Comprehensive intelligent monitoring system for transformer operating state

Abstract

The application discloses a transformer operation state comprehensive intelligent monitoring system, which comprises an information acquisition unit, a preprocessing unit, a core analysis unit, an optimization design unit and an early warning unit. The operation data and environmental data of the transformer are collected by the information acquisition unit, and are preliminarily analyzed by the preprocessing unit. Then, the transformer monitoring model is constructed by the core analysis unit, the influence function between the environmental data and the operation data is built by monitoring the working environment of the transformer, the sensitivity of the transformer to environmental changes is improved, the accuracy of the transformer monitoring model is analyzed by the optimization design unit, the fault indicators are processed in a targeted manner, the accuracy of fault repair and risk response is improved, the prediction risk degree of the transformer is deeply analyzed and evaluated by the early warning unit, and the early warning signal is generated to realize early warning processing of the fault, so that the sudden influence of abnormal climate and environmental changes on the transformer can be responded.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of transformer monitoring, in particular to a comprehensive intelligent monitoring system for the running state of a transformer. BACKGROUND

[0002] A transformer is a device for changing AC voltage by using the principle of electromagnetic induction, and its main components are a primary coil, a secondary coil and a core (magnetic core). Its main functions include voltage transformation, current transformation, impedance transformation, isolation, voltage stabilization (magnetic saturation transformer) and the like.

[0003] The running state of a transformer is affected by its working environment, and the existing comprehensive intelligent monitoring system for the running state of a transformer has the defect of insufficient sensitivity to environmental changes, which leads to insufficient real-time monitoring capability and risk warning capability of the running state of the transformer. Under the influence of abnormal climate conditions, such as high fluctuation degree of the working environment of the transformer or sudden extreme climate, the transformer may have unexpected failures, affecting the continuous and stable operation of the transformer.

[0004] In view of the above technical defects, a solution is proposed. SUMMARY

[0005] The present application aims to solve the problem of insufficient sensitivity to environmental changes in the prior art, and the problem of insufficient real-time monitoring capability and risk warning capability of the running state of a transformer. The running state of a transformer is evaluated by comprehensive analysis of insulation performance, mechanical risk and power stability, improving the comprehensiveness of data processing. The influence function between environmental data and running data is established by monitoring the working environment of the transformer, thereby extending the analysis of the comprehensive intelligent evaluation of the running state of the transformer, improving the sensitivity of the transformer to environmental changes, and analyzing and evaluating the predicted risk degree of the transformer in depth through the early warning prompt unit to cope with the sudden impact of abnormal climate environmental changes on the transformer.

[0006] In order to achieve the above-mentioned purpose, the present application adopts the following technical scheme:

[0007] The comprehensive intelligent monitoring system for the running state of a transformer comprises an information acquisition unit, a preprocessing unit, a core analysis unit, an optimization design unit and an early warning prompt unit, wherein the information acquisition unit, the preprocessing unit, the core analysis unit, the optimization design unit and the early warning prompt unit are signal connected.

[0008] The information acquisition unit is used to acquire the running data and environmental data of the transformer: set the information acquisition period Tc to acquire the running data and environmental data of the transformer at regular intervals.

[0009] The preprocessing unit is used for preliminary analysis of the operation data and environmental data of the transformer, and respectively evaluates the insulation performance, mechanical risk and power stability of the transformer, and the insulation influence degree and mechanical influence degree;

[0010] The core analysis unit comprehensively analyzes the operation data and environmental data of the transformer by constructing a transformer monitoring model, wherein the transformer monitoring model comprises a comprehensive operation analysis submodel, an environmental influence analysis submodel and an operation risk analysis submodel;

[0011] By constructing the comprehensive operation analysis submodel, the influence functions of the insulation performance and mechanical risk of the transformer on the power stability are analyzed in sequence, and then the operation state index of the transformer is comprehensively output;

[0012] By establishing the environmental influence analysis submodel, the change functions between the insulation influence degree and the insulation performance, and the change functions between the mechanical influence degree and the mechanical risk are fitted and constructed, and then the comprehensive influence function between the environmental data of the transformer and the operation state index of the transformer is comprehensively constructed;

[0013] By establishing the operation risk analysis submodel, the model operation state index of the transformer is obtained by monitoring the real-time environmental data of the transformer, the comprehensive risk degree of the transformer is evaluated, and a risk prompt signal is generated;

[0014] The optimization design unit is used for analyzing the precision of the transformer monitoring model: by monitoring the real-time operation data of the transformer, the actual operation state index of the transformer is obtained, so as to compare and evaluate the precision of the transformer monitoring model, and generate an optimization management signal to optimize and upgrade the transformer monitoring model;

[0015] The early warning prompt unit is used for predicting the operation state of the transformer: meteorological prediction data are obtained and a prediction management instruction is generated and sent to the information collection unit, so as to evaluate the prediction risk degree of the transformer, and generate an early warning prompt signal to perform early warning processing on the transformer failure.

[0016] Further, the specific operation process of the preprocessing unit is:

[0017] A1: The operation data include insulation parameters, mechanical parameters and power parameters, wherein the insulation parameters include winding temperature, transformer oil temperature and liquid level; the mechanical parameters include gas pollutant content, vibration amplitude and vibration frequency; and the power parameters include input frequency, input voltage, output frequency and output voltage of the power supply;

[0018] The insulation parameters, mechanical parameters and power parameters are preliminarily analyzed by constructing a first parameter analysis model, so as to respectively evaluate the insulation performance, mechanical risk and power stability of the transformer;

[0019] A2: the environmental data comprises insulation influence parameters and mechanical influence parameters; wherein, the insulation influence parameters comprise environmental temperature and environmental humidity; the mechanical influence parameters comprise instantaneous wind speed and average wind speed;

[0020] The insulation influence parameters and the mechanical influence parameters are analyzed preliminarily by constructing the second parameter analysis model, so as to evaluate the insulation influence degree and the mechanical influence degree.

[0021] Further, the specific process of constructing the first parameter analysis model is as follows:

[0022] A1-1: input the index set P into the first parameter analysis model, wherein the index set P comprises n0 index elements;

[0023] Any index element is marked as i, and the value of the index element i is marked as Zi;

[0024] The standard interval of the index element i is set as [Bp1, Bp2];

[0025] The reference value Ci of the index element i is obtained by combining the value Zi of the index element i and the standard interval [Bp1, Bp2] thereof;

[0026] The evaluation coefficient Xp of the index set P is obtained and output by combining the reference values of the n0 index elements of the index set P;

[0027] A1-2: the insulation parameters, the mechanical parameters and the power supply parameters are analyzed preliminarily, so as to output the insulation performance evaluation coefficient, the mechanical risk evaluation coefficient and the power supply stability evaluation coefficient respectively;

[0028] The insulation parameters are taken as the index set P and input into the first parameter analysis model, the evaluation coefficient of the index set P is obtained and output comprehensively, and is marked as the insulation performance evaluation coefficient Xjy; the evaluation interval of the insulation performance evaluation coefficient Xjy is set, and the insulation performance of the transformer is evaluated by interval comparison;

[0029] The mechanical parameters are taken as the index set P and input into the first parameter analysis model, the evaluation coefficient of the index set P is obtained comprehensively, and is marked as the mechanical risk evaluation coefficient Xjx; the evaluation interval of the mechanical risk evaluation coefficient Xjx is set, and the mechanical risk of the transformer is evaluated by interval comparison;

[0030] The power supply parameters are taken as the index set P and input into the first parameter analysis model, the evaluation coefficient of the index set P is obtained and output comprehensively, and is marked as the power supply stability evaluation coefficient Xdw; the evaluation interval of the power supply stability evaluation coefficient Xdw is set, and the power supply stability of the transformer is evaluated by interval comparison.

[0031] Further, the specific process of constructing the second parameter analysis model is as follows:

[0032] A2-1: input the index set Q into the second parameter analysis model, the index set Q including n1 index elements;

[0033] mark any index element as j, collect the values of the index element j in N0 information collection periods, and mark the value of the index element j in any information collection period as Yj;

[0034] further obtain the mean value JZj of the index element j and the fluctuation coefficient BDj of the index element j, and set the standard interval of the index element j as [Bq1, Bq2];

[0035] obtain the reference value Cj of the index element j by combining the mean value JZj of the index element j and the standard interval [Bq1, Bq2] of the index element j;

[0036] obtain and output the influence coefficient Hq of the index set Q by combining the reference values and the fluctuation coefficients of the n1 index elements of the index set Q;

[0037] A2-2: preliminarily analyze the insulation influence parameters and the mechanical influence parameters, and thus output the insulation influence coefficient and the mechanical influence coefficient respectively;

[0038] wherein the insulation influence parameters are taken as the index set Q, input into the second parameter analysis model, the influence coefficient of the index set Q is comprehensively obtained and output, and is marked as the insulation influence coefficient Hjy; the evaluation interval of the insulation influence coefficient Hjy is set, and the insulation influence degree is evaluated by interval comparison;

[0039] the mechanical influence parameters are taken as the index set Q, input into the second parameter analysis model, the influence coefficient of the index set Q is comprehensively obtained and output, and is marked as the mechanical influence coefficient Hjx; the evaluation interval of the mechanical influence coefficient Hjx is set, and the mechanical influence degree is evaluated by interval comparison.

[0040] Further, the specific process of constructing the comprehensive operation analysis sub-model is as follows:

[0041] B1: collect the insulation performance evaluation coefficient Xjy, the mechanical risk evaluation coefficient Xjx and the power stability evaluation coefficient Xdw of the transformer corresponding to N0 information collection periods;

[0042] B1-1: establish the change curve Sa1 between the insulation performance evaluation coefficient Xjy of the transformer and the power stability evaluation coefficient Xdw, and generate the influence function F1 by fitting;

[0043] B1-2: establish the change curve Sa2 between the mechanical risk evaluation coefficient Xjx of the transformer and the power stability evaluation coefficient Xdw, and generate the influence function F2 by fitting;

[0044] B1-3: Further, through the insulation performance evaluation coefficient Xjy, the mechanical risk evaluation coefficient Xjx and the power supply stability evaluation coefficient Xdw, the operation state index Zyx of the transformer is comprehensively output;

[0045] B1-4: The power supply stability evaluation coefficient Xdw is converted into the influence function F1 and the influence function F2, and the operation state index Zyx of the transformer.

[0046] Further, the specific process of establishing the environmental impact analysis sub-model is:

[0047] B2-1: Fit the change function between the insulation influence degree and the insulation performance;

[0048] The insulation influence coefficient Hjy and the insulation performance evaluation coefficient Xjy of the transformer corresponding to N0 information collection periods are collected, the change curve Sb1 between the insulation influence coefficient Hjy and the insulation performance evaluation coefficient Xjy of the transformer is established, and the change function G1 is fitted and generated;

[0049] B2-2: Fit the change function between the mechanical influence degree and the mechanical risk;

[0050] The mechanical influence coefficient Hjx and the mechanical risk evaluation coefficient Xjx of the transformer corresponding to N0 information collection periods are collected, the change curve Sb2 between the mechanical influence coefficient Hjx and the mechanical risk evaluation coefficient Xjx of the transformer is established, and the change function G2 is fitted and generated;

[0051] B2-3: The comprehensive influence function GZ between the environmental data of the transformer and the operation state index of the transformer is comprehensively constructed, the insulation influence coefficient Hjy and the mechanical influence coefficient Hjx are substituted into the comprehensive influence function GZ, and the operation state index Zyx of the transformer is obtained.

[0052] Further, the specific process of establishing the operation risk analysis sub-model is:

[0053] B3-1: The model operation state index of the transformer is obtained by monitoring the real-time environmental data of the transformer;

[0054] The environmental data of the transformer is collected in real time and substituted into the No. 2 parameter analysis model, the actual insulation influence coefficient Hjy0 and the actual mechanical influence coefficient Hjx0 of the transformer at the current time node are preliminarily analyzed and obtained, and then substituted into the comprehensive influence function GZ, the model operation state index ZyxM of the transformer at the current time node is obtained;

[0055] B3-2: The evaluation interval of the model operation state index ZyxM is set, and the comprehensive risk degree of the transformer is evaluated by interval comparison;

[0056] B3-3: According to the signal level, the corresponding risk visualization presentation is carried out, so as to prompt the background technician to carry out corresponding fault processing on the transformer;

[0057] B3-301: When a first-level risk prompt signal is generated, no processing is performed;

[0058] B3-302: When a second-level risk prompt signal is generated, the analysis comparison judgment fault parameter is fed back, so as to carry out targeted repair processing on the fault parameter;

[0059] B3-303: When a third-level risk prompt signal is generated, synchronous fault repair processing is carried out on the insulation parameter, mechanical parameter and power parameter.

[0060] Further, the specific operation process of the optimization design unit is:

[0061] Set the model comparison period Tm, monitor the running data of the transformer in real time, and substitute it into the first parameter analysis model to preliminarily analyze and obtain the actual insulation performance evaluation coefficient Xjy1, the actual mechanical risk evaluation coefficient Xjx1 and the actual power stability evaluation coefficient Xdw1 of the transformer at the current time node;

[0062] Then substitute it into the comprehensive operation analysis sub-model to calculate, and then comprehensively obtain and output the actual running state index ZyxU of the transformer;

[0063] Compare the actual running state index ZyxU of the transformer with the model running state index ZyxM of the transformer, so as to obtain the model deviation index ΔZ;

[0064] Set the evaluation interval of the model deviation index ΔZ, compare and evaluate the accuracy of the transformer monitoring model, and generate a corresponding optimization management signal;

[0065] Then feed back the optimization management signal to the core processing unit, so as to carry out corresponding optimization and upgrading operation on the transformer monitoring model.

[0066] Further, the specific operation process of the early warning prompt unit is:

[0067] Set the prediction time node to obtain the meteorological prediction data of the prediction time node, and generate a prediction management instruction to send to the information collection unit;

[0068] The information collection unit receives the prediction management instruction and extracts the predicted value of the environmental data from the meteorological prediction data, and then outputs it to the preprocessing unit and the core analysis unit, substitutes the predicted value of the environmental data into the second parameter analysis model, preliminarily analyzes and obtains the predicted insulation influence coefficient HjyY and the predicted mechanical influence coefficient HjxY of the transformer at the prediction time node, and then substitutes it into the comprehensive influence function GZ to obtain the predicted running state index ZyxY of the transformer;

[0069] The risk interval of the predicted operation state index ZyxY of the transformer is set, the predicted risk degree of the transformer is evaluated, and a corresponding early warning prompt signal is generated.

[0070] The early warning prompt unit receives the early warning prompt signal to perform early warning processing on the transformer fault.

[0071] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present application are:

[0072] The present application collects the operation data and environmental data of the transformer through the information collection unit, and performs preliminary analysis through the preprocessing unit, and then constructs a transformer monitoring model through the core analysis unit, comprehensively analyzes and evaluates the operation state of the transformer from multiple angles of insulation performance, mechanical risk and power stability, improves the comprehensiveness of data processing, builds the influence function between environmental data and operation data by monitoring the working environment of the transformer, thereby extending the analysis of the comprehensive intelligent evaluation of the operation state of the transformer and generating a risk prompt signal, and improving the sensitivity of the transformer to environmental changes;

[0073] The present application optimizes the accuracy of the transformer monitoring model through the optimization design unit, generates an optimization management signal for model optimization and upgrading, thereby performing targeted processing on the fault indicators, improving the accuracy of fault repair and risk response, and through the early warning prompt unit, the predicted risk degree of the transformer is deeply analyzed and evaluated, and a early warning prompt signal is generated to realize early warning processing of the fault, so as to cope with the sudden impact of abnormal climate and environmental changes on the transformer. BRIEF DESCRIPTION OF DRAWINGS

[0074] Fig. 1 The module schematic diagram of the present application is shown;

[0075] Fig. 2 The overall flowchart of the present application is shown;

[0076] Fig. 3 The flowchart of the transformer monitoring model of the present application is shown. DETAILED DESCRIPTION

[0077] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0078] Embodiment 1:

[0079] As Figs. 1-3As shown, the transformer operating state comprehensive intelligent monitoring system comprises an information acquisition unit, a preprocessing unit, a core analysis unit, an optimization design unit and a warning prompt unit, wherein the information acquisition unit, the preprocessing unit, the core analysis unit, the optimization design unit and the warning prompt unit are signal-connected;

[0080] The working steps are as follows:

[0081] S1: The information acquisition unit acquires the operating data and environmental data of the transformer: set the information acquisition period Tc to acquire the operating data and environmental data of the transformer at regular intervals;

[0082] The operating data include insulation parameters, mechanical parameters and power supply parameters, wherein the insulation parameters include winding temperature, transformer oil temperature and liquid level; the mechanical parameters include gas pollutant content, vibration amplitude and vibration frequency; the power supply parameters include input frequency, input voltage, output frequency and output voltage of the power supply;

[0083] Among them, the winding temperature and the transformer oil temperature are acquired by a temperature sensor, and the liquid level of the insulating oil is acquired by monitoring the transformer oil tank; there are nw kinds of gas pollutants, such as dust, smoke, etc., the content of any pollutant is acquired by a gas detector, and the gas pollutant content is acquired by accumulation; the vibration amplitude and the vibration frequency are acquired by a vibration sensor; the input frequency, the input voltage, the output frequency and the output voltage of the power supply are acquired by monitoring and collecting the power meter;

[0084] S2: The preprocessing unit preliminarily analyzes the operating data and environmental data of the transformer, and respectively evaluates the insulation performance, mechanical risk and power supply stability of the transformer, as well as the insulation influence degree and mechanical influence degree;

[0085] A1: The insulation parameters, mechanical parameters and power supply parameters are preliminarily analyzed by constructing a first parameter analysis model, so as to respectively evaluate the insulation performance, mechanical risk and power supply stability of the transformer;

[0086] The specific process of constructing the first parameter analysis model is as follows:

[0087] A1-1: input the index set P into the first parameter analysis model, wherein the index set P comprises n0 index elements;

[0088] Mark any index element as i, and mark the numerical value of the index element i as Zi;

[0089] Set the standard interval of the index element i as [Bp1, Bp2];

[0090] Obtain the reference value Ci of the index element i by combining the numerical value Zi of the index element i and the standard interval [Bp1, Bp2] thereof:

[0091] Wherein, φi is the adjustment coefficient of the index element i, and φi is greater than 0; the adjustment coefficient φi is a constant value that makes the reference value Ci of the index element i always greater than 0, and the adjustment coefficient φi is obtained by presetting;

[0092] When the value Zi of the index element i is in the standard interval [Bp1, Bp2], the reference value Ci of the index element i is greater than or equal to the adjustment coefficient φi, indicating that the index element i is normal; otherwise, when the value Zi of the index element i is greater than or less than the standard interval [Bp1, Bp2], the reference value Ci of the index element i is less than the adjustment coefficient φi, indicating that the index element i is abnormal, and the greater the deviation of the value Zi from the standard interval [Bp1, Bp2], the smaller the reference value Ci of the index element i, indicating that the abnormality of the index element i is higher;

[0093] The evaluation coefficient Xp of the index set P is obtained and output by combining the reference values of the n0 index elements of the index set P:

[0094] Wherein, αi is the weight factor coefficient of the reference value Ci of the index element i, and αi is greater than 0; the weight factor coefficient αi is obtained by a large amount of experimental data measurement and presetting; the higher the reference value Ci of the index element i, the higher the evaluation coefficient Xp of the index set P, indicating that the state evaluation of the index set P is better;

[0095] A1-2: Preliminary analysis of insulation parameters, mechanical parameters and power supply parameters, so as to output insulation performance evaluation coefficient, mechanical risk evaluation coefficient and power supply stability evaluation coefficient respectively;

[0096] Wherein, the insulation parameters are taken as the index set P and input into the first parameter analysis model, the insulation parameters include winding temperature, transformer oil temperature and liquid level, then n0 is equal to 3, the values of the winding temperature, the transformer oil temperature and the liquid level are marked as Wr, Wy and Ly respectively; the winding temperature is taken as the index element i, then the reference value of the winding temperature is Cwr; the transformer oil temperature is taken as the index element i, then the reference value of the transformer oil temperature is Cwy; the liquid level is taken as the index element i, then the reference value of the liquid level is Cly; then the evaluation coefficient of the index set P is obtained and output comprehensively, and is marked as the insulation performance evaluation coefficient Xjy, and the evaluation interval of the insulation performance evaluation coefficient Xjy is set, and the insulation performance of the transformer is evaluated by interval comparison, the higher the insulation performance evaluation coefficient Xjy, the better the insulation performance of the transformer, indicating that the winding temperature, the transformer oil temperature and the liquid level are normal;

[0097] The mechanical parameters are taken as the index set P and input into the first parameter analysis model, the mechanical parameters including the gas pollutant content, the vibration amplitude and the vibration frequency, then n0=3, the values of the gas pollutant content, the vibration amplitude and the vibration frequency are marked as Vw, Zf and Zp respectively; the gas pollutant content is taken as the index element i, then the reference value of the gas pollutant content is Cvw; the vibration amplitude is taken as the index element i, then the reference value of the vibration amplitude is Cz f; the vibration frequency is taken as the index element i, then the reference value of the vibration frequency is Czp; then the evaluation coefficient of the index set P is obtained and marked as the mechanical risk evaluation coefficient Xjx, and the evaluation interval of the mechanical risk evaluation coefficient Xjx is set, and the mechanical risk of the transformer is evaluated through interval comparison, when the mechanical risk evaluation coefficient Xjx is higher, it means that the mechanical parameters are in the standard interval, and the influence of the gas pollutant and the vibration on the mechanical structure is smaller, so the mechanical risk of the transformer is lower;

[0098] Among them, the pollutants in the air, such as dust, smoke, etc., may cause the accumulation of transformer surface, thereby affecting the heat dissipation and insulation effect, too much vibration may cause the fastening structure to loosen, affect the electrical connection, and even cause mechanical failure;

[0099] The power supply parameters are taken as the index set P and input into the first parameter analysis model, the power supply parameters including the input frequency, the input voltage, the output frequency and the output voltage of the power supply, then n0=4, the values of the input frequency, the input voltage, the output frequency and the output voltage of the power supply are marked as Ip, Iv, Op and Ov respectively; the input frequency of the power supply is taken as the index element i, then the reference value of the input frequency of the power supply is Cip; the input voltage of the power supply is taken as the index element i, then the reference value of the input voltage of the power supply is Civ; the output frequency of the power supply is taken as the index element i, then the reference value of the output frequency of the power supply is Cop; the output voltage of the power supply is taken as the index element i, then the reference value of the output voltage of the power supply is Cov; then the evaluation coefficient of the index set P is obtained and output, and marked as the power supply stability evaluation coefficient Xdw, and the evaluation interval of the power supply stability evaluation coefficient Xdw is set, and the power supply stability of the transformer is evaluated through interval comparison, when the power supply stability evaluation coefficient Xdw is higher, it means that the power supply parameters are in the standard interval, and the input frequency, the input voltage, the output frequency and the output voltage of the power supply are normal, so the power supply stability performance of the transformer is better;

[0100] A2: The environmental data includes insulation influence parameters and mechanical influence parameters; wherein, the insulation influence parameters include environmental temperature and environmental humidity; the mechanical influence parameters include instantaneous wind speed and average wind speed;

[0101] The insulation influence parameters and the mechanical influence parameters are preliminarily analyzed through the construction of the second parameter analysis model, so as to evaluate the insulation influence degree and the mechanical influence degree;

[0102] The specific process of constructing the second parameter analysis model is as follows:

[0103] A2-1: input the index set Q into the second parameter analysis model, the index set Q including n1 index elements;

[0104] Mark any index element as j, collect the values of the index element j in N0 information collection periods, and mark the value of the index element j in any information collection period as Yj;

[0105] Further, the mean value JZj of the index element j is obtained: The higher the mean value JZj of the index element j, the higher the overall level of the index element j;

[0106] The fluctuation coefficient BDj of the index element j is obtained: The higher the fluctuation coefficient BDj of the index element j, the higher the fluctuation degree of the index element j;

[0107] The standard interval [Bq1, Bq2] of the index element j is set;

[0108] The reference value Cj of the index element j is obtained by combining the mean value JZj and the standard interval [Bq1, Bq2] of the index element j:

[0109] Wherein, φj is the adjustment coefficient of the index element j, and φj is greater than 0; the adjustment coefficient φj is a constant value that makes the reference value Cj of the index element j always greater than 0, and the adjustment coefficient φj is obtained by presetting;

[0110] When the mean value JZj of the index element j is in the standard interval [Bq1, Bq2], the smaller the reference value Cj of the index element j, the more normal the overall level of the index element j; on the contrary, when the mean value JZj of the index element j is greater than or less than the standard interval [Bq1, Bq2], the greater the reference value Cj of the index element j, the more abnormal the overall level of the index element j, and the greater the deviation degree of the mean value JZj from the standard interval [Bq1, Bq2], the greater the reference value Cj of the index element j, indicating that the abnormal degree of the index element j is higher;

[0111] The influence coefficient Hq of the index set Q is obtained and output by combining the reference values and fluctuation coefficients of the n1 index elements of the index set Q:

[0112] Wherein, βj is a weight factor coefficient of the reference value Cj of the index element j, and βj is greater than 0; the weight factor coefficient βj is obtained by a large number of experimental data; when the reference value Cj of the index element j is higher, the influence coefficient Hq of the index set Q is higher, indicating that the influence degree of the index set Q is higher;

[0113] A2-2: preliminary analysis of insulation influence parameters and mechanical influence parameters, thereby respectively outputting insulation influence coefficients and mechanical influence coefficients;

[0114] The insulation influence parameters are taken as the index set Q, and are input into the second parameter analysis model; the insulation influence parameters include environmental temperature and environmental humidity; n1 is equal to 2; the values of the environmental temperature and the environmental humidity are respectively marked as Wh and Sh; the environmental temperature is taken as the index element j, the reference value and the fluctuation coefficient of the environmental temperature are obtained; the environmental humidity is taken as the index element j, the reference value and the fluctuation coefficient of the environmental humidity are obtained; then the influence coefficient of the index set Q is obtained and output, and is marked as the insulation influence coefficient Hjy; the evaluation interval of the insulation influence coefficient Hjy is set, and the insulation influence degree is evaluated through interval comparison; when the insulation influence coefficient Hjy is higher, it indicates that the deviation degree of the insulation influence parameters from the standard interval is higher and the fluctuation degree is larger, and it indicates that the abnormal influence of the environmental temperature and the environmental humidity on the insulation performance of the transformer is higher, and the evaluation of the insulation influence degree is higher; wherein, the environmental temperature range should be moderate; too high or too low temperature will affect the performance of the transformer; too high temperature can cause aging of the winding insulation material, decrease of the tolerance, and increase of the fault risk; too low temperature can cause brittleness of the steel core and the insulation material, and affect the performance of the transformer; the environmental humidity has a direct influence on the insulation material and the equipment state; too high humidity can cause decrease of the insulation performance; too low humidity can cause accumulation of static electricity, and increase of the tripping risk;

[0115] The mechanical influence parameters are taken as the index set Q, and are input into the second parameter analysis model; the mechanical influence parameters include instantaneous wind speed and average wind speed; n1 is equal to 2; the values of the instantaneous wind speed and the average wind speed are respectively marked as Vs and Vf; the instantaneous wind speed is taken as the index element j, the reference value and the fluctuation coefficient of the instantaneous wind speed are obtained; the average wind speed is taken as the index element j, the reference value and the fluctuation coefficient of the average wind speed are obtained; then the influence coefficient of the index set Q is obtained and output, and is marked as the mechanical influence coefficient Hjx; the evaluation interval of the mechanical influence coefficient Hjx is set, and the mechanical influence degree is evaluated through interval comparison; when the mechanical influence coefficient Hjx is higher, it indicates that the deviation degree of the mechanical influence parameters from the standard interval is higher and the fluctuation degree is larger, and it indicates that the abnormal influence of the instantaneous wind speed and the average wind speed on the mechanical structure of the transformer is higher, and the evaluation of the mechanical influence degree is higher; wherein, the transformer can be affected by external actions such as the instantaneous wind speed and the average wind speed, causing mechanical vibration and noise, and thereby potentially threatening the operation stability of the transformer;

[0116] S3: The core analysis unit comprehensively analyzes the operation data and environmental data of the transformer by constructing a transformer monitoring model, wherein the transformer monitoring model comprises a comprehensive operation analysis sub-model, an environmental influence analysis sub-model, and an operation risk analysis sub-model;

[0117] S3-1: By constructing the comprehensive operation analysis sub-model, the influence function of the insulation performance and mechanical risk of the transformer on the power stability is analyzed first, and then the operation state index of the transformer is comprehensively output;

[0118] The specific process of constructing the comprehensive operation analysis sub-model is as follows:

[0119] B1: Collect the insulation performance evaluation coefficient Xjy, the mechanical risk evaluation coefficient Xjx, and the power stability evaluation coefficient Xdw of the transformer corresponding to N0 information collection periods;

[0120] B1-1: By establishing the change curve Sa1 between the insulation performance evaluation coefficient Xjy and the power stability evaluation coefficient Xdw of the transformer, and fitting to generate the influence function F1: Xdw=F1(Xjy), the insulation performance evaluation coefficient Xjy is substituted into the influence function F1, and the power stability evaluation coefficient Xdw is obtained;

[0121] B1-2: By establishing the change curve Sa2 between the mechanical risk evaluation coefficient Xjx and the power stability evaluation coefficient Xdw of the transformer, and fitting to generate the influence function F2: Xdw=F2(Xjx), the mechanical risk evaluation coefficient Xjx is substituted into the influence function F2, and the power stability evaluation coefficient Xdw is obtained;

[0122] Therefore, the power stability evaluation coefficient Xdw is expressed by the insulation performance evaluation coefficient Xjy and the mechanical risk evaluation coefficient Xjx of the transformer:

[0123] B1-3: Then, the operation state index Zyx of the transformer is comprehensively output by the insulation performance evaluation coefficient Xjy, the mechanical risk evaluation coefficient Xjx, and the power stability evaluation coefficient Xdw of the transformer:

[0124]

[0125] Wherein, ω1, ω2 and ω3 are weight factor coefficients of the insulation performance evaluation coefficient Xjy, the mechanical risk evaluation coefficient Xjx and the power supply stability evaluation coefficient Xdw of the transformer respectively, and ω1, ω2 and ω3 are all greater than 0; when the insulation performance evaluation coefficient Xjy of the transformer is higher, the mechanical risk evaluation coefficient Xjx is lower and the power supply stability evaluation coefficient Xdw is higher, then the operation state index Zyx of the transformer is higher, which indicates that the operation state of the transformer is better; through comprehensive analysis of the insulation performance, the mechanical risk and the power supply stability, the comprehensiveness of data processing is improved;

[0126] B1-4: The power supply stability evaluation coefficient Xdw is converted into the influence function F1 and the influence function F2, and then the operation state index Zyx of the transformer is:

[0127] S3-2: By establishing an environmental impact analysis submodel, a change function between the insulation influence degree and the insulation performance is first fitted and constructed, and a change function between the mechanical influence degree and the mechanical risk is further constructed, and then a comprehensive influence function between the environmental data of the transformer and the operation state index of the transformer is constructed comprehensively;

[0128] The specific process of establishing the environmental impact analysis submodel is as follows:

[0129] B2-1: The change function between the insulation influence degree and the insulation performance is fitted and constructed;

[0130] The insulation influence coefficient Hjy and the insulation performance evaluation coefficient Xjy of the transformer corresponding to N0 information collection periods are collected, a change curve Sb1 between the insulation influence coefficient Hjy and the insulation performance evaluation coefficient Xjy of the transformer is established, and a change function G1: Xjy=G1(Hjy) is fitted and generated, the insulation influence coefficient Hjy is substituted into the change function G1, and then the insulation performance evaluation coefficient Xjy is obtained;

[0131] B2-2: The change function between the mechanical influence degree and the mechanical risk is fitted and constructed;

[0132] The mechanical influence coefficient Hjx and the mechanical risk evaluation coefficient Xjx of the transformer corresponding to N0 information collection periods are collected, a change curve Sb2 between the mechanical influence coefficient Hjx and the mechanical risk evaluation coefficient Xjx of the transformer is established, and a change function G2: Xjx=G2(Hjx) is fitted and generated, the mechanical influence coefficient Hjx is substituted into the change function G1, and then the mechanical risk evaluation coefficient Xjx is obtained;

[0133] B2-3: a comprehensive influence function GZ between the environmental data of the transformer and the operation state index of the transformer is constructed, Zyx=GZ(Hjy, Hjx), the insulation influence coefficient Hjy and the mechanical influence coefficient Hjx are substituted into the comprehensive influence function GZ, and the operation state index Zyx of the transformer is obtained:

[0134]

[0135] S3-3: by establishing an operation risk analysis submodel, the model operation state index of the transformer is obtained by monitoring the real-time environmental data of the transformer, the comprehensive risk degree of the transformer is evaluated, and a risk prompt signal is generated;

[0136] The specific process of establishing the operation risk analysis submodel is as follows:

[0137] B3-1: the model operation state index of the transformer is obtained by monitoring the real-time environmental data of the transformer;

[0138] The environmental data of the transformer is collected in real time and substituted into the No. 2 parameter analysis model, the actual insulation influence coefficient Hjy0 and the actual mechanical influence coefficient Hjx0 of the transformer at the current time node are preliminarily analyzed and obtained, and then substituted into the comprehensive influence function GZ, the model operation state index ZyxM of the transformer at the current time node is obtained;

[0139]

[0140] B3-2: the evaluation interval of the model operation state index ZyxM is set, and the comprehensive risk degree of the transformer is evaluated by interval comparison, and the specific process is as follows:

[0141] The evaluation interval of the model operation state index ZyxM is preset as [M1, M2], when the model operation state index ZyxM is higher than the evaluation interval [M1, M2], it is determined that the operation state of the transformer is good, the comprehensive risk degree of the transformer is low, and a first-level risk prompt signal is generated; when the model operation state index ZyxM is in the evaluation interval [M1, M2], it is determined that the operation state of the transformer is general, the comprehensive risk degree of the transformer is intermediate, and a second-level risk prompt signal is generated; when the model operation state index ZyxM is lower than the evaluation interval [M1, M2], it is determined that the operation state of the transformer is poor, the comprehensive risk degree of the transformer is high, and a third-level risk prompt signal is generated;

[0142] B3-3: the corresponding risk visualization is presented according to the signal level, so as to prompt the background technical personnel to perform corresponding fault processing on the transformer;

[0143] B3-301: when a first-level risk prompt signal is generated, no processing is performed;

[0144] B3-302: When the secondary risk prompt signal is generated, feedback to the environmental impact analysis sub-model, and compare to determine the fault parameters:

[0145] B3-302-1: The actual insulation impact coefficient Hjy0 of the transformer at the current time node is obtained and substituted into the change function G1 to obtain the simulated insulation performance evaluation coefficient Xjy0 of the transformer at the current time node: Xjy0 = G1(Hjy0);

[0146] The evaluation threshold of the simulated insulation performance evaluation coefficient Xjy0 is set to m1, and when the simulated insulation performance evaluation coefficient Xjy0 is lower than the evaluation threshold m1, it is determined that the insulation parameter is faulty, and the insulation parameter is processed;

[0147] B3-302-2: The actual mechanical impact coefficient Hjx0 of the transformer at the current time node is obtained and substituted into the change function G2 to obtain the simulated mechanical risk evaluation coefficient Xjx0 of the transformer at the current time node: Xjx0 = G2(Hjx0);

[0148] The evaluation threshold of the simulated mechanical risk evaluation coefficient Xjx0 is set to m2, and when the simulated mechanical risk evaluation coefficient Xjx0 is higher than the evaluation threshold m2, it is determined that the mechanical parameter is faulty, and the mechanical parameter is processed;

[0149] B3-302-3: The actual insulation performance evaluation coefficient Xjy0 and the actual mechanical risk evaluation coefficient Xjx0 of the transformer at the current time node are substituted into the influence function F1 and the influence function F2 respectively to obtain the simulated power supply stability evaluation coefficient Xdw0 of the transformer at the current time node:

[0150] The evaluation threshold of the simulated power supply stability evaluation coefficient Xdw0 is set to m3, and when the simulated power supply stability evaluation coefficient Xdw0 is lower than the evaluation threshold m3, it is determined that the power supply parameter is faulty, and the power supply parameter is processed;

[0151] Through targeted repair processing for fault parameters, the accuracy of fault repair and risk response is improved;

[0152] B3-303: When the tertiary risk prompt signal is generated, the insulation parameter, the mechanical parameter and the power supply parameter are simultaneously subjected to fault repair processing;

[0153] S4: The optimization design unit is used to analyze the accuracy of the transformer monitoring model: the actual operation state index of the transformer is obtained by monitoring the real-time operation data of the transformer, and the accuracy of the transformer monitoring model is compared and evaluated, an optimization management signal is generated to optimize and upgrade the transformer monitoring model, and the specific operation process is:

[0154] S4-1: Set the model comparison period Tm, by monitoring the operation data of the transformer in real time, and substituting it into the first parameter analysis model, the actual insulation performance evaluation coefficient Xjy1, the actual mechanical risk evaluation coefficient Xjx1 and the actual power supply stability evaluation coefficient Xdw1 of the transformer at the current time node are obtained by preliminary analysis;

[0155] S4-2: Substitute it into the comprehensive operation analysis sub-model to calculate, and then obtain and output the actual operation state index ZyxU of the transformer:

[0156] S4-3: Compare and analyze the actual operation state index ZyxU of the transformer with the model operation state index ZyxM of the transformer, so as to obtain the model deviation index ΔZ:

[0157] When the difference between the actual operation state index ZyxU of the transformer and the model operation state index ZyxM is larger, the model deviation index ΔZ is higher, indicating that the deviation degree of the transformer monitoring model is higher, and then it is indicated that the precision of the transformer monitoring model is lower;

[0158] S4-4: Set the evaluation interval of the model deviation index ΔZ, compare and evaluate the precision of the transformer monitoring model, and generate the corresponding optimization management signal;

[0159] The evaluation interval of the preset model deviation index ΔZ has N1, and any evaluation interval is marked as Ur, wherein r is the serial number of the evaluation interval and 1≤r≤N1, when the model deviation index ΔZ is located in the evaluation interval Ur, the precision of the transformer monitoring model is r level, and r level optimization management signal is generated; When the model deviation index ΔZ is lower, the level r of the optimization management signal is higher, indicating that the precision of the transformer monitoring model is higher;

[0160] S4-5: The optimization management signal is fed back to the core processing unit, so as to perform corresponding optimization and upgrading operation on the transformer monitoring model, and the transformer monitoring model is re-modeled, so as to fit and correct the preset value of the weight factor coefficient in the model, realize the comprehensive intelligent monitoring of the transformer operation state through the working environment of the transformer, and improve the sensitivity of the transformer to environmental changes;

[0161] S5: The early warning prompt unit is used for predicting the operation state of the transformer: first, obtain meteorological prediction data through a professional meteorological platform, and generate prediction management instructions and send them to the information collection unit, so as to evaluate the prediction risk degree of the transformer, and generate early warning prompt signals to perform early warning processing on the transformer fault, and the specific operation process is:

[0162] S5-1: Set a prediction time node, obtain meteorological prediction data of the prediction time node through a professional meteorological platform, and generate a prediction management instruction and send it to the information collection unit;

[0163] S5-2: The information collection unit receives the prediction management instruction and extracts the predicted value of the environmental data through the meteorological prediction data, and then outputs it to the preprocessing unit and the core analysis unit, substitutes the predicted value of the environmental data into the second parameter analysis model, preliminarily analyzes to obtain the predicted insulation influence coefficient HjyY and the predicted mechanical influence coefficient HjxY of the transformer at the prediction time node, and then substitutes them into the comprehensive influence function GZ to obtain the predicted operation state index ZyxY of the transformer;

[0164] S5-3: Set the risk interval of the predicted operation state index ZyxY of the transformer, evaluate the predicted risk degree of the transformer, and generate a corresponding early warning prompt signal; the higher the predicted operation state index ZyxY, the better the predicted operation state of the transformer at the prediction time node, so that the evaluation of the predicted risk degree is lower;

[0165] S5-4: The early warning prompt unit receives the early warning prompt signal to perform early warning processing on the transformer fault, and optimizes the maintenance of the insulation parameters, mechanical parameters and power supply parameters to cope with the sudden impact of abnormal climate environment on the transformer.

[0166] In summary, the information collection unit collects the operation data and environmental data of the transformer, the preprocessing unit performs preliminary analysis, and the core analysis unit constructs a transformer monitoring model, which comprehensively analyzes and evaluates the operation state of the transformer from the aspects of insulation performance, mechanical risk and power supply stability, improves the comprehensiveness of data processing, comprehensively and intelligently evaluates the operation state of the transformer by monitoring the working environment of the transformer, and generates a risk prompt signal to improve the sensitivity of the transformer to environmental changes;

[0167] The present application optimizes the design unit to analyze the accuracy of the transformer monitoring model, generates an optimization management signal for model optimization and upgrading, processes the fault indicators, improves the accuracy of fault repair and risk response, and evaluates the predicted risk degree of the transformer through the early warning prompt unit to generate a warning prompt signal to realize early warning processing of the fault, so as to cope with the sudden impact of abnormal climate environment on the transformer.

[0168] The setting of the size of the interval and the threshold value is for easy comparison, and the size of the threshold value depends on the number of sample data and the base number set by the person skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameters and the quantized values.

[0169] The above formulas are all dimensionless values calculated, the formulas are obtained by collecting a large amount of data to simulate a formula of the most recent real situation, and preset parameters in the formulas are set by a person skilled in the art according to actual conditions;

[0170] The above merely describes a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can make equivalent replacement or change according to the technical scheme and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. A comprehensive intelligent monitoring system for the operating state of a transformer, characterized in that: The application relates to a transformer state monitoring and management system, which comprises an information collection unit, a preprocessing unit, a core analysis unit, an optimization design unit and a warning prompt unit, wherein the information collection unit, the preprocessing unit, the core analysis unit, the optimization design unit and the warning prompt unit are signal-connected; The information collection unit is used for collecting operation data and environmental data of the transformer; an information collection period Tc is set to collect the operation data and the environmental data of the transformer at regular time intervals; The preprocessing unit is used for preliminarily analyzing the operation data and the environmental data of the transformer, and respectively evaluating the insulation performance, the mechanical risk and the power supply stability of the transformer, and the insulation influence degree and the mechanical influence degree; The core analysis unit comprehensively analyzes the operation data and the environmental data of the transformer by constructing a transformer monitoring model, wherein the transformer monitoring model comprises a comprehensive operation analysis submodel, an environmental influence analysis submodel and an operation risk analysis submodel; The comprehensive operation analysis submodel is constructed to sequentially analyze the influence function of the insulation performance and the mechanical risk of the transformer on the power supply stability, and then comprehensively output the operation state index of the transformer; The environmental influence analysis submodel is constructed to fit the change function between the insulation influence degree and the insulation performance, and the change function between the mechanical influence degree and the mechanical risk, and then comprehensively construct the comprehensive influence function between the environmental data of the transformer and the operation state index of the transformer; The operation risk analysis submodel is constructed to obtain the model operation state index of the transformer by monitoring the real-time environmental data of the transformer, evaluate the comprehensive risk degree of the transformer and generate a risk prompt signal; The optimization design unit is used for analyzing the precision of the transformer monitoring model; the actual operation state index of the transformer is obtained by monitoring the real-time operation data of the transformer, so that the precision of the transformer monitoring model is compared and evaluated, an optimization management signal is generated to optimize and upgrade the transformer monitoring model; The warning prompt unit is used for predicting the operation state of the transformer; meteorological prediction data are obtained and a prediction management instruction is generated and sent to the information collection unit, so that the prediction risk degree of the transformer is evaluated, and a warning prompt signal is generated to perform early warning processing on the transformer fault.

2. The transformer operating state comprehensive intelligent monitoring system according to claim 1, characterized in that: The specific operation process of the preprocessing unit is as follows: A1: The operation data comprise insulation parameters, mechanical parameters and power supply parameters, wherein the insulation parameters comprise winding temperature, transformer oil temperature and liquid level; the mechanical parameters comprise gas pollutant content, vibration amplitude and vibration frequency; and the power supply parameters comprise input frequency, input voltage, output frequency and output voltage of the power supply; A first parameter analysis model is constructed to preliminarily analyze the insulation parameters, the mechanical parameters and the power supply parameters, so as to respectively evaluate the insulation performance, the mechanical risk and the power supply stability of the transformer; A2: The environmental data comprise insulation influence parameters and mechanical influence parameters; wherein the insulation influence parameters comprise environmental temperature and environmental humidity; and the mechanical influence parameters comprise instantaneous wind speed and average wind speed; A second parameter analysis model is constructed to preliminarily analyze the insulation influence parameters and the mechanical influence parameters, so as to evaluate the insulation influence degree and the mechanical influence degree.

3. The transformer operating state comprehensive intelligent monitoring system according to claim 2, characterized in that: The specific process of constructing the first parameter analysis model is as follows: A1-1: input the index set P into the first parameter analysis model, the index set P including n0 index elements; Mark any index element as i, and the value of the index element i as Zi; Set the standard interval of the index element i as [Bp1, Bp2]; Obtain the reference value Ci of the index element i by combining the value Zi of the index element i and its standard interval [Bp1, Bp2]; Obtain and output the evaluation coefficient Xp of the index set P by combining the reference values of the n0 index elements of the index set P; A1-2: preliminarily analyze the insulation parameters, mechanical parameters and power supply parameters, thereby outputting the insulation performance evaluation coefficient, the mechanical risk evaluation coefficient and the power supply stability evaluation coefficient respectively; Take the insulation parameters as the index set P and input them into the first parameter analysis model, comprehensively obtain and output the evaluation coefficient of the index set P, and mark it as the insulation performance evaluation coefficient Xjy, and then set the evaluation interval of the insulation performance evaluation coefficient Xjy, and evaluate the insulation performance of the transformer by interval comparison; Take the mechanical parameters as the index set P and input them into the first parameter analysis model, comprehensively obtain the evaluation coefficient of the index set P, and mark it as the mechanical risk evaluation coefficient Xjx, and then set the evaluation interval of the mechanical risk evaluation coefficient Xjx, and evaluate the mechanical risk of the transformer by interval comparison; Take the power supply parameters as the index set P and input them into the first parameter analysis model, comprehensively obtain and output the evaluation coefficient of the index set P, and mark it as the power supply stability evaluation coefficient Xdw, and then set the evaluation interval of the power supply stability evaluation coefficient Xdw, and evaluate the power supply stability of the transformer by interval comparison.

4. The transformer operating state comprehensive intelligent monitoring system according to claim 3, characterized in that: The specific process of constructing the second parameter analysis model is as follows: A2-1: input the index set Q into the second parameter analysis model, the index set Q including n1 index elements; Mark any index element as j, and collect the value of the index element j in the N0 information collection periods, and mark the value of the index element j in any information collection period as Yj; Then obtain the mean value JZj of the index element j and the fluctuation coefficient BDj of the index element j, and set the standard interval of the index element j as [Bq1, Bq2]; Obtain the reference value Cj of the index element j by combining the mean value JZj of the index element j and its standard interval [Bq1, Bq2]; Obtain and output the influence coefficient Hq of the index set Q by combining the reference values and the fluctuation coefficients of the n1 index elements of the index set Q; A2-2: preliminarily analyze the insulation influence parameters and the mechanical influence parameters, thereby outputting the insulation influence coefficient and the mechanical influence coefficient respectively; Take the insulation influence parameters as the index set Q and input them into the second parameter analysis model, comprehensively obtain and output the influence coefficient of the index set Q, and mark it as the insulation influence coefficient Hjy; then set the evaluation interval of the insulation influence coefficient Hjy, and evaluate the insulation influence degree by interval comparison; The mechanical influence parameter is taken as the index set Q and is input into the second parameter analysis model, an influence coefficient of the index set Q is comprehensively acquired and output, and is marked as a mechanical influence coefficient Hjx; an evaluation interval of the mechanical influence coefficient Hjx is set, and the mechanical influence degree is evaluated through interval comparison.

5. The transformer operating state comprehensive intelligent monitoring system according to claim 4, characterized in that: The specific process of constructing the comprehensive operation analysis sub-model is as follows: B1: collecting the insulation performance evaluation coefficient Xjy, the mechanical risk evaluation coefficient Xjx and the power stability evaluation coefficient Xdw of the transformer corresponding to N0 information collection periods; B1-1: a change curve Sa1 between the insulation performance evaluation coefficient Xjy and the power stability evaluation coefficient Xdw of the transformer is established, and an influence function F1 is generated by fitting; B1-2: a change curve Sa2 between the mechanical risk evaluation coefficient Xjx and the power stability evaluation coefficient Xdw of the transformer is established, and an influence function F2 is generated by fitting; B1-3: the operation state index Zyx of the transformer is comprehensively output through the insulation performance evaluation coefficient Xjy, the mechanical risk evaluation coefficient Xjx and the power stability evaluation coefficient Xdw of the transformer; B1-4: the power stability evaluation coefficient Xdw is converted into the influence function F1 and the influence function F2 to acquire the operation state index Zyx of the transformer.

6. The transformer operating state comprehensive intelligent monitoring system according to claim 5, characterized in that: The specific process of establishing the environmental influence analysis sub-model is as follows: B2-1: a change function between the insulation influence degree and the insulation performance is fitted and constructed; The insulation influence coefficient Hjy and the insulation performance evaluation coefficient Xjy of the transformer corresponding to N0 information collection periods are collected, a change curve Sb1 between the insulation influence coefficient Hjy and the insulation performance evaluation coefficient Xjy of the transformer is established, and a change function G1 is generated by fitting; B2-2: a change function between the mechanical influence degree and the mechanical risk is fitted and constructed; The mechanical influence coefficient Hjx and the mechanical risk evaluation coefficient Xjx of the transformer corresponding to N0 information collection periods are collected, a change curve Sb2 between the mechanical influence coefficient Hjx and the mechanical risk evaluation coefficient Xjx of the transformer is established, and a change function G2 is generated by fitting; B2-3: a comprehensive influence function GZ between the environmental data of the transformer and the operation state index of the transformer is comprehensively constructed, the insulation influence coefficient Hjy and the mechanical influence coefficient Hjx are substituted into the comprehensive influence function GZ to acquire the operation state index Zyx of the transformer.

7. The transformer operating state comprehensive intelligent monitoring system according to claim 6, characterized in that: The specific process of establishing the operation risk analysis sub-model is as follows: B3-1: the model operation state index of the transformer is acquired by monitoring real-time environmental data of the transformer; Real-time environmental data of the transformer are collected and substituted into the second parameter analysis model, the actual insulation influence coefficient Hjy0 and the actual mechanical influence coefficient Hjx0 of the transformer at a current time node are preliminarily analyzed and acquired, and then are substituted into the comprehensive influence function GZ to acquire the model operation state index ZyxM of the transformer at the current time node; B3-2: an evaluation interval of the model operation state index ZyxM is set, and the comprehensive risk degree of the transformer is evaluated through interval comparison; B3-3: According to the signal level, the corresponding risk visualization presentation is carried out, so as to prompt the background technician to carry out corresponding fault processing on the transformer; B3-301: When a first-level risk prompt signal is generated, no processing is performed; B3-302: When a second-level risk prompt signal is generated, the analysis comparison judgment fault parameter is fed back, so as to carry out targeted repair processing on the fault parameter; B3-303: When a third-level risk prompt signal is generated, synchronous fault repair processing is carried out on the insulation parameter, mechanical parameter and power parameter.

8. The transformer operating state comprehensive intelligent monitoring system according to claim 7, characterized in that: The specific operation process of the optimization design unit is: Set the model comparison period Tm, monitor the running data of the transformer in real time, and substitute it into the first parameter analysis model to preliminarily analyze and obtain the actual insulation performance evaluation coefficient Xjy1, the actual mechanical risk evaluation coefficient Xjx1 and the actual power stability evaluation coefficient Xdw1 of the transformer at the current time node, and then substitute them into the comprehensive operation analysis sub-model to calculate and further comprehensively obtain and output the actual running state index ZyxU of the transformer; Compare and analyze the actual running state index ZyxU of the transformer with the model running state index ZyxM of the transformer, so as to obtain the model deviation index ΔZ; Set the evaluation interval of the model deviation index ΔZ, compare and evaluate the accuracy of the transformer monitoring model, and generate a corresponding optimization management signal; The optimization management signal is fed back to the core processing unit, so as to carry out corresponding optimization and upgrading operation on the transformer monitoring model.

9. The transformer operating state comprehensive intelligent monitoring system according to claim 8, characterized in that: The specific operation process of the early warning prompt unit is: Set the prediction time node to obtain the meteorological prediction data of the prediction time node, and generate a prediction management instruction sent to the information collection unit; The information collection unit receives the prediction management instruction and extracts the predicted value of the environmental data through the meteorological prediction data, and then outputs it to the preprocessing unit and the core analysis unit, substitutes the predicted value of the environmental data into the second parameter analysis model to preliminarily analyze and obtain the predicted insulation influence coefficient HjyY and the predicted mechanical influence coefficient HjxY of the transformer at the prediction time node, and then substitutes them into the comprehensive influence function GZ to obtain the predicted running state index ZyxY of the transformer; Set the risk interval of the predicted running state index ZyxY of the transformer, evaluate the predicted risk degree of the transformer, and generate a corresponding early warning prompt signal; The early warning prompt unit receives the early warning prompt signal to carry out early warning processing on the transformer fault.

Images