Box transformer substation temperature monitoring method and system based on wireless sensing

Abstract

The invention relates to the field of box transformer substation temperature monitoring, in particular to a box transformer substation temperature monitoring method and system based on wireless sensing. A box transformer substation temperature monitoring system based on wireless sensing comprises a risk value acquisition module, an interference value acquisition module and a temperature monitoring module. By adopting the wireless sensing technology, the system can comprehensively cover all temperature monitoring points in the box-type substation, real-time and continuous monitoring of temperature data is achieved, the system can also calculate the comprehensive risk value of each monitoring point, the monitoring frequency is dynamically adjusted according to the comprehensive risk value of each monitoring point, and optimal distribution of resources is achieved; meanwhile, for the monitoring point with the comprehensive risk value larger than the preset deviation value, whether abnormity exists or not can be further judged based on the difference value between the ideal temperature and the actual temperature, the accuracy of temperature data is improved, and the intelligent level of box transformer substation temperature monitoring is greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of box-type transformer temperature monitoring, and in particular to a box-type transformer temperature monitoring method and system based on wireless sensing. Background Art

[0002] With the rapid development of industrialization and urbanization, the stability and reliability of the power system, as an important part of the infrastructure, are crucial to ensuring the normal operation of social and economic activities; the box-type substation (abbreviated as box-type substation) is a key equipment in the power system, responsible for converting high-voltage electricity into a voltage level suitable for civil and commercial use. The normal operating temperature of its internal equipment is directly related to the stability and safety of the power supply.

[0003] In traditional box-type transformer temperature monitoring methods, limited sensors are usually used for temperature monitoring. These methods have the following limitations: due to the influence of environmental factors such as electromagnetic interference, dust, ventilation status and vibration, the initial temperature data collected may have errors, affecting the accuracy of temperature monitoring; at the same time, the fixed monitoring frequency cannot be dynamically adjusted according to actual temperature changes and environmental conditions, which may lead to waste of resources or insufficient monitoring; on the other hand, there is a lack of comprehensive consideration of external environmental factors, and it is impossible to predict the ideal temperature, so that the temperature change trend cannot be effectively predicted and adjusted.

[0004] In order to overcome the above limitations and improve the accuracy, real-time and intelligence level of box-type transformer temperature monitoring, it is necessary to develop a box-type transformer temperature monitoring method and system based on wireless sensing. Summary of the invention

[0005] By adopting wireless sensing technology, the present invention can fully cover all temperature monitoring points in the box-type substation and realize real-time and continuous monitoring of temperature data. The system can also calculate the comprehensive risk value of each monitoring point and dynamically adjust the monitoring frequency according to the comprehensive risk value of the monitoring point to achieve optimal allocation of resources; at the same time, for the monitoring points whose comprehensive risk value is greater than the preset deviation value, it will further judge whether there is an abnormality based on the difference between the ideal temperature and the actual temperature, thereby improving the accuracy of the temperature data and greatly improving the intelligent level of box-type substation temperature monitoring.

[0006] A box transformer temperature monitoring method based on wireless sensing, comprising:

[0007] For any temperature monitoring point in the box-type substation, collect initial temperature data at the monitoring time point of the current temperature monitoring point; perform filtering operations on the collected initial temperature data to obtain accurate temperature data; determine whether the accurate temperature data is within the normal temperature range, if so, no operation; if not, execute alarm response measures;

[0008] At the beginning of any preset update cycle, the electromagnetic interference intensity of the current box-type substation is obtained, and the dust level, ventilation status and vibration intensity of each temperature monitoring point are obtained respectively. Based on the obtained electromagnetic interference intensity and the dust level, ventilation status and vibration intensity of each temperature monitoring point, the comprehensive interference value of each temperature monitoring point is calculated; at the same time, the current external environment temperature, external environment humidity and load of the box-type substation are obtained, and based on the accurate temperature data of each temperature monitoring point obtained in the most recent preset update cycle, the comprehensive risk value of each temperature monitoring point is calculated;

[0009] Set the comprehensive interference threshold. For any temperature monitoring point, if the comprehensive interference value of the current temperature monitoring point is lower than the comprehensive interference threshold, the first-level filtering operation is performed on the initial temperature data collected at the current temperature monitoring point; if the comprehensive interference value of the current temperature monitoring point is higher than or equal to the comprehensive interference threshold, the second-level filtering operation is performed on the initial temperature data collected at the current temperature monitoring point;

[0010] For any temperature monitoring point, the comprehensive risk value of the current temperature monitoring point is used to calculate the temperature monitoring frequency within the preset temperature monitoring frequency adjustment range, and the calculated temperature monitoring frequency is used to collect initial temperature data within the current preset update cycle; a comprehensive risk threshold is set. If the comprehensive risk value of the current temperature monitoring point is higher than the comprehensive risk threshold, the external ambient temperature, external ambient humidity and load are obtained at the current temperature prediction time point, and the obtained external ambient temperature, external ambient humidity and load are input into the ideal temperature model, and the ideal temperature data at the current temperature prediction time point is output; the currently obtained accurate temperature data is compared with the ideal temperature data. If the difference between the accurate temperature data and the ideal temperature data is greater than the preset deviation value, the alarm response measures are executed.

[0011] Preferably, the comprehensive interference value of each temperature monitoring point is calculated, and the specific operation is as follows:

[0012] Based on the obtained electromagnetic interference strength EMI, set the electromagnetic interference threshold EMI threshold ; If the currently acquired electromagnetic interference intensity EMI is less than the electromagnetic interference threshold EMI threshold , then I EMI 0; if the electromagnetic interference intensity EMI is greater than or equal to the electromagnetic interference threshold EMI threshold , then use the formula Calculate the electromagnetic interference value I EMI ;

[0013] Based on the dust level D(i), ventilation status V(i) and vibration intensity Vib(i) of each temperature monitoring point, i=1, 2, ..., n; n represents the number of temperature monitoring points;

[0014] For dust level D(i), if the dust level D(i) at the temperature monitoring point is less than the preset dust level threshold D threshold , then the dust interference value I(i) D is 0 if the dust level D(i) is greater than or equal to the dust level threshold D threshold , then use the formula Calculate and obtain the dust interference value I(i) D ;

[0015] For ventilation status V(i): If the ventilation status V(i) of the temperature monitoring point is within the ideal ventilation range V min To V max If the ventilation interference value is I(i) V is 0; if the ventilation state V(i) is less than or equal to V min , then use the formula Calculate and obtain the ventilation interference value I(i) V ; If the ventilation state V(i) is greater than or equal to V max , then use the formula Calculate and obtain the ventilation interference value I(i) V ;

[0016] For the vibration intensity Vib(i), if the vibration intensity Vib(i) of the current temperature monitoring point is less than the preset vibration intensity threshold Vib threshold , then the vibration disturbance value I(i) Vib is 0; if the vibration intensity Vib(i) is greater than or equal to the vibration intensity threshold Vib threshold , then use the formula Calculate and obtain the vibration disturbance value I(i) Vib ;

[0017] Finally, the formula I(i) = α1I EMI +α2(I(i) D +I(i) V +I(i) Vib ) Calculate and obtain the comprehensive interference value I(i) of each temperature monitoring point; where α1 and α2 are weight coefficients, α1>α2.

[0018] Preferably, the comprehensive risk value of each temperature monitoring point is calculated and obtained, and the specific operation is as follows:

[0019] Based on the external ambient temperature T ext , external ambient humidity H and load L, for external ambient temperature T ext , set the external reference temperature T ref and the maximum acceptable temperature T max ; When the external environment temperature Text Less than or equal to T ref When the ambient temperature risk value R T is 0; when the external environment temperature T ext Located in T ref With T max When the Calculate the ambient temperature risk value R T ; When the external environment temperature T ext Higher than or equal to T max When the ambient temperature risk value R T is 1;

[0020] For the external environment humidity H, set the external reference humidity H ref , using the formula Calculate the environmental humidity risk value R H ; where k is the control coefficient, which is used to control the external environment humidity H to be higher than H ref After that, the environmental humidity risk value R H The rising speed;

[0021] For load L, set the load upper limit L max , using the formula Calculate the load risk value R L ;

[0022] For any temperature monitoring point, based on the accurate temperature data T obtained at the current temperature monitoring point within the most recent preset update period j , j = 1, 2, ..., m; m represents the number of accurate temperature data obtained; using the formula Calculate and obtain the temperature standard deviation σ of the current temperature monitoring point, where is the average value of accurate temperature data; set the normal temperature standard deviation σ under normal conditions ref and the maximum temperature standard deviation σ max , if the temperature standard deviation σ is less than or equal to σ ref , then the temperature fluctuation risk value R σ is 0; if the temperature standard deviation σ is within σ ref and σ max Between, use the formula Calculate the temperature fluctuation risk value Rσ of the current temperature monitoring point; if the temperature standard deviation σ is greater than or equal to σ max , then the temperature fluctuation risk value R σ 1; at the same time, calculate the difference ΔT between the current accurate temperature data and the accurate temperature data obtained at the beginning of the most recent period, based on the preset update period duration t, using the formula Calculate the temperature change rate V ΔT; If the temperature change rate V ΔT If the temperature change rate is lower than the preset threshold, the temperature change risk value R ΔT is 0; if the temperature change rate V ΔT If the temperature change rate is higher than or equal to the preset temperature change rate threshold, the temperature change risk value RΔT is 1; using the formula The comprehensive risk value R of the current temperature monitoring point is calculated, where β1 and β2 are weight coefficients, β1<β2; finally, the comprehensive risk value R(i) of each temperature monitoring point is obtained.

[0023] Preferably, the temperature monitoring frequency is calculated and obtained, and the specific operations are as follows:

[0024] Set the temperature monitoring frequency adjustment range to F min To F max Based on the comprehensive risk value R(i) of all temperature monitoring points, the formula F(i) = F min +R(i)(F max -F min ) calculates and obtains the temperature monitoring frequency applied at each temperature monitoring point in the current time period.

[0025] Preferably, the ideal temperature model is established based on a random forest model, and the specific operations of training the ideal temperature model are as follows:

[0026] Obtain historical temperature data, and use the historical temperature data under normal conditions as training samples, where each training sample contains several historical temperature data and the corresponding external environment temperature, external environment humidity and load; divide all training samples into a training set and a validation set, use the external environment temperature, external environment humidity and load as the input of the ideal temperature model, and use the historical temperature data as the output of the ideal temperature model, and input the training set into the ideal temperature model with parameter initialization for training; use the validation set to validate the ideal temperature model and obtain the validation result; determine whether the validation result meets the preset training conditions, and if so, output the trained ideal temperature model; if not, use the training set to continue training the ideal temperature model.

[0027] Preferably, the filtering method adopted by the first-level filtering operation is sliding mean filtering; and the filtering method adopted by the second-level filtering operation is Kalman filtering.

[0028] Preferably, the comprehensive interference threshold is selected by a particle swarm optimization algorithm, and the specific operation is as follows:

[0029] Step 1: Set the maximum number of iterations and randomly generate a particle swarm. The position of each particle represents a candidate threshold.

[0030] Step 2: For any particle, initialize the velocity of the particle and use the current position as the individual optimal position of the particle; use the formula Calculate the fitness value F of the particle, where S represents the size of the candidate threshold, and w represents the deviation rate of temperature monitoring using the current candidate threshold; the larger the candidate threshold, the smaller the deviation rate, and the larger the fitness value of the candidate threshold;

[0031] Step 3: Traverse all particles and take the position corresponding to the particle with the highest fitness value F as the global optimal position;

[0032] Step 4: For any particle, update the speed and position of the particle, and recalculate the fitness value; if the updated fitness value is higher than the fitness value of the individual best position of the particle, then update the individual best position of the particle to the current position;

[0033] Step 5: Traverse all particles. If the fitness value of the individual best position of any particle is higher than the fitness value of the global best position, the global best position is updated to the individual best position of the particle.

[0034] Step 6: Repeat steps 3 to 5 until the maximum number of iterations is reached. The global best position obtained is the optimal comprehensive interference threshold.

[0035] Preferably, a temperature monitoring system for a box-type transformer based on wireless sensing includes:

[0036] The risk value acquisition module is used to obtain the external environment temperature, external environment humidity and load of the box-type substation, and calculate the comprehensive risk value of each temperature monitoring point based on the accurate temperature data obtained at each temperature monitoring point within the latest preset update cycle;

[0037] The interference value acquisition module is used to obtain the electromagnetic interference intensity of the box-type substation, and at the same time obtain the dust level, ventilation status and vibration intensity of each temperature monitoring point respectively, and calculate the comprehensive interference value of each temperature monitoring point based on the obtained electromagnetic interference intensity and the dust level, ventilation status and vibration intensity of each temperature monitoring point;

[0038] The temperature monitoring module includes a frequency adjustment unit, a filtering unit, an ideal temperature acquisition unit and a temperature monitoring unit; the frequency adjustment unit is used to calculate and obtain the temperature monitoring frequency within a preset temperature monitoring frequency adjustment range using the comprehensive risk value of the temperature monitoring point, and use the calculated temperature monitoring frequency to collect initial temperature data within the current preset update cycle; the filtering unit is used to perform a first-level filtering operation on the initial temperature data collected from the temperature monitoring point when the comprehensive interference value of the temperature monitoring point is lower than the comprehensive interference threshold; when the comprehensive interference value of the temperature monitoring point is higher than or equal to the comprehensive interference threshold, perform a second-level filtering operation on the initial temperature data collected from the current temperature monitoring point; the ideal temperature acquisition unit is used to perform a second-level filtering operation on the initial temperature data collected from the current temperature monitoring point when the comprehensive risk value of the temperature monitoring point is higher than When the comprehensive risk threshold is reached, the external environment temperature, external environment humidity and load are obtained at the temperature prediction time point, and the obtained external environment temperature, external environment humidity and load are input into the ideal temperature model to output the ideal temperature data at the current temperature prediction time point; the temperature monitoring unit is used to collect initial temperature data at the monitoring time point of each temperature monitoring point; a filtering operation is performed on the collected initial temperature data to obtain accurate temperature data; it is determined whether the accurate temperature data is within the normal temperature range, if so, no operation is performed; if not, an alarm response measure is executed; at the same time, when the ideal temperature data exists, the currently acquired accurate temperature data is compared with the ideal temperature data, and if the difference between the accurate temperature data and the ideal temperature data is greater than the preset deviation value, an alarm response measure is executed.

[0039] The present invention has the following advantages:

[0040] 1. By adopting wireless sensing technology, the system of the present invention can comprehensively cover all temperature monitoring points in the box-type substation and realize real-time and continuous monitoring of temperature data. The system can also calculate the comprehensive risk value of each monitoring point and dynamically adjust the monitoring frequency according to the comprehensive risk value of the monitoring point to achieve optimal allocation of resources; at the same time, for the monitoring points whose comprehensive risk value is greater than the preset deviation value, it will further judge whether there is an abnormality based on the difference between the ideal temperature and the actual temperature, thereby improving the accuracy of the temperature data and greatly improving the intelligent level of box-type substation temperature monitoring.

[0041] 2. The present invention combines advanced filtering algorithms and comprehensively considers various environmental factors such as electromagnetic interference, dust level, ventilation status and vibration intensity to calculate the comprehensive interference value of each monitoring point, and selects appropriate filtering strategies accordingly. The system can effectively reduce environmental interference and improve the accuracy of temperature data. This method of comprehensively considering multiple factors enables the system to better adapt to different environmental conditions and improves the accuracy and reliability of monitoring. In addition, the particle swarm optimization algorithm adopted by the system can automatically select the optimal comprehensive interference threshold, further enhancing the adaptability and flexibility of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 The figure is a schematic diagram of the structure of a temperature monitoring system for a box-type transformer based on wireless sensing adopted in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0044] Embodiment 1, a box-type transformer temperature monitoring method based on wireless sensing, comprising:

[0045] For any temperature monitoring point in the box-type substation, the initial temperature data is collected at the monitoring time point of the current temperature monitoring point. These data are original and unprocessed and may be affected by environmental factors; the collected initial temperature data is filtered. The collected initial temperature data may contain noise and errors, so it is necessary to eliminate these interferences through filtering operations to obtain accurate temperature data; determine whether the accurate temperature data is within the normal temperature range. This range is determined based on the safe operation requirements of the box-type substation equipment and historical data experience. If so, no operation is performed; if not, alarm response measures are executed. These measures may include issuing sound and light alarms, sending notifications to operation and maintenance personnel, starting the cooling system, etc., to control the temperature within a safe range as soon as possible;

[0046] At the beginning of any preset update cycle, the electromagnetic interference intensity of the current box-type substation is obtained, and the dust level, ventilation status and vibration intensity of each temperature monitoring point are obtained respectively; there are a large number of electrical equipment inside the box-type substation, which will generate electromagnetic interference during operation; electromagnetic interference may affect the signal transmission of the sensor, resulting in deviations in the collected temperature data; dust will accumulate on the surface of the sensor, affecting the heat conduction between the sensor and the object being measured, resulting in inaccurate temperature readings; the ventilation status directly affects the heat exchange efficiency inside the box-type substation, and poor ventilation will cause heat accumulation, making the temperature reading too high or too low; vibration may cause poor contact between the sensor and the object being measured, or cause fatigue damage to the internal components of the sensor, thereby affecting the stability and accuracy of the temperature measurement; based on the obtained electromagnetic interference intensity and the dust level, ventilation status and vibration intensity of each temperature monitoring point, the comprehensive interference value of each temperature monitoring point is calculated; this value reflects the total degree of interference to each monitoring point;

[0047] At the same time, the current external ambient temperature, external ambient humidity and load of the box-type substation are obtained. The external ambient temperature directly affects the temperature of the equipment inside the box-type substation. In a high temperature environment, the equipment is more likely to overheat; in a low temperature environment, the equipment may produce condensation due to the temperature difference, affecting the insulation performance; a high humidity environment may increase the risk of condensation on the surface of the equipment, reduce the insulation performance, and increase the risk of short circuit and failure; during the operation of the equipment, the load will cause heat to be generated inside the equipment. The greater the load, the more heat is generated, and the higher the temperature of the equipment, thereby increasing the risk of overheating; at the same time, based on the accurate temperature data obtained at each temperature monitoring point in the most recent preset update cycle, the comprehensive risk value of each temperature monitoring point is calculated;

[0048] Set the comprehensive interference threshold. For any temperature monitoring point, if the comprehensive interference value of the current temperature monitoring point is lower than the comprehensive interference threshold, the first-level filtering operation is performed on the initial temperature data collected at the current temperature monitoring point; if the comprehensive interference value of the current temperature monitoring point is higher than or equal to the comprehensive interference threshold, the second-level filtering operation is performed on the initial temperature data collected at the current temperature monitoring point;

[0049] For any temperature monitoring point, the comprehensive risk value of the current temperature monitoring point is used to calculate the temperature monitoring frequency within the preset temperature monitoring frequency adjustment range, and the calculated temperature monitoring frequency is used to collect initial temperature data within the current preset update cycle; a comprehensive risk threshold is set. If the comprehensive risk value of the current temperature monitoring point is higher than the comprehensive risk threshold, the external ambient temperature, external ambient humidity and load are obtained at the current temperature prediction time point, and the obtained external ambient temperature, external ambient humidity and load are input into the ideal temperature model, and the ideal temperature data at the current temperature prediction time point is output; the currently obtained accurate temperature data is compared with the ideal temperature data. If the difference between the accurate temperature data and the ideal temperature data is greater than the preset deviation value, the alarm response measures are executed.

[0050] Calculate the comprehensive interference value of each temperature monitoring point. The specific operations are as follows:

[0051] Based on the obtained electromagnetic interference strength EMI, set the electromagnetic interference threshold EMI threshold ; If the currently acquired electromagnetic interference intensity EMI is less than the electromagnetic interference threshold EMI threshold , then I EMI 0; if the electromagnetic interference intensity EMI is greater than or equal to the electromagnetic interference threshold EMI threshold , then use the formula Calculate the electromagnetic interference value I EMI ;

[0052] Based on the dust level D(i), ventilation status V(i) and vibration intensity Vib(i) of each temperature monitoring point, i=1, 2, ..., n; n represents the number of temperature monitoring points;

[0053] For dust level D(i), if the dust level D(i) at the temperature monitoring point is less than the preset dust level threshold D threshold , then the dust interference value I(i) D is 0 if the dust level D(i) is greater than or equal to the dust level threshold D threshold , then use the formula Calculate and obtain the dust interference value I(i) D ;

[0054] For ventilation status V(i): If the ventilation status V(i) of the temperature monitoring point is within the ideal ventilation range V min To F max If the ventilation interference value is I(i) V is 0; if the ventilation state V(i) is less than or equal to V min , then use the formula Calculate and obtain the ventilation interference value I(i) V ; If the ventilation state V(i) is greater than or equal to V max , then use the formula Calculate and obtain the ventilation interference value I(i) V ;

[0055] For the vibration intensity Vib(i), if the vibration intensity Vib(i) of the current temperature monitoring point is less than the preset vibration intensity threshold Vib threshold , then the vibration disturbance value I(i) Vib is 0; if the vibration intensity Vib(i) is greater than or equal to the vibration intensity threshold Vib threshold , then use the formula Calculate and obtain the vibration disturbance value, (i) Vib ;

[0056] Finally, the formula I(i) = α1I EMI +α2(I(i) D +I(i) V +I(i) Vib ) Calculate and obtain the comprehensive interference value I(i) of each temperature monitoring point; where α1 and α2 are weight coefficients, α1>α2.

[0057] Calculate and obtain the comprehensive risk value of each temperature monitoring point. The specific operations are as follows:

[0058] Based on the external ambient temperature T ext , external ambient humidity H and load L, for external ambient temperature Text , set the external reference temperature T ref and the maximum acceptable temperature T max ; When the external environment temperature T ext Less than or equal to T ref When the ambient temperature risk value R T is 0; when the external environment temperature T ext Located in T ref With T max When between Calculate the ambient temperature risk value R T ; When the external environment temperature T ext Higher than or equal to T max When the ambient temperature risk value R T is 1;

[0059] For the external environment humidity H, set the external reference humidity H ref , using the formula Calculate the environmental humidity risk value R H ; where k is the control coefficient, which is used to control the external environment humidity H to be higher than H ref After that, the environmental humidity risk value R H The rising speed;

[0060] For load L, set the load upper limit L max , using the formula Calculate the load risk value R L ;

[0061] For any temperature monitoring point, based on the accurate temperature data T obtained at the current temperature monitoring point within the most recent preset update period j , j = 1, 2, ..., m; m represents the number of accurate temperature data obtained; using the formula Calculate and obtain the temperature standard deviation σ of the current temperature monitoring point, where is the average value of accurate temperature data; set the normal temperature standard deviation σ under normal conditions ref and the maximum temperature standard deviation σ max , if the temperature standard deviation σ is less than or equal to σ ref , then the temperature fluctuation risk value R σ is 0; if the temperature standard deviation σ is within σ ref and σ max Between, use the formula Calculate and obtain the temperature fluctuation risk value R of the current temperature monitoring point σ ; If the temperature standard deviation σ is greater than or equal to σ max , then the temperature fluctuation risk value R σ1; at the same time, calculate the difference ΔT between the current accurate temperature data and the accurate temperature data obtained at the beginning of the most recent period, based on the preset update period duration t, using the formula Calculate the temperature change rate V ΔT ; If the temperature change rate V ΔT If the temperature change rate is lower than the preset threshold, the temperature change risk value R ΔT is 0; if the temperature change rate V ΔT If the temperature change rate is higher than or equal to the preset temperature change rate threshold, the temperature change risk value R ΔT is 1; using the formula The comprehensive risk value R of the current temperature monitoring point is calculated, where β1 and β2 are weight coefficients, β1<β2; finally, the comprehensive risk value R(i) of each temperature monitoring point is obtained.

[0062] Calculate and obtain the temperature monitoring frequency. The specific operations are as follows:

[0063] Set the temperature monitoring frequency adjustment range to F min To F max Based on the comprehensive risk value R(i) of all temperature monitoring points, the formula F(i) = F min +R(i)(F max -F min ) calculates and obtains the temperature monitoring frequency applied at each temperature monitoring point in the current time period.

[0064] The ideal temperature model is built on the basis of the random forest model. Random forest is an ensemble learning method that improves the accuracy and robustness of predictions by constructing multiple decision trees and integrating their results. During the training process, random forest randomly extracts samples from the original data set to construct each decision tree, and for each decision tree, it only considers a random subset of features when splitting nodes, thereby increasing the diversity between trees. When making predictions, random forest combines the results of all decision trees through majority voting or averaging to obtain the final prediction output. The specific operations for training the ideal temperature model are as follows:

[0065] Obtain historical temperature data, and use the historical temperature data under normal conditions as training samples, where each training sample contains several historical temperature data and the corresponding external environment temperature, external environment humidity and load; divide all training samples into a training set and a validation set, use the external environment temperature, external environment humidity and load as the input of the ideal temperature model, and use the historical temperature data as the output of the ideal temperature model, and input the training set into the ideal temperature model with parameter initialization for training; use the validation set to validate the ideal temperature model and obtain the validation result; determine whether the validation result meets the preset training conditions, and if so, output the trained ideal temperature model; if not, use the training set to continue training the ideal temperature model.

[0066] The filtering method used in the first-level filtering operation is sliding mean filtering; the filtering method used in the second-level filtering operation is Kalman filtering.

[0067] The comprehensive interference threshold is selected by the particle swarm optimization algorithm, which is an optimization technology based on swarm intelligence and is inspired by the foraging behavior of bird flocks. In PSO, each solution is regarded as a "particle" in the search space. The particle adjusts its position and speed in the problem space according to its own experience and the experience of other particles in the group to find the optimal solution. The particle swarm gradually approaches the optimal solution by iteratively updating its position and speed. This process involves two "extreme values" of individuals and groups: the optimal solution found by each particle (individual extreme value) and the optimal solution found by the entire particle swarm (global extreme value). The specific operations are as follows:

[0068] Step 1: Set the maximum number of iterations and randomly generate a particle swarm. The position of each particle represents a candidate threshold.

[0069] Step 2: For any particle, initialize the velocity of the particle and use the current position as the individual optimal position of the particle; use the formula Calculate the fitness value F of the particle, where S represents the size of the candidate threshold, and W represents the deviation rate of temperature monitoring using the current candidate threshold; the larger the candidate threshold, the smaller the deviation rate, and the larger the fitness value of the candidate threshold;

[0070] Step 3: Traverse all particles and take the position corresponding to the particle with the highest fitness value F as the global optimal position;

[0071] Step 4: For any particle, update the speed and position of the particle, and recalculate the fitness value; if the updated fitness value is higher than the fitness value of the individual best position of the particle, then update the individual best position of the particle to the current position;

[0072] Step 5: Traverse all particles. If the fitness value of the individual best position of any particle is higher than the fitness value of the global best position, the global best position is updated to the individual best position of the particle.

[0073] Step 6: Repeat steps 3 to 5 until the maximum number of iterations is reached. The global best position obtained is the optimal comprehensive interference threshold.

[0074] Example 2, a temperature monitoring system for a box transformer based on wireless sensing, such as Figure 1 As shown, including:

[0075] The risk value acquisition module is used to obtain the external environment temperature, external environment humidity and load of the box-type substation, and calculate the comprehensive risk value of each temperature monitoring point based on the accurate temperature data obtained at each temperature monitoring point within the latest preset update cycle;

[0076] The interference value acquisition module is used to obtain the electromagnetic interference intensity of the box-type substation, and at the same time obtain the dust level, ventilation status and vibration intensity of each temperature monitoring point respectively, and calculate the comprehensive interference value of each temperature monitoring point based on the obtained electromagnetic interference intensity and the dust level, ventilation status and vibration intensity of each temperature monitoring point;

[0077] The temperature monitoring module includes a frequency adjustment unit, a filtering unit, an ideal temperature acquisition unit and a temperature monitoring unit; the frequency adjustment unit is used to calculate and obtain the temperature monitoring frequency within a preset temperature monitoring frequency adjustment range using the comprehensive risk value of the temperature monitoring point, and use the calculated temperature monitoring frequency to collect initial temperature data within the current preset update cycle; the filtering unit is used to perform a first-level filtering operation on the initial temperature data collected from the temperature monitoring point when the comprehensive interference value of the temperature monitoring point is lower than the comprehensive interference threshold; when the comprehensive interference value of the temperature monitoring point is higher than or equal to the comprehensive interference threshold, perform a second-level filtering operation on the initial temperature data collected from the current temperature monitoring point; the ideal temperature acquisition unit is used to perform a second-level filtering operation on the initial temperature data collected from the current temperature monitoring point when the comprehensive risk value of the temperature monitoring point is higher than When the comprehensive risk threshold is reached, the external environment temperature, external environment humidity and load are obtained at the temperature prediction time point, and the obtained external environment temperature, external environment humidity and load are input into the ideal temperature model to output the ideal temperature data at the current temperature prediction time point; the temperature monitoring unit is used to collect initial temperature data at the monitoring time point of each temperature monitoring point; a filtering operation is performed on the collected initial temperature data to obtain accurate temperature data; it is determined whether the accurate temperature data is within the normal temperature range, if so, no operation is performed; if not, an alarm response measure is executed; at the same time, when the ideal temperature data exists, the currently acquired accurate temperature data is compared with the ideal temperature data, and if the difference between the accurate temperature data and the ideal temperature data is greater than the preset deviation value, an alarm response measure is executed.

[0078] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A temperature monitoring method for a box transformer based on wireless sensing, characterized in that: include: For any temperature monitoring point in the box-type substation, collect initial temperature data at the monitoring time point of the current temperature monitoring point; Perform filtering operations on the collected initial temperature data to obtain accurate temperature data; determine whether the accurate temperature data is within the normal temperature range, if so, no operation is performed; if not, perform alarm response measures; At the beginning of any preset update cycle, the electromagnetic interference intensity of the current box-type substation is obtained, and the dust level, ventilation status and vibration intensity of each temperature monitoring point are obtained respectively. Based on the obtained electromagnetic interference intensity and the dust level, ventilation status and vibration intensity of each temperature monitoring point, the comprehensive interference value of each temperature monitoring point is calculated; at the same time, the current external environment temperature, external environment humidity and load of the box-type substation are obtained, and based on the accurate temperature data of each temperature monitoring point obtained in the most recent preset update cycle, the comprehensive risk value of each temperature monitoring point is calculated; Set the comprehensive interference threshold. For any temperature monitoring point, if the comprehensive interference value of the current temperature monitoring point is lower than the comprehensive interference threshold, the first-level filtering operation is performed on the initial temperature data collected at the current temperature monitoring point; if the comprehensive interference value of the current temperature monitoring point is higher than or equal to the comprehensive interference threshold, the second-level filtering operation is performed on the initial temperature data collected at the current temperature monitoring point; For any temperature monitoring point, the comprehensive risk value of the current temperature monitoring point is used to calculate the temperature monitoring frequency within the preset temperature monitoring frequency adjustment range, and the calculated temperature monitoring frequency is used to collect initial temperature data within the current preset update period; Set a comprehensive risk threshold. If the comprehensive risk value of the current temperature monitoring point is higher than the comprehensive risk threshold, obtain the external ambient temperature, external ambient humidity and load at the current temperature prediction time point, input the obtained external ambient temperature, external ambient humidity and load into the ideal temperature model, and output the ideal temperature data at the current temperature prediction time point; compare the currently obtained precise temperature data with the ideal temperature data. If the difference between the precise temperature data and the ideal temperature data is greater than the preset deviation value, execute the alarm response measures.

2. According to the method for monitoring temperature of a box transformer based on wireless sensing according to claim 1, it is characterized in that: Calculate the comprehensive interference value of each temperature monitoring point. The specific operations are as follows: Based on the obtained electromagnetic interference strength EMI, set the electromagnetic interference threshold EMI threshold ; If the currently acquired electromagnetic interference intensity EMI is less than the electromagnetic interference threshold EMI threshold , then I EMI 0; if the electromagnetic interference intensity EMI is greater than or equal to the electromagnetic interference threshold EMI threshold , then use the formula Calculate the electromagnetic interference value I EMI ; Based on the dust level D(i), ventilation status V(i) and vibration intensity Vib(i) of each temperature monitoring point, i=1, 2, ..., n; n represents the number of temperature monitoring points; For dust level D(i), if the dust level D(i) at the temperature monitoring point is less than the preset dust level threshold D threshold , then the dust interference value I(i) D is 0 if the dust level D(i) is greater than or equal to the dust level threshold D threshold , then use the formula Calculate and obtain the dust interference value I(i) D ; For ventilation status V(i): If the ventilation status V(i) of the temperature monitoring point is within the ideal ventilation range V min To V max If the ventilation interference value is I(i) V is 0; if the ventilation state V(i) is less than or equal to V min , then use the formula Calculate and obtain the ventilation interference value I(i) V ; If the ventilation state V(i) is greater than or equal to V max , then use the formula Calculate and obtain the ventilation interference value I(i) V ; For the vibration intensity Vib(i), if the vibration intensity Vib(i) of the current temperature monitoring point is less than the preset vibration intensity threshold Vib threshold , then the vibration disturbance value I(i) Vib is 0; if the vibration intensity Vib(i) is greater than or equal to the vibration intensity threshold Vib threshold , then use the formula Calculate and obtain the vibration disturbance value I(i) Vib ; Finally, the formula I(i) = α1I EMI +α2(I(i) D +I(i) V +I(i) Vib ) Calculate and obtain the comprehensive interference value I(i) of each temperature monitoring point; where α1 and α2 are weight coefficients, α1>α2.

3. According to the method for monitoring temperature of a box transformer based on wireless sensing according to claim 2, it is characterized in that: Calculate and obtain the comprehensive risk value of each temperature monitoring point. The specific operations are as follows: Based on the external ambient temperature T ext , external ambient humidity H and load L, for external ambient temperature T ext , set the external reference temperature T ref and the maximum acceptable temperature T max ; When the external environment temperature T ext Less than or equal to T ref When the ambient temperature risk value R T is 0; when the external environment temperature T ext Located in T ref With T max When the Calculate the ambient temperature risk value R T ; When the external environment temperature T ext Higher than or equal to T max When the ambient temperature risk value R T is 1; For the external environment humidity H, set the external reference humidity H ref , using the formula Calculate the environmental humidity risk value R H ; where k is the control coefficient, which is used to control the external environment humidity H to be higher than H ref After that, the environmental humidity risk value R H The rising speed; For load L, set the load upper limit L max , using the formula Calculate the load risk value R L ; For any temperature monitoring point, based on the accurate temperature data T obtained at the current temperature monitoring point within the most recent preset update period j , j = 1, 2, ..., m; m represents the number of accurate temperature data obtained; using the formula Calculate and obtain the temperature standard deviation σ of the current temperature monitoring point, where is the average value of accurate temperature data; Set the normal temperature standard deviation σ under normal conditions ref and the maximum temperature standard deviation σ max , if the temperature standard deviation σ is less than or equal to σ ref , then the temperature fluctuation risk value R σ is 0; if the temperature standard deviation σ is within σ ref and σ max Between, use the formula Calculate and obtain the temperature fluctuation risk value R of the current temperature monitoring point σ ; If the temperature standard deviation σ is greater than or equal to σ max , then the temperature fluctuation risk value R σ 1; at the same time, calculate the difference ΔT between the current accurate temperature data and the accurate temperature data obtained at the beginning of the most recent period, based on the preset update period duration t, using the formula Calculate the temperature change rate V ΔT ; If the temperature change rate V ΔT If the temperature change rate is lower than the preset threshold, the temperature change risk value R ΔT is 0; if the temperature change rate V ΔT If the temperature change rate is higher than or equal to the preset temperature change rate threshold, the temperature change risk value R ΔT is 1; using the formula The comprehensive risk value R of the current temperature monitoring point is calculated, where β1 and β2 are weight coefficients, β1<β2; finally, the comprehensive risk value R(i) of each temperature monitoring point is obtained.

4. The method for monitoring temperature of a box transformer based on wireless sensing according to claim 3 is characterized in that: Calculate and obtain the temperature monitoring frequency. The specific operations are as follows: Set the temperature monitoring frequency adjustment range to F min To F max Based on the comprehensive risk value R(i) of all temperature monitoring points, the formula F(i) = F min +R(i)(F max -F min ) calculates and obtains the temperature monitoring frequency applied at each temperature monitoring point in the current time period.

5. The method for monitoring temperature of a box-type transformer based on wireless sensing according to claim 4 is characterized in that: The ideal temperature model is established based on the random forest model. The specific operations of training the ideal temperature model are as follows: Obtain historical temperature data, and use the historical temperature data under normal conditions as training samples, where each training sample contains several historical temperature data and the corresponding external environment temperature, external environment humidity and load; divide all training samples into a training set and a validation set, use the external environment temperature, external environment humidity and load as the input of the ideal temperature model, and use the historical temperature data as the output of the ideal temperature model, and input the training set into the ideal temperature model with parameter initialization for training; use the validation set to validate the ideal temperature model and obtain the validation result; determine whether the validation result meets the preset training conditions, and if so, output the trained ideal temperature model; if not, use the training set to continue training the ideal temperature model.

6. A temperature monitoring method for a box-type transformer based on wireless sensing according to claim 5, characterized in that: The filtering method used in the first-level filtering operation is sliding mean filtering; the filtering method used in the second-level filtering operation is Kalman filtering.

7. A temperature monitoring method for a box-type transformer based on wireless sensing according to claim 6, characterized in that: The comprehensive interference threshold is selected by particle swarm optimization algorithm. The specific operation is as follows: Step 1: Set the maximum number of iterations and randomly generate a particle swarm. The position of each particle represents a candidate threshold. Step 2: For any particle, initialize the velocity of the particle and use the current position as the individual optimal position of the particle; use the formula Calculate the fitness value F of the particle, where S represents the size of the candidate threshold, and W represents the deviation rate of temperature monitoring using the current candidate threshold; the larger the candidate threshold, the smaller the deviation rate, and the larger the fitness value of the candidate threshold; Step 3: Traverse all particles and take the position corresponding to the particle with the highest fitness value F as the global optimal position; Step 4: For any particle, update the speed and position of the particle, and recalculate the fitness value; if the updated fitness value is higher than the fitness value of the individual best position of the particle, then update the individual best position of the particle to the current position; Step 5: Traverse all particles. If the fitness value of the individual best position of any particle is higher than the fitness value of the global best position, the global best position is updated to the individual best position of the particle. Step 6: Repeat steps 3 to 5 until the maximum number of iterations is reached. The global best position obtained is the optimal comprehensive interference threshold.

8. A temperature monitoring system for a box transformer based on wireless sensing, characterized in that: The system is applied to the box-type transformer temperature monitoring method based on wireless sensing as described in any one of claims 1 to 7 above, comprising: The risk value acquisition module is used to obtain the external environment temperature, external environment humidity and load of the box-type substation, and calculate the comprehensive risk value of each temperature monitoring point based on the accurate temperature data obtained at each temperature monitoring point within the latest preset update cycle; The interference value acquisition module is used to obtain the electromagnetic interference intensity of the box-type substation, and at the same time obtain the dust level, ventilation status and vibration intensity of each temperature monitoring point respectively, and calculate the comprehensive interference value of each temperature monitoring point based on the obtained electromagnetic interference intensity and the dust level, ventilation status and vibration intensity of each temperature monitoring point; The temperature monitoring module includes a frequency adjustment unit, a filtering unit, an ideal temperature acquisition unit and a temperature monitoring unit; the frequency adjustment unit is used to calculate and obtain the temperature monitoring frequency within a preset temperature monitoring frequency adjustment range using the comprehensive risk value of the temperature monitoring point, and use the calculated temperature monitoring frequency to collect initial temperature data within the current preset update cycle; the filtering unit is used to perform a first-level filtering operation on the initial temperature data collected from the temperature monitoring point when the comprehensive interference value of the temperature monitoring point is lower than the comprehensive interference threshold; when the comprehensive interference value of the temperature monitoring point is higher than or equal to the comprehensive interference threshold, perform a second-level filtering operation on the initial temperature data collected from the current temperature monitoring point; the ideal temperature acquisition unit is used to perform a second-level filtering operation on the initial temperature data collected from the current temperature monitoring point when the comprehensive risk value of the temperature monitoring point is higher than When the comprehensive risk threshold is reached, the external environment temperature, external environment humidity and load are obtained at the temperature prediction time point, and the obtained external environment temperature, external environment humidity and load are input into the ideal temperature model to output the ideal temperature data at the current temperature prediction time point; the temperature monitoring unit is used to collect initial temperature data at the monitoring time point of each temperature monitoring point; a filtering operation is performed on the collected initial temperature data to obtain accurate temperature data; it is determined whether the accurate temperature data is within the normal temperature range, if so, no operation is performed; if not, an alarm response measure is executed; at the same time, when the ideal temperature data exists, the currently acquired accurate temperature data is compared with the ideal temperature data, and if the difference between the accurate temperature data and the ideal temperature data is greater than the preset deviation value, an alarm response measure is executed.

Images