A high-voltage generator, a high-voltage generator temperature regulation method and system

Abstract

The present application relates to the technical field of high-voltage generator, and particularly relates to a high-voltage generator, a high-voltage generator temperature adjusting method and system, the method comprising: S1, obtaining a detection temperature of a shell according to a temperature detection device on a high-voltage generator shell; obtaining a coil parameter and a coil current of the high-voltage generator and calculating a coil temperature; obtaining a gas pressure parameter of an insulating gas in the high-voltage generator and calculating a gas temperature; S2, inputting the detection temperature, the coil temperature and the gas temperature at a current time point into a trained neural network model, the model outputting a refrigeration level at the current time point and a predicted refrigeration level at a next time point; S3, controlling a corresponding refrigeration effect according to the refrigeration level at the current time point and the predicted refrigeration level at the next time point to adjust the temperature of the high-voltage generator.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of high-voltage generator, in particular to a high-voltage generator, a high-voltage generator temperature regulation method and system. BACKGROUND

[0002] At present, the extra-high voltage power transmission substation develops rapidly, and the safe and stable operation of the extra-high voltage GIS as the key equipment in the power system is crucial to the reliability and efficiency of the entire power grid. In order to ensure the normal operation of the equipment and study possible problems, it is necessary to carry out long-term live, partial discharge test on the extra-high voltage GIS product, and a high-voltage generator capable of outputting stable extra-high voltage is usually used to provide high-voltage power supply.

[0003] For example, a two-stage voltage regulator controller of an extra-high voltage DC generator with publication number CN108258922A, comprising: a rectifier for rectifying a three-phase alternating current power supply into a direct current power supply; a filter for filtering the rectified direct current power supply; an inverter for converting the filtered direct current power supply into alternating current; a medium frequency transformer for transforming the alternating current into high voltage alternating current; a voltage doubling circuit for rectifying and doubling the voltage of the transformed high voltage alternating current to output high voltage direct current.

[0004] Since the high-voltage generator usually works in a high-power state, a large amount of electrical energy is converted into heat energy, causing the internal temperature to rise. Even in a stable state, heat is generated due to resistance when current passes through components such as coils, especially in parts with large contact resistance. The structure of the extra-high voltage generator is complex and has high sealing performance, so it is difficult for heat to be quickly and effectively dissipated, making it difficult to regulate the temperature inside the high-voltage generator, which can easily cause local temperature accumulation and affect the service life of the high-voltage generator. SUMMARY

[0005] In order to solve the problem that the temperature inside the high-voltage generator is not easy to regulate, which can easily cause local temperature accumulation and affect the service life of the high-voltage generator, the present application provides a high-voltage generator, a high-voltage generator temperature regulation method and system.

[0006] In the first aspect, the present application provides a high-voltage generator temperature regulation method, which adopts the following technical scheme:

[0007] A high-voltage generator temperature regulation method, comprising the following steps:

[0008] S1, obtaining the detection temperature of the shell according to the temperature detection device on the shell of the high-voltage generator;

[0009] Obtaining the coil parameters and coil current of the high-voltage generator and calculating the coil temperature;

[0010] Acquire the gas pressure parameter of the insulation gas in the high-voltage generator and calculate the gas temperature;

[0011] S2, input the detection temperature, coil temperature and gas temperature at the current time point into the trained neural network model, and the model outputs the refrigeration level at the current time point and the predicted refrigeration level at the next time point;

[0012] S3, control the corresponding refrigeration effect according to the refrigeration level at the current time point and the predicted refrigeration level at the next time point to adjust the temperature of the high-voltage generator.

[0013] By adopting the above technical scheme, there are many factors affecting the inside of the high-voltage generator, the temperatures of multiple factors are predicted, and then the refrigeration effect is adjusted according to the predicted structure, so that the temperature in the high-voltage generator can be better adjusted, and the service life of the high-voltage generator is prolonged.

[0014] Optionally, the detection temperature of the shell is acquired according to the temperature detection device on the shell of the high-voltage generator, and the detection temperature of the shell at the i-th time point is calculated according to the temperature of the m-th temperature measurement point of the shell at the i-th time point and the corresponding weight of the m-th temperature measurement point.

[0015] The temperature of the m-th temperature measurement point of the shell at the i-th time point is acquired based on the temperature measurement sensor , m=1, 2, ……M, M is the total number of temperature measurement points, i is the time point, and m is the number of temperature measurement points;

[0016] The temperature of the m-th temperature measurement point of the shell at the i-th time point is set The corresponding weight , m=1, 2, ……M, M is the total number of temperature measurement points, i is the time point, and m is the number of temperature measurement points;

[0017] The detection temperature of the shell at the i-th time point is calculated The calculation formula is: , i is the time point.

[0018] By adopting the above technical scheme, the temperatures of multiple positions are measured, and the weights are configured according to the influence degree of different positions on the whole high-voltage generator, and the comprehensive detection temperature at a moment is calculated.

[0019] Optionally, in step S1, the coil parameters include coil resistance R, coil thermal conductivity K and coil surface area A, the coil temperature includes primary coil temperature and secondary coil temperature, and the calculation formula of the primary coil temperature at the i-th time point is:

[0020]

[0021] The calculation formula of the secondary coil temperature at the i-th time point is:

[0022]

[0023] Wherein, i is a time point, t represents a coil energization time, is a primary coil temperature at the i th time point, is a primary coil circuit at the i th time point, is an initial primary coil temperature, is a secondary coil temperature at the i th time point, is a secondary coil circuit at the i th time point, is an initial secondary coil temperature.

[0024] By adopting the technical scheme, the coil temperature in theory is obtained by calculating the heat generated by the coil and combining the parameters of the coil.

[0025] Optionally, in step S1, the gas pressure parameter of the insulation gas in the high-voltage generator is obtained, and the gas temperature is calculated, including:

[0026] An initial gas pressure parameter P and an initial gas temperature T are obtained;

[0027] A gas pressure parameter at the i th time point is obtained and a formula is constructed with the initial gas pressure parameter P and the initial gas temperature T:

[0028]

[0029] Wherein, i is a time point, V is the volume in the high-voltage generator, is a gas temperature at the i th time point.

[0030] Optionally, the refrigeration levels include zero level, first level, second level and third level, before model training, a judgment rule is established by comprehensively obtaining the detection temperature, the coil temperature and the gas temperature at each time point, the refrigeration level at each time point is judged based on the judgment rule, and the specific steps include:

[0031] A first difference value of the detection temperature and the coil temperature, a second difference value of the gas temperature and the coil temperature, and a third difference value of the coil temperature and the gas temperature at each time point are obtained respectively;

[0032] When the detection temperature, the coil temperature and the gas temperature are all less than a temperature threshold value, the refrigeration level is judged as zero level;

[0033] When one of the detection temperature, the coil temperature and the gas temperature is greater than the temperature threshold value, and at the same time, the difference value related thereto is less than a difference value threshold value, the refrigeration level is judged as first level;

[0034] When one of the detection temperature, the coil temperature and the gas temperature is greater than the temperature threshold value, and at the same time, the difference value related thereto is greater than the difference value threshold value, or when two of the detection temperature, the coil temperature and the gas temperature are greater than the temperature threshold value, the refrigeration level is judged as second level;

[0035] When the measured temperature, the coil temperature and the gas temperature are all greater than the temperature threshold value and / or the first difference, the second difference and the third difference are all greater than the difference threshold value, it is determined that the refrigeration level is level three.

[0036] Optionally, the training process of the neural network model comprises:

[0037] obtaining historical detection temperatures, historical coil temperatures and historical gas temperatures at all historical time points;

[0038] calculating historical refrigeration levels corresponding to the historical time points according to a determination rule of the refrigeration level;

[0039] constructing a data set from the historical detection temperatures, the historical coil temperatures, the historical gas temperatures and the historical refrigeration levels, and dividing the data set into a training set and a test set;

[0040] inputting the historical detection temperature, the historical coil temperature and the historical gas temperature of a historical time point into an initial prediction model for training, and the model output is a historical refrigeration level of the historical time point and a refrigeration level of a next historical time point;

[0041] testing the initial prediction model through the test set, and outputting a predicted refrigeration level of the historical time point and a predicted refrigeration level of the next historical time point;

[0042] determining a corresponding loss value based on the predicted refrigeration level of the historical time point and the predicted refrigeration level of the next historical time point output by the prediction model and a mean square error loss function, and determining whether the loss value is less than a threshold value;

[0043] If the loss value is less than the threshold value, stop training and use the initial prediction model as the corresponding prediction model;

[0044] If the loss value is not less than the threshold value, correct the parameters of the initial prediction model based on the loss value and update the prediction model.

[0045] Optionally, the step of controlling the corresponding refrigeration effect to adjust the temperature of the high-pressure generator according to the refrigeration level of the current time point and the predicted refrigeration level of the next time point comprises:

[0046] determining the level size of the refrigeration level of the current time point and the predicted refrigeration level of the next time point;

[0047] when the refrigeration level of the current time point is not less than the predicted refrigeration level of the next time point, retaining the refrigeration level of the current time point to control the corresponding refrigeration effect to adjust the temperature of the high-pressure generator;

[0048] When the cooling level of the current time point is less than the predicted cooling level of the next time point, the corresponding cooling effect is controlled by using the predicted cooling level to adjust the temperature of the high-voltage generator.

[0049] In a second aspect, a high-voltage generator temperature adjustment system comprises:

[0050] The detection module obtains the detection temperature of the shell detection point through a temperature detection device in the high-voltage generator, obtains the coil parameters of the high-voltage generator and detects the coil current through a current detection device, and detects the gas pressure parameters through a pressure sensor in the high-voltage generator.

[0051] The processing module is connected with the detection module, judges the cooling level corresponding to each historical time point by comprehensively considering the detection temperature, coil temperature and gas temperature obtained at all historical time points, inputs the detection temperature, coil temperature and gas temperature of each time point into a neural network model, and analyzes the cooling level of the current time point and the predicted cooling level of the next time point to generate control information.

[0052] The cooling module adjusts the cooling level according to the control information to adjust the temperature of the high-voltage generator.

[0053] In a third aspect, a high-voltage generator comprises a shell, a core, a primary coil, a secondary coil and the above-mentioned temperature adjustment system, the shell is filled with an insulating medium, the core is fixedly arranged in the shell, the primary coil is wound on the core, the shell is provided with a low-voltage input port connected with the primary coil, the secondary coil is wound on the side of the primary coil, and the shell is provided with a high-voltage output assembly connected with the secondary coil.

[0054] The system is arranged on the shell, the temperature detection device comprises a temperature sensor arranged on the position of the inner wall of the shell where high temperature is easy to occur, the current detection device and the pressure sensor are arranged in the shell.

[0055] The processing module is arranged outside the shell, and the temperature sensor, the current detection device and the pressure sensor are connected with the processing module through the shell.

[0056] The cooling module comprises a controller and a pipeline, the shell is provided with a cooling liquid injection port, the controller is arranged outside the shell and is used to control the injection of the cooling liquid from the cooling liquid injection port, one end of the pipeline is in communication with the cooling liquid injection port, and the pipeline is arranged between the core and the primary coil.

[0057] In summary, the present application has the beneficial technical effects as follows: the model is used to judge and predict the refrigeration level corresponding to the detection temperature, coil temperature and gas temperature, the temperature in the high-voltage generator is adjusted according to the refrigeration effect corresponding to the refrigeration level at the current time point and the next time point, the temperature in the high-voltage generator is flexibly adjusted according to the demand, the generation of excessively high temperature can be prevented to a certain extent in advance, and the service life of the high-voltage generator is prolonged. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 is a high-voltage generator temperature adjustment method flow diagram of an embodiment of the present application.

[0059] Figure 2 is a structural diagram of a high-voltage generator temperature adjustment system of an embodiment of the present application.

[0060] Figure 3 is a structural diagram of a high-voltage generator of an embodiment of the present application.

[0061] Wherein, 1, insulator; 2, high-voltage conducting rod; 3, secondary coil; 4, low-voltage input port; 5, shell; 6, primary coil; 7, cooling pipeline; 8, cooling liquid injection port; 9, iron core. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings Figures 1-3 The present application will be further described in detail.

[0063] Embodiment 1

[0064] Referring to Figure 1 , the high-voltage generator temperature adjustment method of the present embodiment comprises the following steps:

[0065] S1, obtaining the detection temperature of the shell detection point according to the temperature detection device in the high-voltage generator; specifically comprising:

[0066] obtaining the temperature of the mth temperature measurement point of the ith time point based on the temperature measurement sensor , m=1, 2, ……M, M is the total number of temperature measurement points, i is the time point, and m is the number of temperature measurement points;

[0067] setting the temperature of the mth temperature measurement point of the ith time point corresponding weight , m=1, 2, ……M, M is the total number of temperature measurement points, i is the time point, and m is the number of temperature measurement points;

[0068] the detection temperature of the ith time point The calculation formula is: , i is the time point.

[0069] Obtaining the coil parameters and coil current of the high-voltage generator and calculating the coil temperature; the coil parameters include coil resistance R, coil thermal conductivity K and coil surface area A, and the coil temperature includes the primary coil temperature and the secondary coil temperature, and the calculation formula of the primary coil temperature at the i-th time point is:

[0070]

[0071] The calculation formula of the secondary coil temperature at the i-th time point is:

[0072]

[0073] Wherein, i is the time point, t represents the coil energization time, is the primary coil temperature at the i-th time point, is the primary coil circuit at the i-th time point, is the initial primary coil temperature, is the secondary coil temperature at the i-th time point, is the secondary coil circuit at the i-th time point, is the initial secondary coil temperature.

[0074] Obtaining the gas pressure parameter of the insulating gas in the high-voltage generator and calculating the gas temperature includes:

[0075] Obtaining the initial gas pressure parameter P and the initial gas temperature T;

[0076] Obtaining the gas pressure parameter at the i-th time point , and constructing the formula with the initial gas pressure parameter P and the initial gas temperature T:

[0077]

[0078] Wherein, i is the time point, V is the volume in the high-voltage generator, is the gas temperature at the i-th time point.

[0079] S2, input the detection temperature, coil temperature and gas temperature at the current time point into the trained neural network model, and the model outputs the refrigeration level at the current time point and the predicted refrigeration level at the next time point;

[0080] The refrigeration level includes zero level, first level, second level and third level, before the model training, the judgment rule is established based on the obtained detection temperature, coil temperature and gas temperature at each time point, and the refrigeration level at each time point is judged based on the judgment rule, and the specific steps include:

[0081] Respectively obtaining the first difference value of the detection temperature and the coil temperature at each time point, the second difference value of the gas temperature and the coil temperature, and the third difference value of the coil temperature and the gas temperature;

[0082] When the detection temperature, the coil temperature and the gas temperature are all less than the temperature threshold value, it is determined that the refrigeration level is zero level; the zero level corresponds to the case of no refrigeration;

[0083] When one of the detection temperature, the coil temperature and the gas temperature is greater than the temperature threshold value, and the difference related thereto is less than the difference threshold value, it is determined that the refrigeration level is one level; at this time, only one of the three temperatures is abnormal, and there is no sudden rise in temperature, so the lower refrigeration level is enough;

[0084] When one of the detection temperature, the coil temperature and the gas temperature is greater than the temperature threshold value, and the difference related thereto is greater than the difference threshold value, or when two of the detection temperature, the coil temperature and the gas temperature are greater than the temperature threshold value, it is determined that the refrigeration level is two level; there are two abnormal temperatures, or one abnormal temperature and a sudden rise in temperature, so a slightly higher refrigeration level is needed at this time.

[0085] When the detection temperature, the coil temperature and the gas temperature are all greater than the temperature threshold value and / or the first difference, the second difference and the third difference are all greater than the difference threshold value, it is determined that the refrigeration level is three level. This is that all the temperatures are abnormal, or all the temperatures between them are abnormal, so the highest refrigeration level is needed to cool down as soon as possible to avoid damaging the high-voltage generator.

[0086] In this embodiment, the temperature difference of the shell temperature, the coil temperature and the insulation gas temperature at the same time point in the closed high-voltage generator will be different due to various factors. However, under normal circumstances, the coil temperature is usually the highest among the three. Due to the Joule heat generated when the current passes through the coil, the coil temperature may be significantly higher than the shell temperature and the insulation gas temperature.

[0087] The temperature difference between the coil temperature and the shell detection temperature is generally between 25-40 degrees Celsius, especially in high-voltage generators with high power and long running time, the temperature difference may be close to 40 degrees Celsius; the high-voltage generator is in a high-load running state, according to the internal heat dissipation, the temperature difference between the coil temperature and the insulation gas temperature is in the range of 30-50 degrees Celsius; the temperature difference between the shell detection temperature and the insulation gas temperature is relatively small, which may be between 5-20 degrees Celsius, and under the condition of uniform internal heat exchange, the temperature difference may be between 5-10 degrees Celsius; and in some special cases, such as poor gas flow, the temperature difference may be expanded to 15-20 degrees Celsius.

[0088] Therefore, the first difference threshold value between the coil temperature and the shell detection temperature is set to 55 degrees, the second difference between the coil temperature and the gas temperature is set to 65 degrees, and the third difference between the shell detection temperature and the gas temperature is set to 25 degrees.

[0089] The training process of the neural network model comprises:

[0090] obtaining historical detection temperatures, historical coil temperatures and historical gas temperatures at all historical time points;

[0091] calculating historical refrigeration levels corresponding to all historical time points according to a judgment rule of the refrigeration level;

[0092] constructing a data set from all historical detection temperatures, historical coil temperatures, historical gas temperatures and historical refrigeration levels, and dividing the data set into a training set and a test set;

[0093] inputting the historical detection temperature, the historical coil temperature and the historical gas temperature of a historical time point into an initial prediction model for training, and the model output is the historical refrigeration level of the historical time point and the refrigeration level of the next historical time point;

[0094] testing the initial prediction model through the test set, and outputting the predicted refrigeration level of the historical time point and the predicted refrigeration level of the next historical time point;

[0095] determining a corresponding loss value based on the predicted refrigeration level of the historical time point and the predicted refrigeration level of the next historical time point output by the prediction model and a mean square error loss function, and judging whether the loss value is less than a threshold value;

[0096] if the loss value is less than the threshold value, stopping the training and taking the initial prediction model as the corresponding prediction model;

[0097] if the loss value is not less than the threshold value, correcting the parameters of the initial prediction model based on the loss value and updating the prediction model.

[0098] S3, controlling the corresponding refrigeration effect to adjust the temperature of the high-pressure generator according to the refrigeration level of the current time point and the predicted refrigeration level of the next time point, and the specific steps comprise:

[0099] judging the level size of the refrigeration level of the current time point and the predicted refrigeration level of the next time point;

[0100] when the refrigeration level of the current time point is not less than the predicted refrigeration level of the next time point, keeping the refrigeration level of the current time point to control the corresponding refrigeration effect to adjust the temperature of the high-pressure generator; at this time, it is indicated that the temperature abnormality of the current time point is greater than that of the next time point, and the control of the refrigeration effect according to the refrigeration level of the current time point can be performed.

[0101] When the cooling level of the current time point is less than the predicted cooling level of the next time point, the corresponding cooling effect is controlled by the predicted cooling level to adjust the temperature of the high-voltage generator. Especially in the case that the current cooling level may be zero level and the next cooling level is greater than zero level, the cooling treatment of the current time point will be carried out by the cooling level of the next time point, which can effectively prevent or slow down the occurrence of abnormal conditions at the next time point, and is beneficial to prolong the service life of the high-voltage generator.

[0102] Embodiment 2

[0103] With reference to Figure 2 The difference between the present embodiment and embodiment 1 is that the present embodiment provides a high-voltage generator temperature adjusting system, comprising:

[0104] The detection module obtains the detection temperature of the shell detection point through the temperature detection device in the high-voltage generator, obtains the coil parameters of the high-voltage generator and detects the coil current through the current detection device, and detects the gas pressure parameters through the pressure sensor in the high-voltage generator;

[0105] The processing module is connected with the detection module, judges the cooling level corresponding to each historical time point by comprehensively analyzing the detection temperature, coil temperature and gas temperature obtained at all historical time points; inputs the detection temperature, coil temperature and gas temperature of each time point into the neural network model, and the model outputs the cooling level of the current time point and the predicted cooling level of the next time point, and generates control information by analyzing the cooling level of the current time point and the predicted cooling level of the next time point;

[0106] The cooling module adjusts the cooling level according to the control information to adjust the temperature of the high-voltage generator.

[0107] Embodiment 3

[0108] With reference to Figure 3The embodiment is different from the embodiments 1 and 2 in that the embodiment provides a high-voltage generator which comprises a shell, a core, a primary coil, a secondary coil and the temperature adjusting system, the shell is filled with an insulating medium, the insulating medium is 0.5 MPa SF6 gas, the core is fixedly arranged in the shell, the primary coil is wound on the core, a low-voltage input port connected with the primary coil is arranged on the shell, the secondary coil is wound on the periphery of the primary coil, and a high-voltage output assembly connected with the secondary coil is arranged on the shell; the high-voltage output assembly in the embodiment adopts a high-voltage conducting rod and an insulator, the high-voltage conducting rod connects the secondary coil and the insulator, the insulator is embedded on the shell, and a device requiring a high-voltage power supply is connected from the high-voltage output assembly. The low-voltage coil is wound by 10*20 copper wires, the low-voltage coil is wound for 100 turns, the resistance is 0.01Ω, and the low-voltage input current is 195A; the high-voltage coil adopts a 3-winding structure and is used in parallel, the high-voltage winding is wound by copper wires with a diameter of 2 mm, high voltage is induced through the electromagnetic induction principle, U1 / U2=N1 / N2 is obtained, each group of coils is wound for 6500 turns, the resistance of each group of coils is 100Ω, and the resistance of the 3 groups of windings in parallel is 33.3Ω. The low-voltage input voltage is 0-20 kV, the high-voltage output voltage is 1300 kV, the high-voltage output current is 3A, and the highest capacitive load is 20000 pF,

[0109] The system is arranged on the shell, the temperature detection device comprises a temperature sensor, the temperature sensor is arranged on the inner wall of the shell, through simulation of the entire high-voltage generator, the corresponding points prone to abnormally high temperature on the inner wall of the shell are found, and the temperature sensor is arranged corresponding to the points; the current detection device and the pressure sensor are arranged in the shell;

[0110] In another embodiment, a temperature measuring optical fiber can also be arranged, a gas flow channel is arranged in the secondary coil, the temperature measuring optical fiber is arranged in the gas flow channel of the secondary coil, then the temperature measuring optical fiber is used for monitoring the abnormal points of the secondary coil, and analysis is performed in combination with the calculated theoretical value.

[0111] The processing module is arranged outside the shell, the temperature measuring optical fiber, the temperature sensor, the current detection device and the pressure sensor all pass through the shell and are connected with the processing module;

[0112] The refrigeration module comprises a controller and a pipeline, the shell is provided with a cooling liquid injection port, the controller is arranged outside the shell and is used for controlling injection of the cooling liquid from the cooling liquid injection port, one end of the pipeline is in communication with the cooling liquid injection port, the pipeline is arranged between the core and the primary coil, the diameter of the pipeline is 20 mm, the pipeline is filled with circulating cooling liquid, and the controller is used for controlling the flow rate of the cooling liquid and the temperature of the cooling liquid, so that the high-voltage generator is cooled, and the temperature in the high-voltage generator is adjusted.

[0113] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. The methods disclosed in the embodiments are described simply because they correspond to the systems disclosed in the embodiments; relevant details can be found in the method section.

[0114] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0115] The above-disclosed embodiments are merely preferred embodiments of the present invention, but the present invention is not limited thereto. Any non-creative variations that can be conceived by those skilled in the art, as well as any improvements and modifications made without departing from the principles of the present invention, should fall within the protection scope of the present invention.

Claims

1. A method of temperature regulation of a high-voltage generator, characterized in that, The method comprises the following steps: S1, obtaining a detected temperature of the shell according to a temperature detection device on the shell of the high-voltage generator; obtaining the coil parameters and the coil current of the high-voltage generator and calculating the coil temperature; obtaining the gas pressure parameter of the insulating gas in the high-voltage generator and calculating the gas temperature; S2, inputting the detected temperature, the coil temperature and the gas temperature at the current time point into the trained neural network model, and the model outputs the refrigeration level at the current time point and the predicted refrigeration level at the next time point; S3, controlling the corresponding refrigeration effect according to the refrigeration level at the current time point and the predicted refrigeration level at the next time point to adjust the temperature of the high-voltage generator; The refrigeration level includes zero level, first level, second level and third level. Before the model is trained, a judgment rule is established based on the obtained detected temperature, coil temperature and gas temperature at each time point. The refrigeration level at each time point is judged based on the judgment rule. The specific steps include: respectively obtaining a first difference value of the detected temperature and the coil temperature, a second difference value of the gas temperature and the coil temperature, and a third difference value of the coil temperature and the gas temperature at each time point; when the detected temperature, the coil temperature and the gas temperature are all less than a temperature threshold value, the refrigeration level is judged to be zero level; when one of the detected temperature, the coil temperature and the gas temperature is greater than the temperature threshold value, and at the same time, the difference value related thereto is less than a difference threshold value, the refrigeration level is judged to be first level; when one of the detected temperature, the coil temperature and the gas temperature is greater than the temperature threshold value, and at the same time, the difference value related thereto is greater than the difference threshold value, or when two of the detected temperature, the coil temperature and the gas temperature are greater than the temperature threshold value, the refrigeration level is judged to be second level; when the detected temperature, the coil temperature and the gas temperature are all greater than the temperature threshold value and / or the first difference value, the second difference value and the third difference value are all greater than the difference threshold value, the refrigeration level is judged to be third level.

2. The high-voltage generator temperature regulation method according to claim 1, characterized in that, The method comprises the following steps: obtaining, by a temperature measuring sensor, a temperature of an mth temperature measuring point of a shell at an ith time point , m = 1, 2, …, M, M is the total number of temperature measuring points, i is the time point, and m is the number of temperature measuring points; set the temperature of the mth temperature measurement point at the ith time point corresponding weight , m = 1, 2, …, M, M is the total number of temperature measurement points, i is the time point, and m is the number of temperature measurement points the detection temperature at the i-th time point The calculation formula is: , i is the time point.

3. The method of claim 1, wherein the temperature of the high voltage generator is adjusted by, In step S1, the coil parameters include coil resistance R, coil thermal conductivity K and coil surface area A. The coil temperature includes primary coil temperature and secondary coil temperature. The calculation formula of the primary coil temperature at the i-th time point is: The calculation formula of the secondary coil temperature at the i-th time point is: where i is a time point, t represents a coil energizing time, is a primary coil temperature at the i-th time point, is a primary coil current at the i-th time point, is an initial primary coil temperature, is a secondary coil temperature at the i-th time point, is a secondary coil current at the i-th time point, is an initial secondary coil temperature.

4. The high-voltage generator temperature regulation method according to claim 1, characterized in that, In step S1, obtaining the gas pressure parameter of the insulating gas in the high-voltage generator and calculating the gas temperature comprises: obtaining an initial gas pressure parameter P and an initial gas temperature T; obtaining the air pressure parameter at the i-th time point and the initial air pressure parameter P and the initial air temperature T to form an equation: wherein, V is the volume in the high voltage generator at the i time point, is the gas temperature at the i time point.

5. The high-voltage generator temperature regulation method according to claim 1, characterized in that, The training process of the neural network model comprises: obtaining historical detected temperature, historical coil temperature and historical gas temperature at all historical time points; calculating the corresponding historical refrigeration level of all historical time points according to the judgment rule of the refrigeration level; constructing a data set with all the historical detected temperature, historical coil temperature, historical gas temperature and historical refrigeration level, and dividing the data set into a training set and a test set; inputting the historical detected temperature, historical coil temperature and historical gas temperature at a certain historical time point into an initial prediction model for training, and the model output is the historical refrigeration level at the certain historical time point and the refrigeration level at the next historical time point; testing the initial prediction model through the test set, and outputting the predicted refrigeration level at the certain historical time point and the predicted refrigeration level at the next historical time point; determining a corresponding loss value based on the predicted refrigeration level of the certain historical time point, the predicted refrigeration level of the next historical time point and the mean square error loss function, judging whether the loss value is less than a threshold value; if the loss value is less than the threshold value, stopping the training and taking the initial prediction model as the corresponding prediction model; if the loss value is not less than the threshold value, correcting the parameters of the initial prediction model based on the loss value and updating the prediction model.

6. The method of claim 5, wherein the temperature of the high-voltage generator is adjusted by, The step of controlling the corresponding refrigeration effect to adjust the temperature of the high-voltage generator according to the refrigeration level of the current time point and the predicted refrigeration level of the next time point includes: judging the level of the refrigeration level of the current time point and the predicted refrigeration level of the next time point; when the refrigeration level of the current time point is not less than the predicted refrigeration level of the next time point, keeping the refrigeration level of the current time point to control the corresponding refrigeration effect to adjust the temperature of the high-voltage generator; when the refrigeration level of the current time point is less than the predicted refrigeration level of the next time point, using the predicted refrigeration level to control the corresponding refrigeration effect to adjust the temperature of the high-voltage generator.

7. A temperature regulation system for a high voltage generator, characterized by comprising: a detection module, which acquires the detection temperature of the shell detection point through the temperature detection device in the high-voltage generator, acquires the coil parameters of the high-voltage generator and detects the coil current through the current detection device, and detects the gas pressure parameters through the pressure sensor in the high-voltage generator; a processing module, which is connected with the detection module, judges the refrigeration level of each historical time point respectively by comprehensively acquiring the detection temperature, the coil temperature and the gas temperature of all the historical time points; inputs the detection temperature, the coil temperature and the gas temperature of each time point into a neural network model, the model outputs the refrigeration level of the current time point and the predicted refrigeration level of the next time point, and analyzes the refrigeration level of the current time point and the predicted refrigeration level of the next time point to generate control information; a refrigeration module, which adjusts the refrigeration level according to the control information to adjust the temperature of the high-voltage generator; The refrigeration level includes zero level, first level, second level and third level. Before the model training, the judgment rule is established by comprehensively acquiring the detection temperature, the coil temperature and the gas temperature of each time point, the refrigeration level of each time point is judged based on the judgment rule, and the specific steps include: respectively acquiring the first difference value of the detection temperature and the coil temperature, the second difference value of the gas temperature and the coil temperature, and the third difference value of the coil temperature and the gas temperature in each time point; when the detection temperature, the coil temperature and the gas temperature are all less than the temperature threshold value, the refrigeration level is judged as zero level; when one of the detection temperature, the coil temperature and the gas temperature is greater than the temperature threshold value and the difference value related thereto is less than the difference value threshold value at the same time, the refrigeration level is judged as first level; when one of the detection temperature, the coil temperature and the gas temperature is greater than the temperature threshold value and the difference value related thereto is greater than the difference value threshold value at the same time, or when two of the detection temperature, the coil temperature and the gas temperature are greater than the temperature threshold value, the refrigeration level is judged as second level; when the detection temperature, the coil temperature and the gas temperature are all greater than the temperature threshold value and / or the first difference value, the second difference value and the third difference value are all greater than the difference value threshold value, the refrigeration level is judged as third level.

8. A high-voltage generator, characterized by The temperature regulating system comprises a shell, a core, a primary coil, a secondary coil and the temperature regulating system of claim 7, the shell is filled with an insulating medium, the core is fixedly arranged in the shell, the primary coil is wound on the core, a low-voltage input port connected with the primary coil is arranged on the shell, the secondary coil is wound on the side of the primary coil, and a high-voltage output assembly connected with the secondary coil is arranged on the shell. The system is arranged on the shell, the temperature detection device comprises a temperature sensor arranged on the inner wall of the shell at a position where a high temperature point is prone to occur, and the current detection device and the pressure sensor are arranged in the shell. The processing module is arranged outside the shell, and the temperature sensor, the current detection device and the pressure sensor are connected with the processing module through the shell. The refrigeration module comprises a controller and a pipeline, a cooling liquid injection port is formed in the shell, the controller is arranged outside the shell and is used for controlling injection of the cooling liquid from the cooling liquid injection port, one end of the pipeline is in communication with the cooling liquid injection port, and the pipeline is arranged between the core and the primary coil.

Images