928 results about "Transformer oil" patented technology

The invention discloses an environment-adaptive transformer oil sample Raman spectrum detection method and related equipment, and relates to the field of optical sensing systems. The method comprises the following steps: collecting oil sample Raman spectrums and environmental parameters in multiple operation scenes, and constructing a multi-scene spectrum characteristic model and a standard fingerprint database; pre-processing and denoising parameters are adaptively set based on the environmental perception vector, and baseline correction and joint denoising are carried out on the original spectrum; scene discrimination is carried out by fusing the characteristics of peak position, peak height, peak width, integral area and the like, a scene-related component standard spectrum dictionary is generated, and the concentration and confidence of each target component are obtained by adopting constrained spectral line unmixing and quantitative calibration; and driving the fingerprint database and the model to update in combination with quality control indexes such as spectral shape relevancy and residual errors and a drift detection result. The system is composed of a Raman spectrum acquisition module, an environment monitoring module and a data processing module, and can improve the robustness and quantitative precision of Raman detection of transformer oil in a complex environment.

The invention discloses a transformer oil leakage detection method and device based on polarization-ultraviolet fusion imaging, and the device comprises a housing, a support, a camera, an environment induction device, an overall controller, a power supply, a polaroid rotating wheel assembly, a lens seat assembly, and an image processor. And the motor driver, the illumination intensity sensor, the ultraviolet irradiance sensor, the image processor and the power supply are electrically connected with the integral controller respectively. According to the invention, a multi-angle polarization image and a non-polarization image in the same field of view can be acquired, and then polarization characteristics contained in an oil stain fluorescence signal are extracted and effectively distinguished from non-polarization characteristics of background light; in the daytime mode, the transformer oil leakage detection device can identify a target area with relatively high polarization intensity, and through image fusion enhancement processing, non-target background signals are significantly suppressed, and the fluorescence significance of an oil leakage characteristic area is improved.

The invention provides a power transformer free gas detection method and system for material screening based on machine learning, and the method comprises the steps: carrying out the sampling of a compound space, and calculating a material characteristic data set of different doping configurations for the adsorption performance of a target gas through a high-throughput first principle; based on the material characteristic data set, a CGCNN convolution kernel is fused through a GNN framework, and a pre-trained material screening model is obtained and used for predicting the adsorption performance of different doping configurations to target gas; screening by utilizing a pre-trained material screening model to obtain an optimal doping configuration meeting a preset performance threshold value; screening based on the optimal doping configuration to obtain an optimal gas sensitive material; acquiring a response signal of the sensor array to the dissolved gas removed from the transformer oil, identifying the type of the target gas, and estimating the corresponding gas concentration; wherein the sensor array is prepared from an optimal gas sensitive material. According to the invention, rapid, accurate and on-line detection of various key free gases in transformer oil can be realized.

The invention provides a degassing-free transformer oil monitoring device and method. A transparent accommodating pipe is used for being communicated with an external transformer oil supply device; the laser detection assembly is arranged on one side of the transparent accommodating tube; the voltage regulation and control unit is connected with the transparent accommodating tube; the imaging assembly is arranged on one side of the transparent accommodating tube and is used for acquiring optical image data of the Tyndall phenomenon in the transformer oil sample under different voltage conditions; the data processing unit is used for setting a segmentation point in the middle position of a Tyndall light pillar based on the Tyndall light pillar in the image data collected by the imaging assembly, determining a standard deviation according to the ratio of the total number of gray pixels on the left side and the right side of the Tyndall light pillar, and determining a threshold value of the Tyndall light pillar according to the relation between the standard deviation and a preset threshold value range. And judging the existence state and distribution characteristics of the gas in the transformer oil sample. Degassing treatment is not needed, the process is simplified, sample pollution is avoided, and the monitoring accuracy and stability of the gas in the transformer oil are remarkably improved.

The invention belongs to the technical field of power equipment state monitoring, and discloses a transformer oil dissolved gas surface enhanced Raman spectrum in-situ monitoring method and a transformer oil dissolved gas surface enhanced Raman spectrum in-situ monitoring system. The system comprises a membrane degassing unit, a Raman sensing unit, a light path unit, a data processing unit and a calibration early warning unit, the method comprises the following steps: firstly preparing a gold nanostar array sensing substrate with bifunctional groups to realize selective adsorption and signal enhancement of characteristic gas, then completing in-situ oil-gas separation through the membrane degassing unit, collecting a Raman spectrum through the light path unit, and calculating the Raman spectrum of the gold nanostar array sensing substrate according to the Raman spectrum. Spectral analysis and quantification are completed based on multiple algorithms, and fault early warning is achieved in combination with temperature and pressure compensation and an industrial standard threshold value. According to the scheme, the problems of lagging, weak interference resistance and the like of traditional monitoring are solved, high sensitivity and high specificity are achieved, full-process automatic closed-loop monitoring can be achieved, and accurate data support is provided for operation and maintenance of the transformer.

The invention relates to the technical field of oil level monitoring, in particular to a transformer oil level monitoring device and system based on dual-mode ultrasonic sensing. The method comprises the following steps: acquiring transformer oil level and vibration acceleration; threshold value differences are obtained through oil level comparison, and the oil levels are divided; based on different types of oil level fluctuations, quantities and intervals, oil level anomaly comprehensive indexes are obtained; after the over-limit oil level is screened, an oil level abnormal factor is determined, and an oil level abnormal value is obtained in combination with the oil level abnormal comprehensive index; determining a change moment and a reaction time range based on the vibration acceleration; an acceleration oil level influence factor is determined by analyzing correlation, and a fault characteristic value is obtained in combination with an interference factor; and monitoring the oil level through the fault characteristic value. The reliability of the whole monitoring system is improved.

The invention relates to the technical field of measurement and testing, in particular to a leakage detection method and system for transformer oil tank production, and the method comprises the steps: dividing a to-be-detected oil tank into a plurality of test regions; determining an ideal net pressure attenuation rate by using the average pressure drop rate, the pressure change rate corresponding to the temperature change and the volume change amount of the oil tank cavity in the target time period in the target test area; determining a pressure attenuation deviation degree by using the ideal net pressure attenuation rate and the actually measured net pressure attenuation rate; determining an acoustic activity weight by using the acoustic activity degree of the acoustic signal in the acoustic event and the background noise average energy of the plurality of test areas in the acoustic event; and determining a regional leakage comprehensive index by using the acoustic activity weight and the pressure attenuation deviation degree, and further determining the leakage detection condition of the oil tank by using the regional leakage comprehensive index. According to the technical scheme, the accuracy and reliability of leakage detection are effectively improved, environment temperature interference is effectively eliminated, and background noise changes are dynamically adapted.

The oil-immersed transformer oil tank sealing anti-seepage structure comprises a transformer oil tank, a fixing assembly and an end cover are arranged at the upper end of the transformer oil tank, the transformer oil tank is connected with the end cover through the fixing assembly, and a jacking mechanism, a transmission assembly and a pressing assembly are arranged at the upper end of the end cover; the jacking mechanism provides power for the pressing assembly through the transmission assembly. The lower end of the end cover is fixedly connected with a sealing ring and a sealing air bag, the side end of the sealing air bag is fixedly connected with a connecting pipe, the end, away from the sealing air bag, of the connecting pipe penetrates through the end cover and is fixedly connected with the end cover, and the connecting pipe communicates with the pressing assembly. By arranging the fixing assembly, the sealing ring and the sealing air bag, double-layer sealing can be achieved at the joint of the end cover and the transformer oil tank, and therefore leakage of transformer oil is effectively prevented.

The invention provides a transformer oil-water exchanger leakage early warning and protection method based on a cooler body. According to the leakage early warning and protection method for the oil-water exchanger of the transformer based on the cooler body, accurate early warning and positioning can be achieved in the early stage of tiny leakage of the oil-water exchanger of the cooler body, water is effectively prevented from entering a main oil way of the transformer by automatically isolating the faulty cooler, and the leakage of the oil-water exchanger of the transformer is prevented. And the system response speed and the operation reliability are obviously improved.

The invention discloses a transformer oil leakage detection algorithm based on improved YOLOv8, and the algorithm comprises the following steps: obtaining transformer data, carrying out the preprocessing of the transformer data, and obtaining the preprocessed transformer data; extracting the preprocessed transformer data by adopting a backbone network to obtain multi-scale feature data; embedding the simple parameter-free attention module into the original neck network to obtain a target neck network, and inputting the multi-scale feature data into the target neck network for multi-scale fusion to obtain fused feature data; executing a detection task on the fused feature data by adopting a head network to obtain oil leakage detection information; and removing redundant detection frames in the oil leakage detection information by using a non-maximum suppression algorithm to obtain an oil leakage detection result of the transformer. According to the method, the problems that a YOLOv8 model cannot cope with a complex environment and the detection accuracy is not high are solved.

The invention discloses a transformer fault intelligent diagnosis system and method based on a K-Medoid and SMOTE optimization support vector machine, and the system comprises a data collection module which is used for collecting dissolved gas and operation parameters in transformer oil in real time; the data preprocessing module is connected with the data acquisition module and is used for completing data standardization and abnormal value detection; the sample balancing module is connected with the data preprocessing module and used for reducing majority class redundancy through K-Medoid clustering and expanding minority classes through an SMOTE algorithm; the intelligent classification module is connected with the sample balance module and adopts a weighted SVM classifier to train and predict; and the result output and visualization module is connected with the intelligent classification module and is used for displaying the classification result, the confidence coefficient and the key performance indexes. According to the method, the problem of classification performance degradation caused by sample imbalance in the prior art is solved, and the recognition capability of the SVM model on minority class faults and the overall diagnosis precision are improved.

In the system, an electromagnetic simulation module establishes a transformer three-dimensional electromagnetic field model based on a finite element method, and calculates magnetic field distribution, winding eddy current loss and iron core hysteresis loss to obtain loss data; the fluid-solid coupling calculation module is connected with the electromagnetic simulation module so as to simulate a transformer oil flowing and heat conduction process based on computational fluid dynamics; the optimization design module is connected with the fluid-solid coupling calculation module so as to perform parametric modeling and optimization on the winding structure, the cooling channel and the cooling fin layout through the hot spot position and temperature distribution based on a multi-target genetic algorithm and an agent model; the state evaluation and heat dissipation adjustment module is connected with the data interaction layer so as to analyze the health state of the transformer based on sensor data and simulation results, predict the hot-spot temperature trend and dynamically adjust operation parameters of the cooling system through a self-adaptive heat dissipation adjustment algorithm.

The invention discloses a transformer oil sample automatic collection method and device based on vision and impedance control, and the method comprises the steps: determining a relative position between an oil taking injector and an oil taking port in a transformer oil sample collection process, and determining an ideal pose of the oil taking injector controlled by a mechanical arm to reach an oil taking port mark based on the relative position; based on image information collected in real time by an image taking device arranged on the mechanical arm, a space coordinate transformation matrix of the ideal pose at the oil taking opening mark and the actual pose of the image taking device is calculated; on the basis of the ideal pose and the space coordinate transformation matrix which are calculated in real time, according to a control law of pose deviation design, a visual servo control rule for the pose of the oil taking injector is determined; based on interaction between the working environment and the tail end of the mechanical arm, a butt joint depth control rule of an oil taking injector and an oil taking opening is determined through an impedance controller; based on a visual servo control rule and a butt joint depth control rule, butt joint of an oil taking injector and an oil taking opening is controlled, and transformer oil sample collection is carried out.

The invention relates to a transformer insulating oil sampling device, and belongs to the technical field of transformer oil sampling, the transformer insulating oil sampling device comprises an insulating oil transfer cabinet, a male joint, an oil sampling machine vehicle, an oil well pump and a female joint, a transformer oil sampling hole, the insulating oil transfer cabinet and the male joint are sequentially communicated in series through an oil delivery pipe; the oil-well pump and the female joint are fixed on the oil sampling machine vehicle at an interval; and the pump inlet end of the oil-well pump is communicated with the oil outlet end of the female joint through an oil conveying pipe. The oil inlet hole in the insulating oil transfer cabinet is communicated with the oil taking hole of the transformer through the oil conveying pipe, then the female connector fixed to the oil sampling machine vehicle is in butt joint communication with the male connector fixed to the insulating oil transfer cabinet, the male connector and the female connector can be directly in butt joint communication, the space position adjusting operation is omitted, and the working efficiency is improved. The structure of the transformer insulating oil sampling device is simplified, the risk of collision with the transformer is reduced, and the transformer insulating oil sampling efficiency is improved.

A transformer fault severity assessment method based on oil chromatographic analysis and characteristic quantification belongs to the technical field of power equipment fault diagnosis, and comprises the following steps: regularly collecting a transformer oil sample, detecting concentration data of at least one characteristic gas in oil, forming a time sequence data sequence, the method comprises the following steps: constructing a multi-dimensional input feature set as the input of an XGBoost classification model, adjusting and optimizing model parameters, training the model to output a fault type identification result, adopting k-means and SMOTE algorithms to obtain the upper bound and the lower bound of the concentration of each feature gas in a normal state and in each fault type, adopting an SHAP algorithm to calculate the Shapley value of each input feature, and obtaining the fault type identification result. And taking the absolute value of the Shapley value as the importance weight of the feature to the corresponding fault type, and carrying out weighted summation on the difference value to obtain a comprehensive evaluation index. Through the improvement of the algorithm, the fault type and severity of the transformer can be accurately evaluated.

The invention discloses a preparation method of high-temperature transformer oil, and belongs to the technical field of high-temperature transformer oil. Comprising the following steps: 1, carrying out hydrolysis reaction to obtain ricinoleic acid; step 2, carrying out first esterification reaction to obtain a ricinoleic acid esterification product; 3, performing epoxidation reaction to obtain an epoxidized ricinoleic acid esterification product; 4, carrying out ring-opening reaction to obtain ring-opened castor oil; 5, carrying out a second esterification reaction to obtain modified castor oil base oil; and 6, mixing the modified castor oil base oil, a viscosity index improver, an antioxidant, a demulsifying agent and an extreme pressure agent to obtain the high-temperature transformer oil. The invention provides modified castor oil and a preparation method thereof. The high-temperature antioxidant transformer oil prepared by using the modified castor oil as base oil has the characteristics of environmental protection, degradability, high flash point, high breakdown voltage and the like, and also has excellent oxidation stability.

The invention relates to the technical field of power equipment state monitoring, and discloses a transformer oil paper insulation frequency domain response online monitoring system which comprises an interface protection module, a modulation control module, a servo balance module and a diagnosis reconstruction module. The interface protection module extracts end screen leakage current and provides overvoltage protection; the modulation control module tracks the bus voltage phase and generates broadband modulation excitation voltage; the servo balance module generates a reverse servo current through closed loop feedback, counteracts a power frequency component in a leakage current so as to maintain a virtual ground state of a measurement node, and injects broadband excitation in a superposed manner; and the diagnosis reconstruction module synchronously acquires signals, reconstructs data in combination with time sequence information, calculates equivalent complex admittance by utilizing cross-correlation demodulation, and outputs complex capacitance and a dielectric loss factor. According to the invention, the operating voltage is used as a carrier wave, strong power frequency interference is eliminated through hardware servo, and safe online excitation is realized by matching with phase chopping, so that the difficult problem of weak broadband signal extraction is effectively solved, and combined monitoring and diagnosis are realized.

The invention discloses an oil temperature monitoring and regulating method for an oil-immersed transformer, which belongs to the technical field of control optimization, and comprises the following steps: constructing an oil temperature PID control system of the oil-immersed transformer, introducing an improved candle flame optimization algorithm, and setting oil temperature PID control parameters by using the improved candle flame optimization algorithm. And setting three optimal control parameters obtained through optimization by using an improved candle flame optimization algorithm as parameters of an oil temperature PID controller in the PID control system of the oil-immersed transformer, and optimizing the oil temperature control effect of the oil-immersed transformer. According to the method, the oil temperature PID control system of the oil-immersed transformer is constructed, and the improved candle flame optimization algorithm is introduced, so that the oil temperature of the oil-immersed transformer can be recovered to be stable more quickly when working condition disturbance such as load sudden change or environment temperature change occurs, the insulation aging risk caused by long-time high-temperature operation can be reduced, and the service life of the oil-immersed transformer is prolonged. And the operation safety and reliability of the oil-immersed transformer are enhanced.

The invention provides a method for calculating the leakage current of an iron core clamp of a power transformer, which belongs to the technical field of electric power, and comprises the following steps: S1, installing a high-frequency vibration sensor on the wall of an oil tank of the transformer, capturing a vibration voiceprint signal through a multi-mode data synchronous acquisition module of an iron core clamp leakage calculation system, and calculating the leakage current of the iron core clamp; synchronously acquiring three-phase operation current and acoustic emission signals of high and low voltage sides of the transformer; s2, using a voiceprint feature extraction module to extract a feature energy proportion of the vibration signal; s3, establishing a dynamic health reference model by using a load-voiceprint association analysis module, and identifying a mechanical looseness abnormal association mode; s4, performing diagnosis and early warning on the abnormal mode by using a fault diagnosis and early warning module; and S5, based on the severity of the mechanical looseness, calculating the leakage current value of the iron core clamp to the ground through the fault degree-leakage current mapping relation model, and solving the problem that the leakage current value cannot be estimated in advance based on the mechanical looseness state before the leakage current is formed in the prior art.

The utility model relates to the technical field of oil storage structures, in particular to an oil storage structure of an oil-immersed transformer, which comprises a transformer outer protection body, a fixing column is fixedly connected to one side of an inner cavity of the transformer outer protection body, and a filter screen is arranged in an inner cavity of the fixing column. The self-cleaning oil tank has the advantages of filtering injected oil and self-cleaning a filtering structure, and ensures that the oil entering the oil tank is always kept clean through the cooperative use of the fixed column, the filter screen, the linkage telescopic rod, the connecting rod, the moving rod, the fixed block, the rack, the fixed frame, the supporting table, the motor, the rotating rod and the gear; therefore, pollutants are prevented from affecting the heat exchange and insulation performance of oil, the service life of the transformer is effectively prolonged, the safety in the using process is improved, the filter screen can be automatically cleaned, when impurities are gradually accumulated on the inner side and the outer side of the filter screen, the formed pressure difference is gradually increased along with time until the preset threshold value is reached, and the service life of the transformer is prolonged. At the moment, the backwashing process is automatically started.

The invention discloses a transformer oil gas real-time monitoring method and device by combining a negative curvature optical fiber with a Raman spectrum. The transformer oil gas real-time monitoring device comprises an optical fiber transmission module, an optical detection module and a multi-degree-of-freedom adjusting mechanism, the optical fiber transmission module comprises a negative curvature optical fiber with a side hole, an optical fiber connector and an optical fiber protection sleeve, and efficient signal transmission is achieved. The optical detection module comprises a light splitting unit, a signal receiving unit and a signal processing unit, and is provided with an automatic gain control unit electrically connected with the signal receiving unit; the multi-degree-of-freedom adjusting mechanism comprises a rotating platform, a pitching support and a telescopic arm, and signal collection is optimized by adjusting the angle and the position. And an integrated interface unit is arranged between the optical fiber transmission module and the optical detection module. The signal transmission efficiency is improved through the negative curvature optical fiber, the system stability is enhanced through automatic gain control and temperature compensation, the type and concentration of gas dissolved in transformer oil are accurately analyzed, and the efficient real-time monitoring requirement of an electric power system is met.

The invention discloses an oil-immersed transformer temperature sensor intelligent on-line metering calibration system and method. The system comprises an on-line metering acquisition unit, a data processing and communication unit and an intelligent analysis platform. The on-line metering acquisition unit is used for acquiring an internal oil temperature signal and a reference temperature signal of the oil-immersed transformer in real time; the data processing and communication unit is used for receiving and processing the internal oil temperature signal and the reference temperature signal to obtain primary processing data; and the intelligent analysis platform is used for receiving the primary processing data, carrying out on-line comparison and analysis on the primary processing data based on a preset temperature sensor on-line calibration model, and generating on-line metering calibration information for the temperature sensor of the oil-immersed transformer. The transformer oil temperature and the high-precision reference signal are collected and compared in real time, the calibration information is automatically generated through intelligent analysis, online and automatic temperature sensor calibration is achieved, the accuracy of monitoring data is improved, and reliable guarantee is provided for equipment state evaluation.

The invention discloses a main transformer oil temperature abnormity real-time monitoring and statistical analysis method, system, equipment and medium, and belongs to the technical field of power equipment state monitoring and fault diagnosis, and the method comprises the steps: collecting temperature data and voiceprint data, carrying out the preprocessing, splicing the data into a combined oil temperature value with voiceprint, calculating the historical distribution of the current temperature, and carrying out the statistical analysis of the current temperature; outputting a dynamic threshold median, calculating a residual error duration, judging whether an oil temperature abnormal mark is triggered or not, when the oil temperature abnormal mark is triggered and an aging index is obtained, generating a root cause event list, performing efficiency evaluation and cooler regulation and control based on the root cause event list, outputting a regulation and control instruction and a current comprehensive energy efficiency ratio, and performing power regulation and control through a power dispatching data network. And requesting to obtain comprehensive efficiency ratios and oil temperature data of other main transformers in the same partition, generating a regional efficiency thermodynamic table, and carrying out oil temperature anomaly statistics. According to the invention, a closed-loop intelligent monitoring system from accurate sensing to co-processing is constructed, and the initiative and accuracy of main transformer oil temperature abnormity management and the overall safety of a power grid are improved.

The invention relates to the technical field of electrical equipment, in particular to a moisture-proof impregnated paper composite insulating sleeve and a preparation method and application thereof. The moisture-proof impregnated paper composite insulating sleeve provided by the invention comprises an impregnated paper capacitor core body and a moisture-proof enhancement layer on the outer surface, the moisture-proof reinforcing material comprises at least one of glass fiber, polyester and polyolefin, and the polyester comprises polyethylene glycol terephthalate; the impregnated paper capacitor core body comprises a central conductive tube, impregnated paper and a capacitor plate; the impregnated paper and the capacitor plates are alternately coated on the outer surface of the central conductive tube from inside to outside, and the impregnated paper or the capacitor plates are arranged on the outermost layer of the impregnated paper capacitor core body. According to the invention, the moisture-proof reinforcing layer is arranged on the outer surface of the impregnated paper capacitor core body, so that moisture in air during storage and trace moisture in transformer oil during operation can be effectively prevented from invading the impregnated paper capacitor core body, the capacitor core body is prevented from being affected with damp, the electrical property is reduced, the mechanical bending strength is enhanced, and the service life of the capacitor sleeve is prolonged.

The application discloses a kind of key feature transformer health state detection systems, it is related to power equipment monitoring technical field, including: data acquisition module, data processing module, confidence calculation module, fault determination module and result output module;Data acquisition module synchronously acquires transformer oil tank surface vibration signal and neutral point direct current data;Data processing module calculates the phase difference of vibration signal and the coherence coefficient of vibration signal and direct current data;Confidence calculation module calculates mechanical fault confidence;Fault determination module judges whether mechanical fault confidence exceeds preset threshold, exceeds and then carries out winding loosening positioning processing, generates loosening coordinate;Otherwise, the multi-band resampling of vibration signal is carried out and the coherence coefficient data is recalculated;Result output module outputs loosening coordinate or interference warning signal;The application can improve the accuracy of transformer health state detection, ensure the reliability of fault positioning.

The invention relates to the technical field of filtering mechanical purification equipment, and discloses a transformer oil circulating filtering mechanical purification device which comprises a lower push plate, a driven rod is rotationally connected to the end of the lower push plate, an expansion rod is rotationally connected to the lower surface of the driven rod, and a sealing plate is rotationally connected to the end, away from the driven rod, of the expansion rod. Through the arrangement of the storage mechanism, impurities concentrated by a concentration ring and a contraction ring can be pushed into a collection box at the moment, so that the impurities are collected through the collection box, and when certain impurities are collected in the collection box, a worker can pull a handle to pull the collection box towards the direction of the handle; and at the moment, the two sides of the collecting box slide towards the direction of the handle on the inner walls of the sliding rails, so that the collecting box is taken out, impurities in the collecting box are cleared away, the effect of collecting the impurities can be effectively achieved through the arrangement of the collecting box, and meanwhile a worker can take out the collecting box at regular time and clear away the impurities in the collecting box.

The invention relates to a transformer oil acetylene abnormity early warning system and early warning method based on multi-physics field data fusion and a computer readable storage medium, and solves the problem that a transformer fault cannot be accurately judged based on a single parameter in the prior art. According to the technical scheme and technical effects, the system integrates gas, thermal field, electric field and magnetic field sensors, collects acetylene concentration, temperature, electric field intensity, magnetic field and other multi-source data in real time, inputs the data into a deep neural network DNN fusion model after wavelet transform denoising, normalization and machine learning preprocessing, extracts features in combination with PCA, analyzes multi-physics field association, and obtains the multi-source data of acetylene concentration, temperature, electric field intensity, magnetic field and the like. Graded early warning is formed based on DNN output and a dynamic threshold value, the accuracy and predictive capacity of fault diagnosis are effectively improved, accurate monitoring and intelligent early warning are achieved, safe operation of the transformer is guaranteed, meanwhile, due to the modular design, the system can adapt to transformers of different models and complex working condition scenes, and high practicability and expansibility are achieved.

The invention discloses a transformer oil level on-line monitoring and early warning method and system and a medium, and relates to the technical field of transformer oil level monitoring, and the method comprises the steps: obtaining historical data in the same time period; preprocessing the historical data to obtain preprocessed data; constructing an oil level prediction model by adopting an LSTM model and fusing an attention mechanism, and training the oil level prediction model to obtain a trained oil level prediction model; inputting top oil temperature, load current and environment temperature data obtained in real time into the trained oil level prediction model for prediction, and outputting a predicted oil level value and an attention weight; and subtracting the actually measured oil level value from the predicted oil level value to obtain a residual error, analyzing the distribution condition of the residual error and the attention weight to obtain a reason causing the current oil level change, and generating an instructive alarm and an operation and maintenance suggestion. According to the invention, early warning before low oil level alarm or fault caused by tiny leakage is realized, specific fault reasons can be analyzed, and corresponding operation and maintenance suggestions can be given.

The utility model discloses an offshore photovoltaic box transformer substation supporting system, belongs to the technical field of offshore photovoltaic box transformer substations, and can solve the problems that an existing box transformer substation platform is difficult to stably install on the sea and is not suitable for offshore working conditions. The system comprises a plurality of supporting piles; the box transformer substation platform is fixed at the top ends of the plurality of supporting piles and is used for supporting box transformer substation equipment; the box transformer platform is provided with an overhanging part horizontally extending towards one side; the overhanging part is used for being connected with a cable bridge; the cable platform is connected below the box transformer platform and is used for supporting a cable; the oil tank frame is connected below the box-type transformer substation platform, is positioned on one side of the cable platform and is used for mounting a box-type transformer substation oil tank; and the overhauling ladder stand is fixed on the supporting piles, and the top end of the overhauling ladder stand is communicated with the overhauling inlet of the box transformer substation platform. The support is used for supporting and installing the box transformer substation and related equipment on the sea.

The utility model provides a kind of high-strength three-dimensional oil-immersed transformer oil tank, including oil tank shell and top cover, the top cover is fixed in the top of oil tank shell by bolt, the front and back of oil tank shell are symmetrically equipped with multiple radiating fins, the radiating fin is evenly equidistant arrangement, the oil tank shell is also equipped with heat sink device. The heat sink device includes air pump, gas storage pipe and air outlet plate, the air pump is arranged in the lower end of oil tank shell, the output end of the air pump is equipped with connecting air pipe, the gas storage pipe is equipped with two, symmetrically fixed in the front and back of oil tank shell, located the bottom of radiating fin, the connecting air pipe is tee pipe, respectively connected with the output end of air pump and two gas storage pipes, through the synergistic effect of radiating fin, heat sink device and oil tank shell inner wall coolant, multiple levels heat dissipation is realized, greatly improve the heat dissipation efficiency of oil tank, can effectively reduce the temperature of transformer oil, guarantee the insulation performance and service life of transformer.