thereof by adopting an Apriori association algorithm, and delimiting an association time period; calculating a loss value based on the correlation time period, generating a plurality of groups of correlation loss curves, performing linear fitting by taking the low-voltage test value as an independent variable and the loss value as a dependent variable, and simulating a target high-voltage test value to predict a high-voltage loss curve; according to the method, equivalent simulation of a high-voltage working condition can be realized, the test cost and the safety risk are reduced, and efficient energy efficiency evaluation and loss test optimization of the transformer are realized.

Description

Technical Field

[0001] This invention relates to the field of power analysis, and more specifically, to a method and system for testing transformer energy efficiency based on intelligent compensation. Background Technology

[0002] Transformers are a crucial component of power systems, primarily used for voltage transformation during electrical energy transmission. They convert the high-voltage current supplied to the transformer into a low-voltage current to meet the demands of different loads within the power system. Transformers not only improve the efficiency of power transmission but also prevent power loss and energy waste.

[0003] However, traditional methods for analyzing transformer losses and energy efficiency lack simulation analysis methods for low-voltage and high-voltage tests. High-voltage testing requires complex and highly integrated testing equipment, and frequent testing, analysis, and energy efficiency assessments consume significant manpower and resources, resulting in low testing efficiency. Therefore, there is an urgent need for an efficient method for testing transformer energy efficiency. Summary of the Invention

[0004] This invention overcomes the shortcomings of the prior art and proposes a transformer energy efficiency testing method and system based on intelligent compensation.

[0005] The first aspect of this invention provides a transformer energy efficiency testing method based on intelligent compensation, comprising:

[0006] S1: Based on the rated parameter range of the target transformer, set multiple low-voltage test values, and generate a test cycle based on each low-voltage test value;

[0007] S2: During a test cycle, the current I, voltage V, input power P1 and output power P2 are monitored in real time, and the parameters I and V are intelligently compensated and measured based on the reactive power compensation device.

[0008] S3: Set multiple time nodes based on the test cycle. For a time node, extract data points from parameters I, V, P1, and P2 to form a data item. Form a set of data items for one test cycle, and form multiple sets of data items based on multiple test cycles.

[0009] S4: Based on the Apriori association algorithm, analyze data items with strong associations in multiple sets of data items, mark the corresponding association time nodes, and set the association time period based on the association time nodes;

[0010] S5: Based on multiple test cycles, calculate the loss values ​​of parameters I, V, P1, and P2 corresponding to the associated time periods to obtain multiple associated loss curves. Use the low-pressure test value as the independent variable and the loss value as the dependent variable to perform linear fitting. After the fitting is completed, set the target high-pressure test value to simulate high-pressure loss prediction and generate the predicted high-pressure loss curve.

[0011] S6: Energy efficiency assessment by predicting high-voltage loss curves.

[0012] In this solution, S1 specifically refers to:

[0013] Based on the rated parameters of the target transformer, the test voltage range is set;

[0014] Set multiple low-voltage test values ​​within the test voltage range to ensure that the voltage span between the multiple low-voltage test values ​​is consistent;

[0015] Multiple test cycles are set based on multiple low-voltage test values.

[0016] In this solution, S2 specifically refers to:

[0017] During a monitoring cycle, low-voltage no-load and load tests are performed, and current I, voltage V, input power P1 and output power P2 are monitored and acquired in real time.

[0018] The reactive power compensation device dynamically tracks parameter V and calculates the real-time load rate. Based on the fluctuation of parameter V and the load rate, current compensation is performed, and the parameter status after compensation is recorded.

[0019] In this solution, S3 specifically refers to:

[0020] Based on a test cycle, N time nodes are set to ensure that the interval between adjacent time nodes is consistent. At a time node, the corresponding parameters I, V, P1, and P2 are extracted to form multi-dimensional data, and a data item is generated based on the multi-dimensional data. N data items are generated based on N time nodes and integrated into a set of data items.

[0021] Multiple sets of data items are generated based on multiple test cycles.

[0022] In this solution, S4 specifically refers to:

[0023] Based on the Apriori association algorithm, minimum support and minimum confidence are set, and multiple sets of data items are used as a dataset.

[0024] In the dataset, support is calculated based on a single data item and a combination of two data items. Frequent itemsets are selected based on the minimum support.

[0025] In the frequent itemset, the confidence score of the data items is calculated, and strong association rules are selected by using the minimum confidence score.

[0026] In strong association rules, two data items that are related are selected;

[0027] Determine whether two data items exist in different test periods. If so, mark the time nodes corresponding to the two data items to obtain the first associated time node and the second associated time node.

[0028] The first associated time node and the second associated time node are used as starting points to form associated time periods;

[0029] All associated time periods were analyzed based on strong association rules.

[0030] In this solution, S5 specifically refers to:

[0031] In one test cycle, all related time periods are extracted to calculate the loss values ​​of parameters I, V, P1, and P2. The loss values ​​are statistically analyzed to obtain the related loss curves. Multiple related loss curves are obtained based on multiple test cycles.

[0032] Within the associated time period, a time node is randomly selected, and the corresponding loss value is extracted from multiple associated loss curves to obtain multiple loss values.

[0033] The loss sequence is obtained by sorting the low-voltage test values ​​corresponding to the loss values. The low-voltage test values ​​are used as independent variables and the loss values ​​are used as dependent variables to perform linear regression fitting and obtain the fitting curve.

[0034] Set a target high-voltage test value, predict the corresponding loss value based on the target high-voltage test value in the fitted curve, and mark it as the simulated high-voltage predicted loss value;

[0035] Within the relevant time period, simulated high-voltage loss prediction is performed for each existing time node. The obtained high-voltage loss prediction values ​​are sorted and statistically analyzed to obtain the predicted high-voltage loss curve.

[0036] In this solution, S6 specifically refers to:

[0037] By predicting the high-voltage loss curve and combining it with the actual low-voltage loss results, the energy consumption of the target transformer under no-load and under-load conditions is assessed, and an energy consumption assessment report is generated.

[0038] A second aspect of the present invention also provides a transformer energy efficiency testing system based on intelligent compensation. The system includes a memory and a processor. The memory includes a transformer energy efficiency testing program based on intelligent compensation. When the processor executes the transformer energy efficiency testing program based on intelligent compensation, it performs the following steps:

[0039] S1: Based on the rated parameter range of the target transformer, set multiple low-voltage test values, and generate a test cycle based on each low-voltage test value;

[0040] S2: During a test cycle, the current I, voltage V, input power P1 and output power P2 are monitored in real time, and the parameters I and V are intelligently compensated and measured based on the reactive power compensation device.

[0041] S3: Set multiple time nodes based on the test cycle. For a time node, extract data points from parameters I, V, P1, and P2 to form a data item. Form a set of data items for one test cycle, and form multiple sets of data items based on multiple test cycles.

[0042] S4: Based on the Apriori association algorithm, analyze data items with strong associations in multiple sets of data items, mark the corresponding association time nodes, and set the association time period based on the association time nodes;

[0043] S5: Based on multiple test cycles, calculate the loss values ​​of parameters I, V, P1, and P2 corresponding to the associated time periods to obtain multiple associated loss curves. Use the low-pressure test value as the independent variable and the loss value as the dependent variable to perform linear fitting. After the fitting is completed, set the target high-pressure test value to simulate high-pressure loss prediction and generate the predicted high-pressure loss curve.

[0044] S6: Energy efficiency assessment by predicting high-voltage loss curves.

[0045] A third aspect of the present invention also provides a computer-readable storage medium comprising a transformer energy efficiency testing program based on intelligent compensation, wherein when the transformer energy efficiency testing program based on intelligent compensation is executed by a processor, the steps of the transformer energy efficiency testing method based on intelligent compensation as described in any of the preceding claims are implemented.

[0046] This invention discloses a transformer energy efficiency testing method and system based on intelligent compensation. Multiple low-voltage test cycles are set based on the target transformer voltage range. During the test cycles, current I, voltage V, input power P1, and output power P2 are monitored in real time, and I and V parameters are dynamically compensated using a reactive power compensation device. Data items for I, V, P1, and P2 at each time point are extracted to construct a multi-cycle dataset. The Apriori correlation algorithm is used to mine strongly correlated data items and their time points, defining correlated time periods. Loss values ​​are calculated based on the correlated time periods, generating multiple sets of correlated loss curves. Linear fitting is performed with low-voltage test values ​​as independent variables and loss values ​​as dependent variables, and the target high-voltage test value is simulated to predict the high-voltage loss curve. Energy efficiency assessment is completed based on the predicted curves. This invention can achieve equivalent simulation of high-voltage operating conditions, reduce testing costs and safety risks, and realize efficient energy efficiency assessment and loss testing optimization of transformers. Attached Figure Description

[0047] Figure 1 A flowchart of a transformer energy efficiency testing method based on intelligent compensation according to the present invention is shown;

[0048] Figure 2 A block diagram of a transformer energy efficiency testing system based on intelligent compensation according to the present invention is shown. Detailed Implementation

[0049] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0050] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0051] Figure 1 A flowchart of a transformer energy efficiency testing method based on intelligent compensation according to the present invention is shown.

[0052] like Figure 1 As shown, the first aspect of the present invention provides a transformer energy efficiency testing method based on intelligent compensation, comprising:

[0053] S1: Based on the rated parameter range of the target transformer, set multiple low-voltage test values, and generate a test cycle based on each low-voltage test value;

[0054] S2: During a test cycle, the current I, voltage V, input power P1 and output power P2 are monitored in real time, and the parameters I and V are intelligently compensated and measured based on the reactive power compensation device.

[0055] S3: Set multiple time nodes based on the test cycle. For a time node, extract data points from parameters I, V, P1, and P2 to form a data item. Form a set of data items for one test cycle, and form multiple sets of data items based on multiple test cycles.

[0056] S4: Based on the Apriori association algorithm, analyze data items with strong associations in multiple sets of data items, mark the corresponding association time nodes, and set the association time period based on the association time nodes;

[0057] S5: Based on multiple test cycles, calculate the loss values ​​of parameters I, V, P1, and P2 corresponding to the associated time periods to obtain multiple associated loss curves. Use the low-pressure test value as the independent variable and the loss value as the dependent variable to perform linear fitting. After the fitting is completed, set the target high-pressure test value to simulate high-pressure loss prediction and generate the predicted high-pressure loss curve.

[0058] S6: Energy efficiency assessment by predicting high-voltage loss curves.

[0059] According to an embodiment of the present invention, S1 specifically includes:

[0060] Based on the rated parameters of the target transformer, the test voltage range is set;

[0061] Set multiple low-voltage test values ​​within the test voltage range to ensure that the voltage span between the multiple low-voltage test values ​​is consistent;

[0062] Multiple test cycles are set based on multiple low-voltage test values.

[0063] It should be noted that low-voltage test values ​​are generally 220V, 380V, etc., and there are usually more than 5 low-voltage test values. High-voltage test values ​​are generally above 1000V. Each test cycle corresponds to one low-voltage test scheme.

[0064] According to an embodiment of the present invention, step S2 specifically includes:

[0065] During a monitoring cycle, low-voltage no-load and load tests are performed, and current I, voltage V, input power P1 and output power P2 are monitored and acquired in real time.

[0066] The reactive power compensation device dynamically tracks parameter V and calculates the real-time load rate. Based on the fluctuation of parameter V and the load rate, current compensation is performed, and the parameter status after compensation is recorded.

[0067] It should be noted that the no-load and load tests can effectively analyze the corresponding energy efficiency. In the no-load test, monitoring and no-load loss calculation are performed based on parameters I and V. In the load test, load rate and load loss calculation are performed based on parameters P1 and P2. When conducting no-load or load tests, if the voltage fluctuation exceeds ±2% or the load rate is >80%, current compensation can be performed through a reactive power compensation device to reduce the actual transformer transmission loss. The status of the parameters after compensation is monitored and recorded in real time, improving the accuracy of load monitoring.

[0068] According to an embodiment of the present invention, step S3 specifically includes:

[0069] Based on a test cycle, N time nodes are set to ensure that the interval between adjacent time nodes is consistent. At a time node, the corresponding parameters I, V, P1, and P2 are extracted to form multi-dimensional data, and a data item is generated based on the multi-dimensional data. N data items are generated based on N time nodes and integrated into a set of data items.

[0070] Multiple sets of data items are generated based on multiple test cycles.

[0071] It should be noted that a data item includes multi-dimensional data such as parameters I, V, P1, and P2. In subsequent correlation analysis, the correlation degree between data items can be calculated based on multi-dimensional parameters. During long testing cycles, it is possible to quickly filter out data items and related time nodes with relationships, and use the data segments corresponding to the related time nodes as the basic data segments for high-voltage simulation. This can effectively simulate the transformer state during high-voltage testing and has a certain simulation accuracy. It enables high-voltage energy efficiency assessment and prediction under low-voltage conditions, effectively improving the accuracy of transformer energy efficiency assessment, reducing the high-voltage measurement process, simplifying the testing process, and reducing the consumption of manpower and resources.

[0072] According to an embodiment of the present invention, step S4 specifically includes:

[0073] Based on the Apriori association algorithm, minimum support and minimum confidence are set, and multiple sets of data items are used as a dataset.

[0074] In the dataset, support is calculated based on a single data item and a combination of two data items. Frequent itemsets are selected based on the minimum support.

[0075] In the frequent itemset, the confidence score of the data items is calculated, and strong association rules are selected by using the minimum confidence score.

[0076] In strong association rules, two data items that are related are selected;

[0077] Determine whether two data items exist in different test periods. If so, mark the time nodes corresponding to the two data items to obtain the first associated time node and the second associated time node.

[0078] The first associated time node and the second associated time node are used as starting points to form associated time periods;

[0079] All associated time periods were analyzed based on strong association rules.

[0080] It is worth mentioning that during the low-voltage testing and loss analysis of transformers, the test parameters fluctuate due to factors such as equipment and environmental conditions. High-voltage test simulations based on the original measurement data often result in significant errors. Therefore, this invention measures and monitors parameters from multiple continuous low-voltage test schemes, uses the Apriori correlation algorithm to perform correlation analysis on parameter data items, and selects correlated time periods. Data corresponding to these correlated time periods is extracted and mined, and high-voltage simulation and loss prediction are performed on the mined data. This enables high-voltage simulation energy efficiency assessment of transformers, improving transformer testing and energy efficiency assessment efficiency, reducing the investment of manpower and resources during testing, and enhancing testing safety and efficiency.

[0081] In multi-test-cycle analysis, since the test duration is consistent, after obtaining the parameter data for each test cycle, the parameter data can be compared by matching the start and end points of the monitoring time. The associated time period is a certain period within the test cycle, which includes multiple time nodes.

[0082] According to an embodiment of the present invention, S5 specifically includes:

[0083] In one test cycle, all related time periods are extracted to calculate the loss values ​​of parameters I, V, P1, and P2. The loss values ​​are statistically analyzed to obtain the related loss curves. Multiple related loss curves are obtained based on multiple test cycles.

[0084] Within the associated time period, a time node is randomly selected, and the corresponding loss value is extracted from multiple associated loss curves to obtain multiple loss values.

[0085] The loss sequence is obtained by sorting the low-voltage test values ​​corresponding to the loss values. The low-voltage test values ​​are used as independent variables and the loss values ​​are used as dependent variables to perform linear regression fitting and obtain the fitting curve.

[0086] Set a target high-voltage test value, predict the corresponding loss value based on the target high-voltage test value in the fitted curve, and mark it as the simulated high-voltage predicted loss value;

[0087] Within the relevant time period, simulated high-voltage loss prediction is performed for each existing time node. The obtained high-voltage loss prediction values ​​are sorted and statistically analyzed to obtain the predicted high-voltage loss curve.

[0088] It should be noted that each associated time period includes multiple time nodes. Within each test cycle, each time node corresponds to four data points: parameters I, V, P1, and P2. The associated loss curves include two types: no-load loss and load loss. The appropriate curve is selected for analysis based on the energy efficiency assessment requirements. Each associated loss curve corresponds to one test cycle and also to one low-voltage test value. The loss sequence is a data sequence obtained by sorting the loss values ​​based on the sorting results of the corresponding low-voltage test values.

[0089] The correlation loss curve is a statistical curve based on the correlation time period, which includes loss values ​​corresponding to multiple time points.

[0090] According to an embodiment of the present invention, step S6 specifically includes:

[0091] By predicting the high-voltage loss curve and combining it with the actual low-voltage loss results, the energy consumption of the target transformer under no-load and under-load conditions is assessed, and an energy consumption assessment report is generated.

[0092] According to an embodiment of the present invention, it further includes:

[0093] Based on the predicted high voltage loss curve, the no-load loss curve and the load loss curve are obtained;

[0094] Based on the time nodes corresponding to the associated time periods, the loss values ​​of the corresponding time nodes are extracted from the no-load loss curve and the load loss curve, and sorted according to the time dimension to form the first loss sequence and the second loss sequence.

[0095] The correlation between the first loss sequence and the second loss sequence is determined based on the grey relational analysis method, and the correlation coefficient of the corresponding data point at each time node is calculated. If the correlation coefficient is lower than the preset value, the corresponding data points in the two sequences are removed as outliers, and the first loss sequence and the second loss sequence are updated.

[0096] Anomalies are eliminated at all time points, and the predicted high-voltage loss curve is updated based on the updated first and second loss sequences.

[0097] It should be noted that the first loss sequence and the second loss sequence correspond to the no-load loss curve and the load loss curve, respectively. The preset value is 0.1; data points below this value are considered to be unrelated. Updating the predicted high-voltage loss curve includes updating both the no-load loss curve and the load loss curve.

[0098] It is worth mentioning that in high-voltage simulation loss prediction, there is a certain correlation between no-load loss and load loss. However, the predicted loss curve often contains some noise. This invention uses the grey relational analysis method to determine the relationship state and marks data points with lower correlation states as outliers for removal. This effectively filters out abnormal data points in high-voltage prediction, thereby obtaining effective predicted loss values ​​and improving the accuracy and efficiency of energy efficiency assessment.

[0099] Figure 2 A block diagram of a transformer energy efficiency testing system based on intelligent compensation according to the present invention is shown.

[0100] A second aspect of the present invention also provides a transformer energy efficiency testing system 2 based on intelligent compensation. The system includes a memory 21 and a processor 22. The memory 21 includes a transformer energy efficiency testing program based on intelligent compensation. When the processor 22 executes the transformer energy efficiency testing program based on intelligent compensation, it performs the following steps:

[0101] S1: Based on the rated parameter range of the target transformer, set multiple low-voltage test values, and generate a test cycle based on each low-voltage test value;

[0102] S2: During a test cycle, the current I, voltage V, input power P1 and output power P2 are monitored in real time, and the parameters I and V are intelligently compensated and measured based on the reactive power compensation device.

[0103] S3: Set multiple time nodes based on the test cycle. For a time node, extract data points from parameters I, V, P1, and P2 to form a data item. Form a set of data items for one test cycle, and form multiple sets of data items based on multiple test cycles.

[0104] S4: Based on the Apriori association algorithm, analyze data items with strong associations in multiple sets of data items, mark the corresponding association time nodes, and set the association time period based on the association time nodes;

[0105] S5: Based on multiple test cycles, calculate the loss values ​​of parameters I, V, P1, and P2 corresponding to the associated time periods to obtain multiple associated loss curves. Use the low-pressure test value as the independent variable and the loss value as the dependent variable to perform linear fitting. After the fitting is completed, set the target high-pressure test value to simulate high-pressure loss prediction and generate the predicted high-pressure loss curve.

[0106] S6: Energy efficiency assessment by predicting high-voltage loss curves.

[0107] According to an embodiment of the present invention, S1 specifically includes:

[0108] Based on the rated parameters of the target transformer, the test voltage range is set;

[0109] Set multiple low-voltage test values ​​within the test voltage range to ensure that the voltage span between the multiple low-voltage test values ​​is consistent;

[0110] Multiple test cycles are set based on multiple low-voltage test values.

[0111] It should be noted that low-voltage test values ​​are generally 220V, 380V, etc., and there are usually more than 5 low-voltage test values. High-voltage test values ​​are generally above 1000V. Each test cycle corresponds to one low-voltage test scheme.

[0112] According to an embodiment of the present invention, step S2 specifically includes:

[0113] During a monitoring cycle, low-voltage no-load and load tests are performed, and current I, voltage V, input power P1 and output power P2 are monitored and acquired in real time.

[0114] The reactive power compensation device dynamically tracks parameter V and calculates the real-time load rate. Based on the fluctuation of parameter V and the load rate, current compensation is performed, and the parameter status after compensation is recorded.

[0115] It should be noted that the no-load and load tests can effectively analyze the corresponding energy efficiency. In the no-load test, monitoring and no-load loss calculation are performed based on parameters I and V. In the load test, load rate and load loss calculation are performed based on parameters P1 and P2. When conducting no-load or load tests, if the voltage fluctuation exceeds ±2% or the load rate is >80%, current compensation can be performed through a reactive power compensation device to reduce the actual transformer transmission loss. The status of the parameters after compensation is monitored and recorded in real time, improving the accuracy of load monitoring.

[0116] According to an embodiment of the present invention, step S3 specifically includes:

[0117] Based on a test cycle, N time nodes are set to ensure that the interval between adjacent time nodes is consistent. At a time node, the corresponding parameters I, V, P1, and P2 are extracted to form multi-dimensional data, and a data item is generated based on the multi-dimensional data. N data items are generated based on N time nodes and integrated into a set of data items.

[0118] Multiple sets of data items are generated based on multiple test cycles.

[0119] It should be noted that a data item includes multi-dimensional data such as parameters I, V, P1, and P2. In subsequent correlation analysis, the correlation degree between data items can be calculated based on multi-dimensional parameters. During long testing cycles, it is possible to quickly filter out data items and related time nodes with relationships, and use the data segments corresponding to the related time nodes as the basic data segments for high-voltage simulation. This can effectively simulate the transformer state during high-voltage testing and has a certain simulation accuracy. It enables high-voltage energy efficiency assessment and prediction under low-voltage conditions, effectively improving the accuracy of transformer energy efficiency assessment, reducing the high-voltage measurement process, simplifying the testing process, and reducing the consumption of manpower and resources.

[0120] According to an embodiment of the present invention, step S4 specifically includes:

[0121] Based on the Apriori association algorithm, minimum support and minimum confidence are set, and multiple sets of data items are used as a dataset.

[0122] In the dataset, support is calculated based on a single data item and a combination of two data items. Frequent itemsets are selected based on the minimum support.

[0123] In the frequent itemset, the confidence score of the data items is calculated, and strong association rules are selected by using the minimum confidence score.

[0124] In strong association rules, two data items that are related are selected;

[0125] Determine whether two data items exist in different test periods. If so, mark the time nodes corresponding to the two data items to obtain the first associated time node and the second associated time node.

[0126] The first associated time node and the second associated time node are used as starting points to form associated time periods;

[0127] All associated time periods were analyzed based on strong association rules.

[0128] It is worth mentioning that during the low-voltage testing and loss analysis of transformers, the test parameters fluctuate due to factors such as equipment and environmental conditions. High-voltage test simulations based on the original measurement data often result in significant errors. Therefore, this invention measures and monitors parameters from multiple continuous low-voltage test schemes, uses the Apriori correlation algorithm to perform correlation analysis on parameter data items, and selects correlated time periods. Data corresponding to these correlated time periods is extracted and mined, and high-voltage simulation and loss prediction are performed on the mined data. This enables high-voltage simulation energy efficiency assessment of transformers, improving transformer testing and energy efficiency assessment efficiency, reducing the investment of manpower and resources during testing, and enhancing testing safety and efficiency.

[0129] In multi-test-cycle analysis, since the test duration is consistent, after obtaining the parameter data for each test cycle, the parameter data can be compared by matching the start and end points of the monitoring time. The associated time period is a certain period within the test cycle, which includes multiple time nodes.

[0130] According to an embodiment of the present invention, S5 specifically includes:

[0131] In one test cycle, all related time periods are extracted to calculate the loss values ​​of parameters I, V, P1, and P2. The loss values ​​are statistically analyzed to obtain the related loss curves. Multiple related loss curves are obtained based on multiple test cycles.

[0132] Within the associated time period, a time node is randomly selected, and the corresponding loss value is extracted from multiple associated loss curves to obtain multiple loss values.

[0133] The loss sequence is obtained by sorting the low-voltage test values ​​corresponding to the loss values. The low-voltage test values ​​are used as independent variables and the loss values ​​are used as dependent variables to perform linear regression fitting and obtain the fitting curve.

[0134] Set a target high-voltage test value, predict the corresponding loss value based on the target high-voltage test value in the fitted curve, and mark it as the simulated high-voltage predicted loss value;

[0135] Within the relevant time period, simulated high-voltage loss prediction is performed for each existing time node. The obtained high-voltage loss prediction values ​​are sorted and statistically analyzed to obtain the predicted high-voltage loss curve.

[0136] It should be noted that each associated time period includes multiple time nodes. Within each test cycle, each time node corresponds to four data points: parameters I, V, P1, and P2. The associated loss curves include two types: no-load loss and load loss. The appropriate curve is selected for analysis based on the energy efficiency assessment requirements. Each associated loss curve corresponds to one test cycle and also to one low-voltage test value. The loss sequence is a data sequence obtained by sorting the loss values ​​based on the sorting results of the corresponding low-voltage test values.

[0137] The correlation loss curve is a statistical curve based on the correlation time period, which includes loss values ​​corresponding to multiple time points.

[0138] According to an embodiment of the present invention, step S6 specifically includes:

[0139] By predicting the high-voltage loss curve and combining it with the actual low-voltage loss results, the energy consumption of the target transformer under no-load and under-load conditions is assessed, and an energy consumption assessment report is generated.

[0140] A third aspect of the present invention also provides a computer-readable storage medium comprising a transformer energy efficiency testing program based on intelligent compensation, wherein when the transformer energy efficiency testing program based on intelligent compensation is executed by a processor, the steps of the transformer energy efficiency testing method based on intelligent compensation as described in any of the preceding claims are implemented.

[0141] This invention discloses a transformer energy efficiency testing method and system based on intelligent compensation. Multiple low-voltage test cycles are set based on the target transformer voltage range. During the test cycles, current I, voltage V, input power P1, and output power P2 are monitored in real time, and I and V parameters are dynamically compensated using a reactive power compensation device. Data items for I, V, P1, and P2 at each time point are extracted to construct a multi-cycle dataset. The Apriori correlation algorithm is used to mine strongly correlated data items and their time points, defining correlated time periods. Loss values ​​are calculated based on the correlated time periods, generating multiple sets of correlated loss curves. Linear fitting is performed with low-voltage test values ​​as independent variables and loss values ​​as dependent variables, and the target high-voltage test value is simulated to predict the high-voltage loss curve. Energy efficiency assessment is completed based on the predicted curves. This invention can achieve equivalent simulation of high-voltage operating conditions, reduce testing costs and safety risks, and realize efficient energy efficiency assessment and loss testing optimization of transformers.

[0142] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0143] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0144] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0145] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0146] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0147] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A transformer energy efficiency testing method based on intelligent compensation, characterized in that, include: S1: Based on the rated parameter range of the target transformer, set multiple low-voltage test values, and generate a test cycle based on each low-voltage test value; S2: During a test cycle, the current I, voltage V, input power P1 and output power P2 are monitored in real time, and the parameters I and V are intelligently compensated and measured based on the reactive power compensation device. S3: Set multiple time nodes based on the test cycle. For a time node, extract data points from parameters I, V, P1, and P2 to form a data item. Form a set of data items for one test cycle, and form multiple sets of data items based on multiple test cycles. S4: Based on the Apriori association algorithm, analyze data items with strong associations in multiple sets of data items, mark the corresponding association time nodes, and set the association time period based on the association time nodes; S5: Based on multiple test cycles, calculate the loss values ​​of parameters I, V, P1, and P2 corresponding to the associated time periods to obtain multiple associated loss curves. Use the low-pressure test value as the independent variable and the loss value as the dependent variable to perform linear fitting. After the fitting is completed, set the target high-pressure test value to simulate high-pressure loss prediction and generate the predicted high-pressure loss curve. S6: Energy efficiency assessment by predicting high-voltage loss curves.

2. The transformer energy efficiency testing method based on intelligent compensation according to claim 1, characterized in that, Specifically, S1 is: Based on the rated parameters of the target transformer, the test voltage range is set; Set multiple low-voltage test values ​​within the test voltage range to ensure that the voltage span between the multiple low-voltage test values ​​is consistent; Multiple test cycles are set based on multiple low-voltage test values.

3. The transformer energy efficiency testing method based on intelligent compensation according to claim 1, characterized in that, Specifically, S2 is: During a monitoring cycle, low-voltage no-load and load tests are performed, and current I, voltage V, input power P1 and output power P2 are monitored and acquired in real time. The reactive power compensation device dynamically tracks parameter V and calculates the real-time load rate. Based on the fluctuation of parameter V and the load rate, current compensation is performed, and the parameter status after compensation is recorded.

4. The transformer energy efficiency testing method based on intelligent compensation according to claim 1, characterized in that, Specifically, S3 is: Based on a test cycle, N time nodes are set to ensure that the interval between adjacent time nodes is consistent. At a time node, the corresponding parameters I, V, P1, and P2 are extracted to form multi-dimensional data, and a data item is generated based on the multi-dimensional data. N data items are generated based on N time nodes and integrated into a set of data items. Multiple sets of data items are generated based on multiple test cycles.

5. The transformer energy efficiency testing method based on intelligent compensation according to claim 1, characterized in that, Specifically, S4 is: Based on the Apriori association algorithm, minimum support and minimum confidence are set, and multiple sets of data items are used as a dataset. In the dataset, support is calculated based on a single data item and a combination of two data items. Frequent itemsets are selected based on the minimum support. In the frequent itemset, the confidence score of the data items is calculated, and strong association rules are selected by using the minimum confidence score. In strong association rules, two data items that are related are selected; Determine whether two data items exist in different test periods. If so, mark the time nodes corresponding to the two data items to obtain the first associated time node and the second associated time node. The first associated time node and the second associated time node are used as starting points to form associated time periods; All associated time periods were analyzed based on strong association rules.

6. The transformer energy efficiency testing method based on intelligent compensation according to claim 1, characterized in that, Specifically, S5 is: In one test cycle, all related time periods are extracted to calculate the loss values ​​of parameters I, V, P1, and P2. The loss values ​​are statistically analyzed to obtain the related loss curves. Multiple related loss curves are obtained based on multiple test cycles. Within the associated time period, a time node is randomly selected, and the corresponding loss value is extracted from multiple associated loss curves to obtain multiple loss values. The loss sequence is obtained by sorting the low-voltage test values ​​corresponding to the loss values. The low-voltage test values ​​are used as independent variables and the loss values ​​are used as dependent variables to perform linear regression fitting and obtain the fitting curve. Set a target high-voltage test value, predict the corresponding loss value based on the target high-voltage test value in the fitted curve, and mark it as the simulated high-voltage predicted loss value; Within the relevant time period, simulated high-voltage loss prediction is performed for each existing time node. The obtained high-voltage loss prediction values ​​are sorted and statistically analyzed to obtain the predicted high-voltage loss curve.

7. The transformer energy efficiency testing method based on intelligent compensation according to claim 1, characterized in that, Specifically, S6 is: By predicting the high-voltage loss curve and combining it with the actual low-voltage loss results, the energy consumption of the target transformer under no-load and under-load conditions is assessed, and an energy consumption assessment report is generated.

8. A transformer energy efficiency testing system based on intelligent compensation, characterized in that, The system includes a memory and a processor. The memory includes a transformer energy efficiency testing program based on intelligent compensation. When the processor executes the transformer energy efficiency testing program based on intelligent compensation, it performs the following steps: S1: Based on the rated parameter range of the target transformer, set multiple low-voltage test values, and generate a test cycle based on each low-voltage test value; S2: During a test cycle, the current I, voltage V, input power P1 and output power P2 are monitored in real time, and the parameters I and V are intelligently compensated and measured based on the reactive power compensation device. S3: Set multiple time nodes based on the test cycle. For a time node, extract data points from parameters I, V, P1, and P2 to form a data item. Form a set of data items for one test cycle, and form multiple sets of data items based on multiple test cycles. S4: Based on the Apriori association algorithm, analyze data items with strong associations in multiple sets of data items, mark the corresponding association time nodes, and set the association time period based on the association time nodes; S5: Based on multiple test cycles, calculate the loss values ​​of parameters I, V, P1, and P2 corresponding to the associated time periods to obtain multiple associated loss curves. Use the low-pressure test value as the independent variable and the loss value as the dependent variable to perform linear fitting. After the fitting is completed, set the target high-pressure test value to simulate high-pressure loss prediction and generate the predicted high-pressure loss curve. S6: Energy efficiency assessment by predicting high-voltage loss curves.

9. A transformer energy efficiency testing system based on intelligent compensation according to claim 8, characterized in that, Specifically, S1 is: Based on the rated parameters of the target transformer, the test voltage range is set; Set multiple low-voltage test values ​​within the test voltage range to ensure that the voltage span between the multiple low-voltage test values ​​is consistent; Multiple test cycles are set based on multiple low-voltage test values.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a transformer energy efficiency testing program based on intelligent compensation. When the transformer energy efficiency testing program based on intelligent compensation is executed by a processor, it implements the steps of the transformer energy efficiency testing method based on intelligent compensation as described in any one of claims 1 to 7.

Images