A method for monitoring humidity inside a 35kV box-type transformer

Abstract

The present invention relates to the field of humidity monitoring technology, and discloses a humidity monitoring method for the interior of a 35kV box-type transformer. The method includes establishing multiple wind distribution models based on wind speed data and wind speed data; selecting a corresponding wind distribution model based on real-time environmental parameters, and setting a humidity threshold for each monitoring point based on the selected wind distribution model; collecting the primary humidity value of each humidity monitoring point based on a sampling time node; correcting the primary humidity value of each humidity monitoring point based on environmental data, and generating a secondary humidity value based on the correction result; generating a total humidity evaluation value inside the transformer based on the secondary humidity value and the humidity threshold of each humidity monitoring point; and generating a control instruction based on the total humidity evaluation value inside the transformer. The method improves the accuracy of humidity monitoring, realizes intelligent monitoring and early warning, optimizes the allocation of monitoring resources, and provides a visual humidity management tool through a digital twin model.

Description

Technical Field

[0001] The present invention relates to the technical field of humidity monitoring, and in particular to a method for monitoring humidity inside a 35kV box-type transformer. Background Art

[0002] With the development of smart grids and automated monitoring technologies, real-time monitoring of environmental parameters for power equipment such as transformers has become an industry trend. Advances in humidity monitoring technology allow for more precise tracking and control of humidity levels within transformers, thereby preventing potential failures and extending equipment life.

[0003] 35kV box-type transformers are common power equipment, widely used in urban distribution networks and industrial power supply systems. Monitoring internal humidity is crucial to ensuring safe operation of transformers. Excessive humidity can cause aging, corrosion, and short circuits in insulation materials, thereby affecting transformer performance and lifespan. Therefore, effective monitoring and control of internal humidity in box-type transformers is a crucial measure to ensure power system stability. Summary of the Invention

[0004] The purpose of the present invention is to accurately monitor the humidity changes inside the transformer box by establishing a wind distribution model and a humidity change model, and to detect potential humidity anomalies in advance through an early warning mechanism that combines multiple thresholds with change rate judgment, thereby effectively ensuring the safe operation of the box-type transformer.

[0005] In order to achieve the above object, the present invention provides a method for monitoring humidity inside a 35kV box-type transformer, comprising:

[0006] Based on the wind speed data and the wind speed data, multiple wind distribution models are established, the corresponding wind distribution model is selected according to the real-time environmental parameters, and the humidity threshold of each monitoring point is set according to the selected wind distribution model;

[0007] The transformer internal humidity sampling period is set according to the real-time environmental parameters, and multiple sampling time nodes are generated, where the sampling time node is the starting point of the sampling period;

[0008] The first-level humidity value of each humidity monitoring point is collected according to the sampling time node, and the first-level humidity value of each humidity monitoring point is corrected according to the real-time environmental parameters, and the second-level humidity value is generated according to the correction result;

[0009] Based on the secondary humidity values ​​and humidity thresholds of each humidity monitoring point, a total evaluation value of the internal humidity of the transformer is generated, and a control instruction is generated according to the total evaluation value of the internal humidity of the transformer.

[0010] In some embodiments of the present invention, the establishing of multiple wind distribution models includes:

[0011] Obtain historical data on wind speed and direction entering the box-type transformer;

[0012] The historical data of wind speed entering the box-type transformer are divided into grade intervals to generate a wind speed grade dataset W, where W = [W1, W2…Wi…Wn];

[0013] Based on historical wind direction data, a wind direction interval dataset AC is established for the interior of the box-type transformer, where AC = [AC1, AC2…ACi…ACm]

[0014] Where Wi represents the i-th wind speed level, ACi represents the i-th wind direction interval, n represents the number of wind speed levels, and m represents the number of wind direction intervals;

[0015] Input the wind speed level Wi and wind direction range ACi inside the box-type transformer, and collect the wind speed level at each monitoring point inside the box-type transformer;

[0016] Generate a wind distribution model Mi when the input wind speed level is Wi and the wind direction range is ACi.

[0017] In some embodiments of the present invention, the collecting of the primary humidity value of each humidity monitoring point includes:

[0018] Select the corresponding wind distribution model according to the real-time environmental parameters, and set the humidity threshold of each monitoring point according to the selected wind distribution model;

[0019] Monitor the real-time data of humidity changes at the current monitoring point and divide the monitored real-time data of humidity changes into intervals;

[0020] Evaluate the humidity of each interval and generate the first-level humidity value set H of the current sampling period, where H = [H1, H2…Hi…Hn];

[0021] Where Hi represents the humidity evaluation value of the i-th monitoring point in the current sampling period; n represents the total number of monitoring points.

[0022] In some embodiments of the present invention, generating the secondary humidity value includes:

[0023] Obtain historical environmental data outside the box-type transformer and historical humidity data at each monitoring point inside the box-type transformer;

[0024] Evaluate the external historical environmental data of the box-type transformer to obtain the external environment score S, and obtain the evaluation data set H of each monitoring point inside the box-type transformer under the wind distribution model Mi;

[0025] Calculate the variance D(H) of the evaluation data set H at each monitoring point inside the box-type transformer;

[0026] Based on the variance D(H) of the evaluation data set H, a first humidity compensation data set C is generated, where C = [C1, C2…Ci…Cn];

[0027] Wherein, Ci represents the first humidity compensation value of the i-th monitoring point, and n represents the number of monitoring points;

[0028] The first humidity compensation data set C is the compensation value of the internal humidity monitoring data under the wind distribution model Mi when the external environment score S is used.

[0029] Correcting the primary humidity value set H based on the first humidity compensation data set C, and generating a second humidity real-time change data set D according to the correction result, where D = [D1, D2…Di…Dn];

[0030] Wherein, Di represents the second real-time temperature data value of the i-th monitoring point, and n represents the number of monitoring points.

[0031] In some embodiments of the present invention, the step of selecting and setting the transformer internal humidity sampling period according to the real-time environmental parameters includes:

[0032] Based on historical data, set the humidity change threshold B, B = [B1, B2, B3];

[0033] Wherein, B1 represents the first threshold value of humidity change, B2 represents the second threshold value of humidity change, and B3 represents the third threshold value of humidity change;

[0034] Count the number of humidity changes exceeding the threshold on a certain date, and based on the statistical results, generate the humidity change degree index k for the current date, k = n1 + 2*(n2-n1) + 3*(n3-n2-n1);

[0035] Where n1 represents the number of humidity changes on the current date that exceed the first threshold, n2 represents the number of humidity changes on the current date that exceed the second threshold, and n3 represents the number of humidity changes on the current date that exceed the third threshold;

[0036] Based on the humidity change index k of the current date, generate the humidity change index dataset K of the current year, K = [k1, k2…ki…kn];

[0037] Among them, ki represents the humidity change index value on the i-th day, and n represents the number of days;

[0038] Based on the value of ki, the levels are divided and the length of the sampling period is adjusted.

[0039] In some embodiments of the present invention, adjusting the duration of the sampling period includes:

[0040] Based on the humidity change index value ki on the i-th day, the level is divided, the average value μ and standard deviation σ of ki are obtained, and multiple types of monitoring periods are generated;

[0041] If ki∈(0,μ-σ), it is the first type of monitoring period, and the monitoring frequency is set to a*(1+b);

[0042] If ki∈(μ-σ,μ+σ), it is the second type of monitoring period, and the monitoring frequency is set to a;

[0043] If ki∈(μ+σ,∞), it is the third type of detection period, and the monitoring frequency is set to a*(1-b);

[0044] Where a is the preset sampling period, and b is the proportional coefficient.

[0045] In some embodiments of the present invention, generating a total evaluation value of the internal humidity of the transformer based on the secondary humidity value and the humidity threshold of each humidity monitoring point includes:

[0046] Obtain the second humidity real-time change data set D of each humidity monitoring point at the current sampling time node, where D = [D1, D2…Di…Dn];

[0047] And determine the humidity threshold A of each monitoring point according to the selected wind distribution model, A=[A1,A2…Ai…An];

[0048] Where Ai represents the humidity threshold of the i-th monitoring point, and n represents the number of monitoring points;

[0049] Combined with the humidity threshold A of each monitoring point and the second humidity real-time change dataset D of each humidity monitoring point at the current sampling time node, an evaluation value E of each humidity monitoring point at the current sampling time node is generated, and a transformer control strategy is generated based on the evaluation value of each humidity monitoring point at the current sampling time node.

[0050] In some embodiments of the present invention, generating a transformer control strategy includes:

[0051] Calculate the difference F between the second humidity real-time change data set D of each humidity monitoring point at the current sampling time node and the humidity threshold A of each monitoring point, where F = [F1, F2…Fi…Fn];

[0052] Fi=Di-Ai;

[0053] Fi represents the difference between the second humidity real-time change data set Di of each humidity monitoring point at the current sampling time node and the humidity threshold Ai of each monitoring point;

[0054] If Fi is a positive number, the current humidity monitoring point is marked as 1;

[0055] If Fi is a negative number, the current humidity monitoring point is marked as 0;

[0056] Count the sum of the marks of each humidity monitoring point at the current sampling time node to generate the evaluation value E of each humidity monitoring point at the current sampling time node;

[0057] Based on the evaluation value E of each humidity monitoring point at the current sampling time node; generate the proportion V of each humidity monitoring point exceeding the threshold;

[0058] V = E / n;

[0059] Where n represents the total number of monitoring points;

[0060] Generate a transformer control strategy based on the value of V.

[0061] In some embodiments of the present invention, generating a transformer control strategy based on the value of V includes:

[0062] Generate a multi-stage transformer control strategy based on the value of V;

[0063] If V>v1, then generate the first-level transformer control strategy;

[0064] If V>v2; then generate the secondary transformer control strategy;

[0065] If V>v3, a three-level transformer control strategy is generated;

[0066] Among them, v1, v2, and v3 are constants set according to the internal historical data of the box-type transformer.

[0067] In some embodiments of the present invention, the multi-stage transformer control strategy includes:

[0068] Primary transformer control strategy: controls the monitoring cycle and generates alarm instructions;

[0069] Secondary transformer control strategy: controls the monitoring period and generates dehumidification instructions;

[0070] Three-level transformer control strategy: generating shutdown commands.

[0071] Compared with the prior art, the humidity monitoring method for a 35kV box-type transformer according to the embodiment of the present invention has the following advantages:

[0072] By establishing a wind distribution model through the box opening parameters and box structure parameters, we can accurately grasp the impact of external wind on the humidity distribution inside the box, determine the reasonable layout of humidity monitoring points under different wind speed conditions, and thus improve the accuracy of humidity monitoring.

[0073] By constructing a wind distribution model, important basic data is provided for subsequent humidity monitoring, equipment operation status evaluation, etc.

[0074] By obtaining the humidity evaluation value of each monitoring point, the humidity distribution inside the transformer can be fully reflected. This helps operation and maintenance personnel to promptly identify monitoring points with abnormal humidity so that targeted measures can be taken, such as adjusting the operating parameters of the dehumidification equipment or conducting key inspections in specific areas, thereby ensuring the normal operation of the transformer.

[0075] The humidity change index k for the current date is generated by counting the number of humidity changes exceeding the threshold on a certain date. The exceedance of different thresholds is comprehensively considered and calculated according to the importance of different thresholds, which can more comprehensively and accurately reflect the overall degree of humidity change on that day.

[0076] By generating evaluation values ​​for each humidity monitoring point, a quantitative assessment of the humidity condition inside the transformer from multiple monitoring points to the whole is achieved; this helps operation and maintenance personnel to accurately understand the overall humidity condition inside the transformer.

[0077] By monitoring humidity changes in real time and dynamically adjusting control strategies based on monitoring data, precise humidity monitoring and targeted control strategies can effectively maintain the humidity inside the transformer within an appropriate range, thereby ensuring the insulation performance of the transformer and improving the reliability and stability of the entire power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 This is a flow chart of a humidity monitoring method inside a 35kV box-type transformer disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0079] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0080] In the description of this application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on this application.

[0081] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. Throughout this application, unless otherwise specified, "plurality" means two or more.

[0082] In the description of this application, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal connections between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0083] like Figure 1 As shown, the embodiment of the present invention discloses a method for monitoring humidity inside a 35kV box-type transformer, comprising:

[0084] Based on the wind speed data and the wind speed data, multiple wind distribution models are established, the corresponding wind distribution model is selected according to the real-time environmental parameters, and the humidity threshold of each monitoring point is set according to the selected wind distribution model;

[0085] The transformer internal humidity sampling period is set according to the real-time environmental parameters, and multiple sampling time nodes are generated, where the sampling time node is the starting point of the sampling period;

[0086] The first-level humidity value of each humidity monitoring point is collected according to the sampling time node, and the first-level humidity value of each humidity monitoring point is corrected according to the real-time environmental parameters, and the second-level humidity value is generated according to the correction result;

[0087] Based on the secondary humidity values ​​and humidity thresholds of each humidity monitoring point, a total evaluation value of the internal humidity of the transformer is generated, and a control instruction is generated according to the total evaluation value of the internal humidity of the transformer.

[0088] In this embodiment, the wind distribution model is created:

[0089] Input parameters include the size of the enclosure openings and the enclosure structure (e.g., material thermal conductivity, gap size, and location). CFD (Computational Fluid Dynamics) simulations are performed using fluid dynamics software (e.g., ANSYS Fluent) to generate wind speed distribution maps, taking into account the effects of changes in wind speed and direction on airflow.

[0090] Combined with the wind distribution model, the impact of air flow on the humidity inside the box is considered.

[0091] The model needs to consider the relationship between temperature, humidity and material hygroscopicity, and use an appropriate physical model to describe the diffusion process of water molecules.

[0092] Match the real-time wind speed data with the pre-established wind speed level dataset. For example, if the wind speed level dataset is divided into 0-2m / s as W1, 2-5m / s as W2, 5-10m / s as W3, etc., when the real-time wind speed is 3m / s, it is matched to W2.

[0093] Taking into account the fluctuation of wind speed, if the wind speed is near the boundary of adjacent levels (such as around 2m / s), the weighted average method can be used to determine a more appropriate level.

[0094] The quantified wind direction data is matched with the established wind direction interval dataset. For example, if the wind direction is northeast by east, and the wind direction interval dataset contains the northeast by east interval (such as AC3), the wind direction interval is determined.

[0095] If the wind direction is at the boundary of two intervals, the interval to which it belongs is determined based on the historical trend of the wind direction or the set priority rules.

[0096] Select a wind distribution model

[0097] Based on the matched wind speed level and wind direction range, the corresponding model is selected from the established multiple wind distribution models. For example, if the wind speed level is W2 and the wind direction range is AC3, the corresponding Mij wind distribution model (i represents the wind speed level number and j represents the wind direction range number) is selected.

[0098] When establishing a wind distribution model, a set of initial humidity thresholds is pre-set for each monitoring point under each wind distribution model. These initial thresholds can be determined based on transformer design parameters, operating experience, and relevant standards. For example, for a monitoring point under a specific wind distribution model, the lower threshold for relative humidity can be set at 30% and the upper threshold at 70%.

[0099] Dynamic Adjustment

[0100] The initial humidity threshold is dynamically adjusted based on real-time environmental parameter changes and the transformer's operating status. If the temperature rises, the saturated humidity of the air will increase, and the upper limit of the humidity threshold can be appropriately raised. Conversely, if the temperature drops, the upper limit of the humidity threshold can be appropriately lowered.

[0101] Consider the load inside the transformer. If the load is high, the transformer will generate more heat, which may affect the humidity distribution. In this case, the humidity threshold may also need to be fine-tuned. For example, if the load increases by 20%, the upper and lower limits of the humidity threshold can be adjusted by ±5%.

[0102] Special situation handling

[0103] In the event of extreme weather conditions (such as high humidity after heavy rain or snowstorms) or abnormal conditions within the transformer (such as localized leakage), the humidity threshold is adjusted according to pre-set rules. For example, after a rainstorm, the upper limit of the humidity threshold is raised to 80%, while humidity changes are closely monitored. Once the humidity shows a downward trend and approaches the normal threshold range, the humidity is gradually restored to the normal threshold setting.

[0104] Example 1:

[0105] The establishment of multiple wind distribution models includes:

[0106] Obtain historical data on wind speed and direction entering the box-type transformer;

[0107] The historical data of wind speed entering the box-type transformer are divided into grade intervals to generate a wind speed grade dataset W, where W = [W1, W2…Wi…Wn];

[0108] Based on historical wind direction data, a wind direction interval dataset AC is established for the interior of the box-type transformer, where AC = [AC1, AC2…ACi…ACm]

[0109] Where Wi represents the i-th wind speed level, ACi represents the i-th wind direction interval, n represents the number of wind speed levels, and m represents the number of wind direction intervals;

[0110] Input the wind speed level Wi and wind direction range ACi inside the box-type transformer, and collect the wind speed level at each monitoring point inside the box-type transformer;

[0111] Generate a wind distribution model Mi when the input wind speed level is Wi and the wind direction range is ACi.

[0112] In this embodiment, historical wind speed and direction data is obtained by installing an anemometer and a wind vane at the entrance of the box-type transformer. This allows us to obtain historical wind speed and direction data entering the box-type transformer. For example, we collected data continuously for one year, with a collection interval of every 10 minutes, and obtained a large number of wind speed and direction numerical records.

[0113] Wind speed level interval division: divide the wind speed historical data into level intervals according to actual needs; for example, divide the wind speed into 5 levels:

[0114] 0-2m / s is W1 (breeze level); 2-5m / s is W2 (light breeze level); 5-10m / s is W3 (moderate wind level); 10-15m / s is W4 (strong wind level);

[0115] Above 15m / s is W5 (hurricane level); thus, a wind speed level data set W = [W1, W2, W3, W4, W5] is generated;

[0116] Establishment of wind direction interval dataset. For historical wind direction data, we establish wind direction interval dataset AC. For example, taking 45° as an interval, the 360° wind direction is divided into 8 intervals:

[0117] 0-45° is AC1; 45-90° is AC2; 90-135° is AC3; 135-180° is AC4; 180-225° is AC5; 225-270° is AC6; 270-315° is AC7; 315-360° is AC8;

[0118] So AC=[AC1,AC2,AC3,AC4,AC5,AC6,AC7,AC8];

[0119] Internal wind speed data collection: When the wind speed level inside the box-type transformer is W3 (medium wind level, 5-10m / s) and the wind direction range is AC3 (90-135°), we collect wind speed data at various monitoring points inside the box-type transformer (for example, at key locations such as the perimeter, top, and bottom of the transformer). Assuming there are 10 monitoring points, the collected wind speed levels may vary due to the influence of the box-type transformer's internal structure.

[0120] Wind force distribution model generation: Based on the wind speed data collected at each monitoring point, we can generate a wind force distribution model M3 for the input wind speed level W3 and wind direction range AC3. This model can be represented by a matrix or function, such as a 10×10 matrix. The elements in the matrix represent the wind speed relationship or energy transfer relationship between each monitoring point, reflecting the wind force distribution inside the box-type transformer under this specific wind speed and direction input.

[0121] Example 2:

[0122] The collection of the primary humidity value of each humidity monitoring point includes:

[0123] Select the corresponding wind distribution model according to the real-time environmental parameters, and set the humidity threshold of each monitoring point according to the selected wind distribution model;

[0124] Monitor the real-time data of humidity changes at the current monitoring point and divide the monitored real-time data of humidity changes into intervals;

[0125] Evaluate the humidity of each interval and generate the first-level humidity value set H of the current sampling period, where H = [H1, H2…Hi…Hn];

[0126] Where Hi represents the humidity evaluation value of the i-th monitoring point in the current sampling period; n represents the total number of monitoring points.

[0127] In this embodiment, the wind distribution model is selected

[0128] In a large industrial plant, multiple devices are operating, and humidity monitoring points are distributed throughout different areas. Real-time environmental parameters include wind speed (5 m / s) and wind direction (120°). Based on these real-time environmental parameters, we selected the previously established wind distribution model M3 (which corresponds to wind speeds of 5-10 m / s and directions of 90-135°).

[0129] Humidity threshold setting

[0130] Based on the selected wind distribution model M3, we set humidity thresholds for each humidity monitoring point. For example, for monitoring point 1, located in a corner of the factory, where wind has a relatively small impact, we set a humidity threshold of ±3% as normal, with any change outside this range a cause for concern. For monitoring point 2, located in the center of the factory near a ventilation opening, where wind may cause increased water vapor exchange, we set a humidity threshold of ±5% as normal.

[0131] Humidity data monitoring

[0132] A humidity sensor is installed at each humidity monitoring point, and real-time humidity data is monitored with a sampling period of 10 minutes. For example, the real-time humidity data at monitoring point 1 within a sampling period may start at 50%, rise to 52%, then fall to 49%, and so on.

[0133] Interval division and humidity evaluation

[0134] For monitoring point 1, we divide the humidity range into intervals. For example, 48%-52% is considered normal, while values ​​below 48% or above 52% are considered abnormal. We then evaluate humidity based on the distribution of the real-time humidity data within these intervals. If the majority of the data falls within the normal range, the humidity evaluation value H1 is high, assuming it is 80 (out of 100). If a significant amount of data falls within the abnormal range, the humidity evaluation value is low.

[0135] We operate in this way for all n monitoring points. For example, the humidity evaluation value of monitoring point 2 is H2, the humidity evaluation value of monitoring point 3 is H3, and so on. Finally, the first-level humidity value set H = [H1, H2, H3...Hi...Hn] for the current sampling period is generated.

[0136] Example 3:

[0137] The generation of the secondary humidity value includes:

[0138] Obtain historical environmental data outside the box-type transformer and historical humidity data at each monitoring point inside the box-type transformer;

[0139] Evaluate the external historical environmental data of the box-type transformer to obtain the external environment score S, and obtain the evaluation data set H of each monitoring point inside the box-type transformer under the wind distribution model Mi;

[0140] Calculate the variance D(H) of the evaluation data set H at each monitoring point inside the box-type transformer;

[0141] Based on the variance D(H) of the evaluation data set H, a first humidity compensation data set C is generated, where C = [C1, C2…Ci…Cn];

[0142] Wherein, Ci represents the first humidity compensation value of the i-th monitoring point, and n represents the number of monitoring points;

[0143] The first humidity compensation data set C is the compensation value of the internal humidity monitoring data under the wind distribution model Mi when the external environment score S is used.

[0144] Correcting the primary humidity value set H based on the first humidity compensation data set C, and generating a second humidity real-time change data set D according to the correction result, where D = [D1, D2…Di…Dn];

[0145] Wherein, Di represents the second real-time temperature data value of the i-th monitoring point, and n represents the number of monitoring points.

[0146] Through these steps, the humidity dataset D' more accurately reflects the actual humidity changes inside the box, improving the reliability and accuracy of the monitoring system. This correction process fully utilizes historical data and model predictions, effectively reducing measurement errors caused by environmental factors.

[0147] In this example, historical environmental data for the box-type transformer's exterior is obtained from a weather station, including temperature, wind speed, rainfall, and other data from the past year. This data is recorded hourly. Simultaneously, historical humidity data from the past year is obtained from humidity sensors installed at various monitoring points (assuming there are five monitoring points) inside the box-type transformer.

[0148] External environment score

[0149] Evaluate the historical external environmental data of the box-type transformer. For example, if the external temperature is relatively stable (fluctuations within ±5°C), the wind speed is low (average wind speed less than 3m / s), and the rainfall is low (monthly rainfall less than 50mm), the external environment score S = 80 (out of 100). If the external environment fluctuates significantly, the score will be reduced accordingly.

[0150] Internal evaluation dataset acquisition

[0151] Based on the current wind distribution model Mi (assuming it is the previously mentioned M3 model), obtain the evaluation dataset H for each monitoring point inside the box-type transformer. For example, under the M3 model, for five monitoring points, the evaluation dataset H = [H1 = 70, H2 = 75, H3 = 80, H4 = 65, H5 = 72] obtained using the previous humidity evaluation method is obtained.

[0152] Based on the variance D(H) of the evaluation dataset H, the first humidity compensation dataset C is generated. Assuming that we have an empirically established functional relationship, when D(H) = 21.84, the first humidity compensation dataset C generated for the five monitoring points is C = [C1 = 2, C2 = 3, C3 = 1, C4 = 4, C5 = 2].

[0153] Humidity dataset correction: The primary humidity value set H is corrected based on the first humidity compensation dataset C. For monitoring point 1, the corrected humidity evaluation value D1 = H1 + C1 = 70 + 2 = 72; for monitoring point 2, D2 = H2 + C2 = 75 + 3 = 78; and so on.

[0154] Based on the correction results, a second real-time humidity change dataset D = [D1 = 72, D2 = 78, D3 = 81, D4 = 69, D5 = 74] was generated. This dataset D more accurately reflects the actual humidity changes inside the cabinet. By considering factors such as the external environment score, the variance of the internal humidity data, and the humidity compensation value, the reliability and accuracy of the humidity monitoring system were effectively improved.

[0155] Example 4:

[0156] The step of selecting and setting the transformer internal humidity sampling period according to the real-time environmental parameters includes:

[0157] Based on historical data, set the humidity change threshold B, B = [B1, B2, B3];

[0158] Wherein, B1 represents the first threshold value of humidity change, B2 represents the second threshold value of humidity change, and B3 represents the third threshold value of humidity change;

[0159] Count the number of humidity changes exceeding the threshold on a certain date, and based on the statistical results, generate the humidity change degree index k for the current date, k = n1 + 2*(n2-n1) + 3*(n3-n2-n1);

[0160] Where n1 represents the number of humidity changes on the current date that exceed the first threshold, n2 represents the number of humidity changes on the current date that exceed the second threshold, and n3 represents the number of humidity changes on the current date that exceed the third threshold;

[0161] Based on the humidity change index k of the current date, generate the humidity change index dataset K of the current year, K = [k1, k2…ki…kn];

[0162] Among them, ki represents the humidity change index value on the i-th day, and n represents the number of days;

[0163] Based on the value of ki, the levels are divided and the length of the sampling period is adjusted.

[0164] In this embodiment, the humidity change threshold setting stage

[0165] Set the humidity change threshold B based on historical data. Suppose we analyze the humidity data inside the box-type transformer over the past five years and divide the humidity change threshold into three levels.

[0166] B1 (the first threshold value of humidity change) is set to 5%, that is, when the humidity change exceeds 5% within a sampling period, it is considered to have reached the first threshold value.

[0167] B2 (the second humidity change threshold) is set to 10%, indicating that the second threshold is reached when the humidity change exceeds 10%.

[0168] B3 (the third humidity change threshold) is set to 15%. If the humidity changes by more than 15%, the third threshold is reached.

[0169] Data statistics and indicator calculation

[0170] Count the number of humidity changes exceeding the threshold in a day. Assume that the humidity sensor collects data once an hour, for a total of 24 times a day.

[0171] By analyzing the data, we found that the number of times n1 the humidity changed exceeded the first threshold (5%) was 5. The number of times n2 the humidity changed exceeded the second threshold (10%) was 3 (n2 includes some of the cases in n1, because exceeding the second threshold necessarily exceeds the first threshold). The number of times n3 the humidity changed exceeded the third threshold (15%) was 1 (again, n3 includes some of the cases in n2 and n1).

[0172] Calculate the humidity change index k on the current date according to the formula:

[0173] k = n1 + 2*(n2 - n1) + 3*(n3 - n2 - n1)

[0174] Substitute the values:

[0175] k = 5 + 2*(3 - 5) + 3*(1 - 3 - 5)

[0176] k = 5 + 2*(-2) + 3*(-7)

[0177] k = 5 - 4 - 21 = -20

[0178] Dataset generation

[0179] According to the above method, calculate for each day of the whole year to obtain the dataset K of the humidity change degree index for the current year. Assume that October 1st is the i-th day, and the calculated ki = -20. As the index values of each day of the whole year are calculated, finally K = [k1, k2... ki... kn], where n is 365;

[0180] Grade division and sampling period adjustment

[0181] Divide the values of ki into grades. For example, we divide the values of ki into three grades:

[0182] When ki ≥ 0, it is considered that the humidity change degree is high, and the sampling period is shortened from the original 1 hour to 30 minutes to more timely monitor the humidity change.

[0183] When -10 < ki < 0, it is considered that the humidity change degree is medium, and the sampling period remains unchanged at 1 hour.

[0184] When ki ≤ -10, it is considered that the humidity change degree is low, and the sampling period is extended to 2 hours, which can reduce unnecessary data collection and also monitor the humidity change within an acceptable accuracy range.

[0185] Example 5:

[0186] The duration of adjusting the sampling period includes:

[0187] Based on the humidity change degree index value ki of the i-th date, conduct grade division, obtain the average value μ and standard deviation σ of ki, and generate multiple types of monitoring time periods;

[0188] If ki ∈ (0, μ - σ), it is the first type of monitoring time period, and the monitoring frequency is set to a*(1 + b);

[0189] If ki ∈ (μ - σ, μ + σ), it is the second type of monitoring time period, and the monitoring frequency is set to a;

[0190] If ki∈(μ+σ,∞), it is the third type of detection period, and the monitoring frequency is set to a*(1-b);

[0191] Where a is the preset sampling period, and b is the proportional coefficient.

[0192] Example 6:

[0193] The method of generating a total evaluation value of the internal humidity of the transformer based on the secondary humidity value and the humidity threshold of each humidity monitoring point includes:

[0194] Obtain the second humidity real-time change data set D of each humidity monitoring point at the current sampling time node, where D = [D1, D2…Di…Dn];

[0195] And determine the humidity threshold A of each monitoring point according to the selected wind distribution model, A=[A1,A2…Ai…An];

[0196] Where Ai represents the humidity threshold of the i-th monitoring point, and n represents the number of monitoring points;

[0197] Combined with the humidity threshold A of each monitoring point and the second humidity real-time change dataset D of each humidity monitoring point at the current sampling time node, an evaluation value E of each humidity monitoring point at the current sampling time node is generated, and a transformer control strategy is generated based on the evaluation value of each humidity monitoring point at the current sampling time node.

[0198] In this embodiment:

[0199] Assuming the current sampling time is 10:00 AM on October 28, 2024, we obtain a second real-time humidity change dataset D for each humidity monitoring point. For example, if there are five humidity monitoring points inside a box-type transformer, then D = [D1 = 60, D2 = 65, D3 = 58, D4 = 62, D5 = 55], where the numerical value represents the humidity value at each monitoring point (assuming the unit is %).

[0200] The humidity threshold A of each monitoring point is determined based on the selected wind distribution model (assuming it is the M3 model mentioned above). Due to factors such as the location of the monitoring point and its relationship with the ventilation opening, for these five monitoring points:

[0201] A1 (humidity threshold of monitoring point 1) is set to a humidity value of 58-62% as normal, with an evaluation value of 80 (out of 100). 5 points will be deducted for every 1% deviation outside this range.

[0202] A2 (humidity threshold of monitoring point 2) is set to a humidity value of 60-65% as normal, with an evaluation value of 85. 3 points will be deducted for every 1% deviation outside the range.

[0203] A3 (humidity threshold of monitoring point 3) is set to a humidity value of 55-60% as normal, with an evaluation value of 90. 4 points will be deducted for every 1% deviation outside the range.

[0204] A4 (humidity threshold of monitoring point 4) is set to a humidity value of 59-63% as normal, with an evaluation value of 82. 2 points will be deducted for every 1% deviation outside the range.

[0205] A4 (humidity threshold of monitoring point 5) is set to a humidity value of 53-58% as normal, with an evaluation value of 88. 3 points will be deducted for every 1% deviation outside the range.

[0206] For monitoring point 1, D1=60. According to the evaluation standard of A1, the humidity value is within the normal range, so E1=80.

[0207] At monitoring point 2, D2=65. The normal range of A2 is 60-65%. The humidity value is just on the boundary, so E2=85.

[0208] At monitoring point 3, D3=58, the normal range of A3 is 55-60%, the humidity value is within the normal range, and E3=90.

[0209] The D4 of monitoring point 4 is 62, the normal range of A4 is 59-63%, the humidity value is within the normal range, and E4 is 82.

[0210] The D5 of monitoring point 5 is 55, the normal range of A5 is 53-58%, the humidity value is within the normal range, and E5 is 88.

[0211] Example 7:

[0212] The generation of the transformer control strategy includes:

[0213] Calculate the difference F between the second humidity real-time change data set D of each humidity monitoring point at the current sampling time node and the humidity threshold A of each monitoring point, where F = [F1, F2…Fi…Fn];

[0214] Fi=Di-Ai;

[0215] Fi represents the difference between the second humidity real-time change data set Di of each humidity monitoring point at the current sampling time node and the humidity threshold Ai of each monitoring point;

[0216] If Fi is a positive number, the current humidity monitoring point is marked as 1;

[0217] If Fi is a negative number, the current humidity monitoring point is marked as 0;

[0218] Count the sum of the marks of each humidity monitoring point at the current sampling time node to generate the evaluation value E of each humidity monitoring point at the current sampling time node;

[0219] Based on the evaluation value E of each humidity monitoring point at the current sampling time node; generate the proportion V of each humidity monitoring point exceeding the threshold;

[0220] V = E / n;

[0221] Where n represents the total number of monitoring points;

[0222] Generate a transformer control strategy based on the value of V.

[0223] Example 8:

[0224] The transformer control strategy is generated based on the value of V, including:

[0225] Generate a multi-stage transformer control strategy based on the value of V;

[0226] If V>v1, then generate the first-level transformer control strategy;

[0227] If V>v2; then generate the secondary transformer control strategy;

[0228] If V>v3, a three-level transformer control strategy is generated;

[0229] Among them, v1, v2, and v3 are constants set according to the internal historical data of the box-type transformer.

[0230] Example 9:

[0231] The multi-stage transformer control strategy includes:

[0232] Primary transformer control strategy: controls the monitoring cycle and generates alarm instructions;

[0233] Secondary transformer control strategy: controls the monitoring period and generates dehumidification instructions;

[0234] Three-level transformer control strategy: generating shutdown commands.

[0235] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention. These improvements and substitutions should also be regarded as the scope of protection of the present invention.

Claims

1. A humidity monitoring method inside a 35kV box-type transformer, characterized in that: include: Establish multiple wind distribution models based on wind speed and direction data, select the corresponding wind distribution model according to real-time environmental parameters, and set the humidity threshold of each monitoring point according to the selected wind distribution model; The internal humidity sampling period of the transformer is set according to the real-time environmental parameters, including: setting a humidity change threshold B based on historical data, B = [B1, B2, B3]; wherein B1 represents the first humidity change threshold, B2 represents the second humidity change threshold, and B3 represents the third humidity change threshold; counting the number of humidity changes exceeding the threshold on a certain date, and generating a humidity change index k for the current date based on the statistical results, k = n1 + 2 * (n2 - n1) + 3 * (n3 - n2 - n1); wherein n1 represents the number of humidity changes exceeding the first threshold on the current date, n2 represents the number of humidity changes exceeding the second threshold on the current date, and n3 represents the number of humidity changes exceeding the third threshold on the current date; based on the humidity change index k on the current date, generating a humidity change index dataset K for the current year, K = [ k1, k2…ki…kn]; wherein ki represents the humidity change index value on the i-th date, and n represents the number of dates; based on the value of ki, the level division is performed, and the duration of the sampling period is adjusted, including: based on the humidity change index value ki on the i-th date, the level division is performed, the average value μ and standard deviation σ of ki are obtained, and multiple types of monitoring periods are generated; if ki∈(0, μ-σ), it is the first type of monitoring period, and the monitoring frequency is set to a*(1+b); if ki∈(μ-σ, μ+σ), it is the second type of monitoring period, and the monitoring frequency is set to a; if ki∈(μ+σ,∞), it is the third type of detection period, and the monitoring frequency is set to a*(1-b); wherein a is the preset sampling period duration, and b is the proportional coefficient; and multiple sampling time nodes are generated, wherein the sampling time node is the starting point of the sampling period; The first-level humidity value of each humidity monitoring point is collected according to the sampling time node, and the first-level humidity value of each humidity monitoring point is corrected according to the real-time environmental parameters, and the second-level humidity value is generated according to the correction result, including: obtaining the external historical environmental data of the box-type transformer and the historical humidity data of each monitoring point inside the box-type transformer; evaluating the external historical environmental data of the box-type transformer to obtain the external environmental score S, and obtaining the evaluation data set H of each monitoring point inside the box-type transformer under the wind distribution model Mi; calculating the variance D(H) of the evaluation data set H of each monitoring point inside the box-type transformer; and based on the variance of the evaluation data set H D(H) generates a first humidity compensation dataset C, where C = [C1, C2…Ci…Cn]; where Ci represents the first humidity compensation value of the i-th monitoring point, and n represents the number of monitoring points. The first humidity compensation dataset C is the compensation value of the internal humidity monitoring data under the wind distribution model Mi when the external environment score is S. The primary humidity value set H is corrected based on the first humidity compensation dataset C, and a second humidity real-time change dataset D is generated based on the correction result, where D = [D1, D2…Di…Dn]; where Di represents the second real-time humidity data value of the i-th monitoring point, and n represents the number of monitoring points. Based on the secondary humidity values ​​and humidity thresholds of each humidity monitoring point, a total evaluation value of the internal humidity of the transformer is generated, and a control instruction is generated according to the total evaluation value of the internal humidity of the transformer.

2. The humidity monitoring method inside the 35kV box-type transformer according to claim 1, characterized in that: The collection of the primary humidity value of each humidity monitoring point includes: Select the corresponding wind distribution model according to the real-time environmental parameters, and set the humidity threshold of each monitoring point according to the selected wind distribution model; Monitor the real-time data of humidity changes at the current monitoring point and divide the monitored real-time data of humidity changes into intervals; Evaluate the humidity of each interval and generate the first-level humidity value set H of the current sampling period, where H = [H1, H2…Hi…Hn]; Where Hi represents the humidity evaluation value of the i-th monitoring point in the current sampling period; n represents the total number of monitoring points.

3. The humidity monitoring method inside the 35kV box-type transformer according to claim 1, characterized in that: The method of generating a total evaluation value of the internal humidity of the transformer based on the secondary humidity value and the humidity threshold of each humidity monitoring point includes: Obtain the second humidity real-time change data set D of each humidity monitoring point at the current sampling time node, where D = [D1, D2…Di…Dn]; And determine the humidity threshold A of each monitoring point according to the selected wind distribution model, A=[A1, A2…Ai…An]; Where Ai represents the humidity threshold of the i-th monitoring point, and n represents the number of monitoring points; Combined with the humidity threshold A of each monitoring point and the second humidity real-time change dataset D of each humidity monitoring point at the current sampling time node, an evaluation value E of each humidity monitoring point at the current sampling time node is generated, and a transformer control strategy is generated based on the evaluation value of each humidity monitoring point at the current sampling time node.

4. The humidity monitoring method inside the 35kV box-type transformer according to claim 1, characterized in that: The generation of the transformer control strategy includes: Calculate the difference F between the second humidity real-time change data set D of each humidity monitoring point at the current sampling time node and the humidity threshold A of each monitoring point, where F = [F1, F2...Fi...Fn]; Fi=Di-Ai; Fi represents the difference between the second humidity real-time change data set Di of each humidity monitoring point at the current sampling time node and the humidity threshold Ai of each monitoring point; If Fi is a positive number, the current humidity monitoring point is marked as 1; If Fi is a negative number, the current humidity monitoring point is marked as 0; Count the sum of the marks of each humidity monitoring point at the current sampling time node to generate the evaluation value E of each humidity monitoring point at the current sampling time node; Based on the evaluation value E of each humidity monitoring point at the current sampling time node; generate the proportion V of each humidity monitoring point exceeding the threshold; V=E / n; Where n represents the total number of monitoring points; Generate a transformer control strategy based on the value of V.

5. The humidity monitoring method inside the 35kV box-type transformer according to claim 4, characterized in that: The transformer control strategy is generated based on the value of V, including: Generate a multi-stage transformer control strategy based on the value of V; If V>v1, then generate the first-level transformer control strategy; If V>v2; then generate the secondary transformer control strategy; If V>v3, a three-level transformer control strategy is generated; Among them, v1, v2, and v3 are constants set according to the internal historical data of the box-type transformer.

6. The humidity monitoring method inside the 35kV box-type transformer according to claim 5, characterized in that: The multi-stage transformer control strategy includes: Primary transformer control strategy: controls the monitoring cycle and generates alarm instructions; Secondary transformer control strategy: controls the monitoring period and generates dehumidification instructions; Three-level transformer control strategy: generating shutdown commands.

Images