A temperature and humidity regulation method and system for a new energy substation

Abstract

The present application relates to the technical field of temperature and humidity regulation of new energy transformer substations, and particularly relates to a temperature and humidity regulation method and system for new energy transformer substations, which realizes the deep integration of temperature and humidity regulation and fault tolerance through a complete control process of data acquisition, fuzzy control prerequisite variable calculation, membership value calculation, control law synthesis, hierarchical start-stop control strategy, fan fault diagnosis, fan fault handling, and closed-loop circulation, and can accurately adapt to the strong coupling characteristics of the multi-chamber environment of new energy transformer substations.

Description

Technical Field

[0001] This invention relates to the field of temperature and humidity control technology for new energy substations, and particularly to a method and system for temperature and humidity control in new energy substations. Background Technology

[0002] Key equipment in new energy substations, such as secondary protection, control, and communication systems, have stringent requirements for operating temperature and humidity. Precise temperature and humidity control directly affects the stability and lifespan of the equipment. Currently, the industry commonly uses traditional distributed PID control or simple threshold start-stop control schemes. These schemes have a core flaw: they lack the ability to adapt to the strong coupling characteristics of multi-chamber environments and the fault-tolerant control mechanism after fan failure.

[0003] In traditional solutions, the temperature and humidity control of the transformer room, high-voltage room, and low-voltage room is independent, failing to consider the mutual influence of environmental parameters in each room. This makes it difficult to cope with complex and ever-changing operating conditions, resulting in insufficient control precision and frequent local overheating or condensation problems. Furthermore, fan fault detection relies solely on thermal relays or air switches, which cannot accurately pinpoint the fault type. Moreover, a single fan failure can cause a power outage on the entire circuit, shutting down other working fans on the same circuit and paralyzing the temperature and humidity control system. In severe cases, this can lead to high-temperature tripping of the substation protection system, affecting overall power supply safety.

[0004] Therefore, there is an urgent need for an intelligent temperature and humidity control solution that can adapt to the strong coupling characteristics of multiple chambers and has intelligent fault diagnosis and fault-tolerant regulation functions for fan failures, so as to solve the core defects of traditional control methods. Summary of the Invention

[0005] The technical problem to be solved by this invention is to provide a method and system for temperature and humidity control in new energy substations, which can solve the core problems of insufficient control accuracy in multi-room strongly coupled environments and easy system paralysis after fan failure in traditional control methods, and realize precise control of temperature and humidity in multiple rooms and fault-tolerant operation.

[0006] To solve the above-mentioned technical problems, the first technical solution adopted by the present invention is as follows:

[0007] A method for temperature and humidity control in a new energy substation includes the following steps:

[0008] S1. Collect temperature and humidity values ​​from multiple electrical rooms within the substation;

[0009] S2. Based on the temperature and humidity values ​​of multiple electrical rooms collected in step S1, calculate multiple fuzzy control prerequisite variables for the TS fuzzy model.

[0010] S3. Based on the fuzzy control premise variables calculated in step S2, calculate the membership values ​​of each fuzzy control premise variable on each fuzzy set in the TS fuzzy model.

[0011] S4. Based on the membership values ​​calculated in step S3, synthesize the global control law through the TS fuzzy model to generate control commands for the coordinated control of the fans in each electrical room.

[0012] S5. Based on the control instructions generated in step S4, execute the layered start-stop control strategy for fans in different locations in each electrical room;

[0013] S6. Collect the current and voltage values ​​of the fans in each electrical room, and perform fault diagnosis on each fan based on the collected current and voltage values ​​to determine whether each fan has malfunctioned.

[0014] S7. If the fan does not malfunction, return to step S1 and continue to the next cycle; if the fan malfunctions, proceed to step S8.

[0015] S8. Disconnect the faulty fan and reconstruct the TS fuzzy model online;

[0016] S9. Based on the reconstructed TS fuzzy model, readjust the operating time of the remaining normal fans;

[0017] S10. Return to step S1 and continue executing the next control loop.

[0018] The second technical solution adopted in this invention is:

[0019] A temperature and humidity control system for a new energy substation, used to implement the above-mentioned temperature and humidity control method, the temperature and humidity control system comprising:

[0020] Multiple temperature and humidity sensors are arranged in multiple electrical rooms to collect temperature and humidity values ​​in each electrical room;

[0021] Multiple fans are located in multiple electrical rooms;

[0022] Multiple current transformers and multiple voltage transformers are installed in the power supply circuit of each wind turbine to collect the current and voltage values ​​of each wind turbine.

[0023] An edge controller, used to run a multivariable TS fuzzy control algorithm, is electrically connected to multiple temperature and humidity sensors, multiple fans, multiple current transformers, and multiple voltage transformers to execute the steps of the above method.

[0024] The beneficial effects of this invention are as follows:

[0025] Through the complete control process of data acquisition, calculation of fuzzy control premise variables, calculation of membership values, synthesis of control laws, hierarchical start-stop control strategy, fan fault diagnosis, fan fault handling, and closed-loop circulation, this solution realizes the deep integration of temperature and humidity regulation and fault tolerance, and can accurately adapt to the strong coupling characteristics of the multi-room environment of new energy substations; the global control law synthesized by the T-S fuzzy model enables the coordinated operation of multi-room fans, greatly improving the accuracy and consistency of temperature and humidity regulation, and effectively avoiding local overheating or condensation problems; the hierarchical start-stop control strategy can differentially control the operation of fans according to the structural characteristics and temperature and humidity requirements of different electrical rooms, reducing the frequent start-stop phenomenon of fans, which not only reduces energy consumption but also extends the service life of fans; the fault diagnosis and fault tolerance control link can quickly identify fan faults and only cut off the faulty fans. Through model reconstruction and adjustment of the remaining fan operation time, it avoids the risk of system paralysis caused by all shutdowns due to a single fault in traditional control methods, ensuring the continuity and stability of temperature and humidity control, and overall enhancing the adaptive ability and operation reliability of the system, providing a strong guarantee for the safe and stable operation of key equipment in new energy substations. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is the step flowchart of the temperature and humidity regulation method for the new energy substation of the present invention;

[0027] Figure 2 is the overall connection block diagram of the temperature and humidity regulation system for the new energy substation of the present invention;

[0028] Figure 3 is the design flowchart of the multi-variable T-S fuzzy controller for the temperature and humidity regulation method for the new energy substation of the present invention;

[0029] Figure 4 is the hierarchical start-stop strategy flowchart of the temperature and humidity regulation method for the new energy substation of the present invention;

[0030] Figure 5 is the schematic diagram of fault diagnosis and only cutting off the faulty fan for the temperature and humidity regulation method for the new energy substation of the present invention;

[0031] Figure 6 is the software flowchart of the temperature and humidity regulation method for the new energy substation of the present invention;

[0032] Figure 7 is the connection block diagram of the temperature and humidity regulation system for the new energy substation of the present invention;

[0033] Reference Signs Explanation:

[0034] 1. Temperature and humidity sensor; 2. Fan; 3. Current transformer; 4. Voltage transformer; 5. Edge controller. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0036] Please refer to Figure 1 The first technical solution adopted in this invention is:

[0037] A method for temperature and humidity control in a new energy substation includes the following steps:

[0038] S1. Collect temperature and humidity values ​​from multiple electrical rooms within the substation;

[0039] S2. Based on the temperature and humidity values ​​of multiple electrical rooms collected in step S1, calculate multiple fuzzy control prerequisite variables for the TS fuzzy model.

[0040] S3. Based on the fuzzy control premise variables calculated in step S2, calculate the membership values ​​of each fuzzy control premise variable on each fuzzy set in the TS fuzzy model.

[0041] S4. Based on the membership values ​​calculated in step S3, synthesize the global control law through the TS fuzzy model to generate control commands for the coordinated control of the fans in each electrical room.

[0042] S5. Based on the control instructions generated in step S4, execute the layered start-stop control strategy for fans in different locations in each electrical room;

[0043] S6. Collect the current and voltage values ​​of the fans in each electrical room, and perform fault diagnosis on each fan based on the collected current and voltage values ​​to determine whether each fan has malfunctioned.

[0044] S7. If the fan does not malfunction, return to step S1 and continue to the next cycle; if the fan malfunctions, proceed to step S8.

[0045] S8. Disconnect the faulty fan and reconstruct the TS fuzzy model online;

[0046] S9. Based on the reconstructed TS fuzzy model, readjust the operating time of the remaining normal fans;

[0047] S10. Return to step S1 and continue executing the next control loop.

[0048] As can be seen from the above description, the beneficial effects of the present invention are as follows:

[0049] This solution realizes the deep integration of temperature and humidity regulation and fault tolerance through a complete control process including data acquisition, calculation of fuzzy control premise variables, calculation of membership values, synthesis of control laws, hierarchical start-stop control strategy, fan fault diagnosis, fan fault handling, and closed-loop circulation. It can accurately adapt to the strong coupling characteristics of the multi-room environment in new energy substations. The global control law synthesized by the T-S fuzzy model enables the coordinated operation of multi-room fans, greatly improving the accuracy and consistency of temperature and humidity regulation, and effectively avoiding local overheating or condensation problems. The hierarchical start-stop control strategy can differentially control the operation of fans according to the structural characteristics and temperature and humidity requirements of different electrical rooms, reducing the frequent start-stop phenomenon of fans, thereby reducing energy consumption and extending the service life of fans. The fault diagnosis and fault tolerance control link can quickly identify fan faults and only cut off the faulty fans. Through model reconstruction and adjustment of the remaining fan operation time, it avoids the risk of system paralysis caused by all shutdowns due to a single fault in the traditional control method, ensuring the continuity and stability of temperature and humidity control, and overall enhancing the adaptive ability and operation reliability of the system, providing a strong guarantee for the safe and stable operation of key equipment in new energy substations.

[0050] Further, the multiple fuzzy control premise variables in step S2 include the maximum global temperature deviation, the maximum global humidity deviation, and the transformer room temperature rise rate.

[0051] The calculation formula for the maximum global temperature deviation is as follows:

[0052] ;

[0053] Where, is the maximum global temperature deviation, is the temperature deviation of the transformer room, is the temperature deviation of the high-voltage room, is the temperature deviation of the low-voltage room;

[0054] The calculation formula for the maximum global humidity deviation is as follows:

[0055] ;

[0056] Where, is the maximum global humidity deviation, is the humidity deviation of the transformer room, is the humidity deviation of the high-voltage room, is the humidity deviation of the low-voltage room;

[0057] ] The calculation formula for the transformer room temperature rise rate is as follows:

[0058] ;

[0059] Where, is the transformer room temperature rise rate, This represents the temperature rise of the transformer room per unit time. Unit of time.

[0060] As described above, the three fuzzy control prerequisite variables comprehensively reflect the temperature and humidity status of the substation from different dimensions. The maximum global temperature deviation and the maximum global humidity deviation focus on the most severe deviations from the set values ​​in each electrical room, ensuring that control commands can address key issues in a targeted manner and avoid affecting equipment operation due to local temperature and humidity exceeding limits. The transformer room temperature rise rate accurately captures the temperature change trend of the main heat-generating areas, enabling the controller to predict temperature change risks in advance and effectively suppress overheating problems caused by rapid temperature increases. The combination of the three fuzzy control prerequisite variables provides comprehensive, accurate, and targeted input information for the TS fuzzy model, ensuring the rationality and effectiveness of subsequent control law synthesis, and further improving the response speed and control accuracy of the entire control system to complex environmental changes.

[0061] Furthermore, in step S6, fault diagnosis of each fan based on the collected current and voltage values ​​includes: comparing the collected current value with a preset rated current value, and comparing the collected voltage value with a preset rated voltage value, and determining the fault type of the fan according to a preset four-category criterion; the four-category criterion includes:

[0062] like and If so, the diagnosis is an open circuit;

[0063] like and If the decrease is less than 5%, it is diagnosed as overload;

[0064] like And it rose rapidly, at the same time If the sudden drop exceeds 10%, it is diagnosed as a short circuit;

[0065] like and If the upstream power supply is normal, the diagnosis is a fan contactor fault.

[0066] in, This represents the current value of the fan. This is the voltage value of the fan. This is the rated current value. This is the rated voltage value.

[0067] As described above, the five-category fault diagnosis method based on current-voltage characteristics, compared to traditional fault detection methods that rely solely on thermal relays or air switches, can accurately distinguish between five different types of faults: open circuit, overload, short circuit, fan failure, and power supply loss. This provides clear and specific criteria for on-site fault diagnosis and maintenance, avoiding the waste of time and costs caused by blind repairs. The criteria for each fault type are based on the quantitative characteristics of current and voltage, with clear judgment logic and rapid response, enabling quick diagnosis after a fan failure occurs, ensuring timely disconnection of the faulty fan and preventing the fault from escalating. By accurately identifying the fault type, unnecessary shutdowns due to misjudgment can be avoided. At the same time, it provides accurate fault information support for the operation adjustment of remaining fans in subsequent fault-tolerant control, ensuring more targeted airflow distribution and control strategy adjustments, and guaranteeing the stability of temperature and humidity control under fault conditions.

[0068] Furthermore, in step S3, the membership values ​​of each fuzzy control premise variable on each fuzzy set in the TS fuzzy model are calculated using the triangular membership function.

[0069] As described above, the triangular membership function has the advantages of simple calculation process and low computational load, enabling it to quickly process fuzzy control premise variables, meet the real-time temperature and humidity control requirements of new energy substations, ensure that control commands can be generated and executed in a timely manner, and avoid control lag caused by calculation delays. The smooth transition of the fuzzy boundary of this membership function can effectively avoid the problem of abrupt changes in control commands caused by fuzzy set partitioning, keeping the fan operation status and temperature and humidity changes stable, reducing temperature and humidity oscillations, and further improving control accuracy. At the same time, the triangular membership function has good sensitivity to changes in input variables, can accurately capture subtle fluctuations in premise variables, and ensure that the TS fuzzy model can quickly respond to small changes in ambient temperature and humidity, adjust the control strategy in a timely manner, and keep the temperature and humidity stable within the set range, ensuring the stability of the equipment operating environment.

[0070] Furthermore, the calculation process of the membership function of the triangle includes:

[0071] The actual value of each of the fuzzy control premise variables is normalized to a preset standard universe of discourse;

[0072] Within the predefined standard universe of discourse, three triangular membership functions are defined for each fuzzy control premise variable, successively covering its negative value range, the range near zero, and the positive value range.

[0073] Substituting the normalized fuzzy control premise variables into their corresponding three triangular membership functions yields three membership values.

[0074] As described above, normalizing the premise variables of fuzzy control can effectively eliminate the influence of differences in the dimensions of different variables, ensuring that the maximum global temperature deviation, the maximum global humidity deviation, and the transformer room temperature rise rate have equal weight and influence in the TS fuzzy model. This avoids control deviations caused by differences in the numerical range of dependent variables and ensures the rationality of the control logic. Defining three triangular membership functions covering different intervals for each premise variable can comprehensively cover all possible states of the variable. Regardless of whether the premise variable is in a negative value, near zero, or in a positive range, a corresponding fuzzy set can be found and the membership value can be accurately calculated, greatly improving the model's coverage and adaptability to complex environments.

[0075] Furthermore, the normalized standard universe of discourse for the fuzzy control premise variables is [-1, 1];

[0076] The vertices of the three membership functions of the triangle are located near the negative boundary, zero, and positive boundary of the preset standard universe of discourse, respectively.

[0077] As can be seen from the above description, setting the normalized standard universe of discourse to [-1, 1] is a moderate range that can fully reflect the degree of deviation and trend of the premise variables, and facilitates the design and calculation of membership functions, thereby reducing the complexity of the algorithm and improving the computational efficiency. The vertices of the three triangular membership functions are located near the negative boundary, zero point and positive boundary of the standard universe of discourse, respectively, so that the membership functions can cover the entire universe of discourse evenly and symmetrically, ensuring the consistency and rationality of the fuzzy set partitioning, and avoiding control deviations caused by uneven fuzzy set partitioning.

[0078] Furthermore, step S4 specifically involves:

[0079] Based on the membership value calculated in step S3, the normalized activation degree of each fuzzy rule in multiple TS fuzzy models is calculated.

[0080] Based on the normalized activation degree, the local linear control laws corresponding to multiple fuzzy rules are weighted and averaged to synthesize a global control law; the output of the global control law is the duty cycle of each wind turbine.

[0081] As described above, the calculation of normalized activation can accurately reflect the matching degree between the current temperature and humidity state and each fuzzy rule, providing a scientific and reasonable basis for the weighted fusion of local linear control laws, ensuring that the global control law can adapt to the current environmental state, and avoiding the control limitations caused by single rule control; by synthesizing the global control law through weighted averaging, the control effects of multiple fuzzy rules can be organically combined, giving full play to the advantages of the TS fuzzy model in handling nonlinear and strongly coupled systems, realizing the coordinated control of multi-chamber fans, and effectively solving the problem of energy waste or inaccurate control caused by the uncoordinated operation of each fan in traditional distributed control;

[0082] The global control law output is the duty cycle of each fan. By adjusting the duty cycle, the fan speed can be continuously adjusted. Compared with traditional on / off control, it can control the air volume output more precisely, which not only improves the accuracy of temperature and humidity control, but also avoids mechanical wear caused by frequent fan start-stop, extends the fan life and reduces energy consumption.

[0083] Furthermore, in step S5, a tiered start-stop control strategy is implemented for fans in different locations within each electrical room, specifically including the following steps:

[0084] S51. Identify the type of the target electrical room based on the control command and the temperature and humidity values ​​of multiple electrical rooms collected in step S1;

[0085] S52. If the target electrical room is a transformer room, then execute the first layering strategy:

[0086] Based on the temperature deviation of the transformer room, the bottom crossflow fan of the transformer room is controlled by duty cycle modulation; if the ambient temperature exceeds the temperature setting value of a higher level, the wall axial flow fan and the top axial flow fan of the transformer room are started in sequence according to the preset priority.

[0087] S53. If the target electrical room is a high-voltage room or a low-voltage room, then execute the second stratification strategy:

[0088] When the temperature inside the cabinet exceeds the standard, the cabinet fan should be activated first.

[0089] When the indoor humidity exceeds the standard, the cabinet fan and heater will start simultaneously, and the wall fan will start after a preset delay.

[0090] As described above, differentiated layered start-up and shutdown strategies were designed to address the different structural characteristics and temperature and humidity control requirements of the transformer room, high-voltage room, and low-voltage room. This achieved precise control tailored to local conditions and avoided the unreasonable regulation problems caused by a uniform control strategy. The transformer room uses a duty cycle modulation method to control the bottom crossflow fan, effectively reducing the number of fan start-ups and shutdowns and extending fan life. The wall-mounted axial flow fans and top-mounted axial flow fans are started sequentially according to priority, ensuring that airflow is distributed as needed and avoiding unnecessary energy waste. In the high-voltage room and low-voltage room, the start-up priority of the cabinet fans is placed before the wall-mounted fans. This quickly solves the local heat dissipation problem of the equipment inside the cabinet, preventing excessively high cabinet temperatures from affecting equipment performance. The design of simultaneously starting the fans and heaters and delaying the start-up of the wall-mounted fans when humidity exceeds the standard effectively prevents heat and humidity conflicts, avoids the impact of condensation on the equipment insulation performance, and further improves the scientific and rational nature of the regulation.

[0091] Furthermore, in step S8, the TS fuzzy model is reconstructed online, specifically including:

[0092] Remove the matrix column corresponding to the faulty wind turbine from the input matrix corresponding to each fuzzy rule of the TS fuzzy model to obtain the input matrix after the fault.

[0093] Based on the obtained input matrix after the fault, the normalized activation degree and global control law of each fuzzy rule in the TS fuzzy model are recalculated.

[0094] As described above, the online reconstruction method of the TS fuzzy model is simple and efficient. It only requires removing the matrix column corresponding to the faulty fan from the input matrix, without changing the overall structure and other parameters of the model, thus ensuring the continuity of control and avoiding the control interruption problem during the model reconstruction process. The reconstructed input matrix can accurately adapt to the number and layout of the remaining normal fans. By recalculating the normalized activation degree and global control law, the control command can quickly adapt to the system state after the fault, ensuring the accuracy and stability of temperature and humidity control and avoiding local temperature and humidity exceeding the standard due to fan failure.

[0095] Please refer to Figure 7 The second technical solution adopted in this invention is as follows:

[0096] A temperature and humidity control system for a new energy substation, used to implement the above-mentioned temperature and humidity control method, the temperature and humidity control system comprising:

[0097] Multiple temperature and humidity sensors 1 are arranged in multiple electrical rooms to collect the temperature and humidity values ​​of each electrical room;

[0098] Multiple fans 2 are respectively arranged in multiple electrical rooms;

[0099] Multiple current transformers 3 and multiple voltage transformers 4 are respectively installed in the power supply circuit of each fan to collect the current and voltage values ​​of each fan.

[0100] Edge controller 5 is used to run the multivariable TS fuzzy control algorithm. It is electrically connected to multiple temperature and humidity sensors 1, multiple fans 2, multiple current transformers 3 and multiple voltage transformers 4 respectively, and executes the steps of the above method.

[0101] As can be seen from the above description, the beneficial effects of the present invention are as follows:

[0102] The multi-location temperature and humidity sensors can comprehensively and accurately collect temperature and humidity data from different areas of each electrical room, providing reliable and comprehensive basic data support for the control algorithm and avoiding control deviations caused by incomplete data collection; the independent current transformers and voltage transformers configured for each fan can collect fan operation data in real time and accurately, providing an accurate data source for fault diagnosis and ensuring the accuracy of fault type judgment.

[0103] Please refer to Figures 1 to 6 Embodiment 1 of the present invention is as follows:

[0104] Please refer to Figure 1 and Figure 6 A method for temperature and humidity control in a new energy substation includes the following steps:

[0105] S1. Collect temperature and humidity values ​​from multiple electrical rooms within the substation;

[0106] S2. Based on the temperature and humidity values ​​of multiple electrical rooms collected in step S1, calculate multiple fuzzy control prerequisite variables for the TS fuzzy model.

[0107] The multiple fuzzy control prerequisite variables in step S2 include the maximum global temperature deviation, the maximum global humidity deviation, and the transformer room temperature rise rate;

[0108] The formula for calculating the maximum global temperature deviation is as follows:

[0109] ;

[0110] in, This represents the maximum global temperature deviation. For transformer room temperature deviation, For the temperature deviation in the high-pressure chamber, Temperature deviation in the low-pressure chamber;

[0111] The formula for calculating the maximum global humidity deviation is as follows:

[0112] ;

[0113] in, This represents the maximum global humidity deviation. For humidity deviation in the transformer room, For humidity deviation in the high-pressure room, Humidity deviation in the low-pressure chamber;

[0114] The formula for calculating the transformer's room temperature rise rate is as follows:

[0115] ;

[0116] in, For the transformer room temperature rise rate, This represents the temperature rise of the transformer room per unit time. Unit of time.

[0117] S3. Based on the fuzzy control premise variables calculated in step S2, calculate the membership values ​​of each fuzzy control premise variable on each fuzzy set in the TS fuzzy model (TS stands for Takagi-Sugeno).

[0118] In step S3, the membership values ​​of each fuzzy control premise variable on each fuzzy set in the TS fuzzy model are calculated using the triangular membership function.

[0119] The calculation process of the membership function of the triangle includes:

[0120] The actual value of each of the fuzzy control premise variables is normalized to a preset standard universe of discourse;

[0121] Within the predefined standard universe of discourse, three triangular membership functions are defined for each fuzzy control premise variable, successively covering its negative value range, the range near zero, and the positive value range.

[0122] Substituting the normalized fuzzy control premise variables into their corresponding three triangular membership functions yields three membership values.

[0123] The normalized standard universe of discourse for the fuzzy control premise variables is [-1, 1];

[0124] The vertices of the three membership functions of the triangle are located near the negative boundary, zero, and positive boundary of the preset standard universe of discourse, respectively.

[0125] S4. Based on the membership values ​​calculated in step S3, synthesize the global control law through the TS fuzzy model to generate control commands for the coordinated control of the fans in each electrical room.

[0126] Step S4 is as follows:

[0127] Based on the membership value calculated in step S3, the normalized activation degree of each fuzzy rule in multiple TS fuzzy models is calculated.

[0128] Based on the normalized activation degree, the local linear control laws corresponding to multiple fuzzy rules are weighted and averaged to synthesize a global control law; the output of the global control law is the duty cycle of each wind turbine.

[0129] S5. Based on the control instructions generated in step S4, execute the layered start-stop control strategy for fans in different locations in each electrical room;

[0130] In step S5, a tiered start-stop control strategy is implemented for fans in different locations within each electrical room, specifically including the following steps:

[0131] S51. Identify the type of the target electrical room based on the control command and the temperature and humidity values ​​of multiple electrical rooms collected in step S1;

[0132] S52. If the target electrical room is a transformer room, then execute the first layering strategy:

[0133] Based on the temperature deviation of the transformer room, the bottom crossflow fan of the transformer room is controlled by duty cycle modulation; if the ambient temperature exceeds the temperature setting value of a higher level, the wall axial flow fan and the top axial flow fan of the transformer room are started in sequence according to the preset priority.

[0134] S53. If the target electrical room is a high-voltage room or a low-voltage room, then execute the second stratification strategy:

[0135] When the temperature inside the cabinet exceeds the standard, the cabinet fan should be activated first.

[0136] When the indoor humidity exceeds the standard, the cabinet fan and heater will start simultaneously, and the wall fan will start after a preset delay.

[0137] S6. Collect the current and voltage values ​​of the fans in each electrical room, and perform fault diagnosis on each fan based on the collected current and voltage values ​​to determine whether each fan has malfunctioned.

[0138] In step S6, fault diagnosis of each fan based on the collected current and voltage values ​​includes: comparing the collected current value with a preset rated current value, and comparing the collected voltage value with a preset rated voltage value, and determining the fault type of the fan according to a preset four-category criterion; the four-category criterion includes:

[0139] like and If so, the diagnosis is an open circuit;

[0140] like and If the decrease is less than 5%, it is diagnosed as overload;

[0141] like And it rose rapidly, at the same time If the sudden drop exceeds 10%, it is diagnosed as a short circuit;

[0142] like and If the upstream power supply is normal, the diagnosis is a fan contactor fault.

[0143] in, This represents the current value of the fan. This is the voltage value of the fan. This is the rated current value. This is the rated voltage value.

[0144] S7. If the fan does not malfunction, return to step S1 and continue to the next cycle; if the fan malfunctions, proceed to step S8.

[0145] S8. Disconnect the faulty fan and reconstruct the TS fuzzy model online;

[0146] In step S8, the TS fuzzy model is reconstructed online, specifically including:

[0147] Remove the matrix column corresponding to the faulty wind turbine from the input matrix corresponding to each fuzzy rule of the TS fuzzy model to obtain the input matrix after the fault.

[0148] Based on the obtained input matrix after the fault, the normalized activation degree and global control law of each fuzzy rule in the TS fuzzy model are recalculated.

[0149] S9. Based on the reconstructed TS fuzzy model, readjust the operating time of the remaining normal fans;

[0150] S10. Return to step S1 and continue executing the next control loop.

[0151] This embodiment provides a method for temperature and humidity control in a new energy substation, applicable to photovoltaic power station step-up substations that include a transformer room, a high-voltage room, and a low-voltage room.

[0152] Please refer to Figure 2 System hardware configuration:

[0153] 1. Deployment of temperature and humidity sensors:

[0154] High-pressure chamber: Deploy 3 sets of temperature and humidity sensors, installed inside the cabinet, in the center of the chamber, and on the wall respectively;

[0155] Low-pressure chamber: Deploy 3 sets of temperature and humidity sensors, installed in the cabinet, in the center of the chamber, and on the wall respectively;

[0156] Transformer room: One set of temperature and humidity sensors is installed on the wall and in the center. Three sets of PT100 resistance temperature sensors are installed on the transformer windings to accurately collect the temperature of the transformer windings.

[0157] 2. Wind turbine deployment:

[0158] High-pressure chamber: One cooling fan is installed inside the cabinet, and one axial flow fan is installed on the wall;

[0159] Low-pressure compartment: One cooling fan is installed inside the cabinet, and one axial flow fan is installed on the wall.

[0160] Transformer room: Six cross-flow cooling fans (independently controlled by contactors) are installed at the bottom, and one axial flow fan is installed on the top and one on the wall.

[0161] 3. Current transformers: Each wind turbine's power supply circuit is equipped with current transformers and voltage transformers to collect the wind turbine's current value in real time. and voltage value .

[0162] 4. Edge Controller: Adopts a dual-core ARM+DSP architecture, supports IEC-61850 and Modbus-TCP communication interfaces, and is electrically connected to various sensors, fans, and transformers for running control algorithms and processing data.

[0163] I. The relevant design of the TS fuzzy model is as follows (please refer to...) Figure 3 ):

[0164] 1. Selection of precondition variables for fuzzy control:

[0165] Maximum global temperature deviation : ;

[0166] Maximum global humidity deviation : ;

[0167] Transformer room temperature rise rate : .

[0168] 2. Determination of fuzzy membership function:

[0169] For each fuzzy control premise variable, three fuzzy sets are defined: {N (negative), Z (zero), P (positive)}, and the triangular membership function is used to normalize to the standard universe of discourse [-1, 1].

[0170] Maximum global temperature deviation The basic universe of discourse is [-10℃, 10℃], which is normalized to the standard universe of discourse [-1, 1].

[0171] Maximum global humidity deviation The basic universe of discourse is [-20% RH, 20% RH], which is normalized to the standard universe of discourse [-1, 1].

[0172] Transformer room temperature rise rate The basic universe of discourse is [-2℃ / s, 2℃ / s], which is normalized to the standard universe of discourse [-1, 1].

[0173] Global temperature deviation maximum value Is ±10℃, the basic universe of discourse is {-10, 10}, the fuzzy subset universe of discourse is {-10, 0, 10}, and the corresponding fuzzy subsets are {N (negative), Z (zero), P (positive)}.

[0174] Global humidity deviation maximum value Is ±20RH, the basic universe of discourse is {-20, 20}, the fuzzy subset universe of discourse is {-20, 0, 20}, and the corresponding fuzzy subsets are {N (negative), Z (zero), P (positive)}.

[0175] Transformer room temperature rise rate Is ±2℃ / 1s, the basic universe of discourse is {-2, 2}, the fuzzy subset universe of discourse is {-2, 0, 2}, and the corresponding fuzzy subsets are {N (negative), Z (zero), P (positive)}.

[0176] After regional normalization, it corresponds to a triangular membership function. For example, the global temperature deviation maximum value The triangular vertex height of the triangular membership function of the fuzzy subset Z (zero) is equal to 1 (membership is 100%), corresponding to the global temperature deviation maximum value being equal to zero. When the deviation is -10℃ < T < 0℃, the membership corresponds to the hypotenuse of the left side of the triangle. When the deviation is -10℃, the membership is 0 (0%). When the deviation is positive, the situation is similar. For temperature deviations of -10℃ < T < 10℃, the corresponding membership can be calculated.

[0177] Extract the temperature and humidity values sampled at each moment and control them according to the maximum value of the global temperature deviation, the maximum value of the global humidity deviation, and the temperature rise rate of the transformer room (if the maximum value is within the control target range, the other temperature and humidity values are also within the control target range). Use the triangular membership function for easy calculation. Normalize these three input variables to the range of [-1, 1], and the corresponding membership degree is 0 - 100%. For example, define the Z (zero) fuzzy set: when the measured maximum temperature deviation is a positive deviation of +10°C or a negative deviation of -10°C, the membership degree is 0 (i.e., 0%). When the temperature deviation is 0°C, the membership degree is 1 (i.e., 100%). When the maximum temperature deviation is 0°C < T < 10°C (or -10°C < T < 0°C), calculate the membership degree according to the two hypotenuses of the triangle. Each temperature and humidity value sampled in real time corresponds to a membership degree value in a corresponding fuzzy set, which is used to predict the next temperature and humidity state value , the output value of the global temperature and humidity state at the current moment , real-time weight etc. (see the following three formulas). And calculate the duty cycle control amount of the fan based on this .

[0178] is the prediction of the control system for the future, which enables the controller to "foresee" the future consequences of the current actions and thus make more intelligent and accurate decisions. This is one of the key technologies for this invention to overcome the shortcomings of traditional methods such as "poor adaptability and inaccurate regulation", achieving the leap from "passive reaction" to "active prediction and optimization".

[0179] 3. T-S fuzzy rules (a total of 3×3×3 = 27 examples):

[0180] Rule 1: If is N (negative), and is N (negative), and is N (negative), then:

[0181] ;

[0182] ;

[0183] Rule 2: If is N (negative), and is N (negative), and is P (positive), then:

[0184] ;

[0185] ;

[0186] Similarly, until Rule 27: If It is N (negative), and It is N (negative), and If Z (zero), then:

[0187] ;

[0188] ;

[0189] in, It is the transpose of the matrix and belongs to a 6×1 column state vector;

[0190] This is a duty cycle speed adjustment command for each fan.

[0191] These 27 fuzzy rules represent all possible permutations and combinations of the fuzzy control premise variables. For any temperature and humidity measurement value within the aforementioned target control range, the specific membership values ​​of its three fuzzy control premise variables can be calculated, and it can be determined which of these 27 rules it belongs to. After determining the specific rule, the state-space equation can be used to predict the next temperature and humidity state value. and the current global temperature and humidity status output value. This enables output control.

[0192] 4. Global output of the fuzzy system:

[0193] The local linear model described by the 27 rules mentioned above is then analyzed using fuzzy weights. Weighted averaging yields the final global state-space equations for control and prediction, which can be updated at any time and used to predict the next temperature and humidity state values. (Predicted temperature and humidity values ​​for 6 rooms), calculate the global state feedback control law. (i.e., the control parameters of each fan speed / duty cycle), and online reconfiguration after a fan failure. , This enables airflow redistribution. It can also calculate the current global temperature and humidity output values. (generally (i.e., measured temperature and humidity), and real-time weights. (Used to combine 27 local matrices) , , (Synthesize a global matrix).

[0194] ;

[0195] ;

[0196] ;

[0197] ;

[0198] The physical meaning of each symbol in the above formula is shown in Table 1:

[0199] Table 1. Physical Meaning of Symbols

[0200]

[0201] In the formula, , , , These four matrices together form the core of a typical state-space model, which is used to describe the dynamic behavior of a control system. Under the TS fuzzy control framework, a complex nonlinear temperature and humidity system of a new energy substation is represented as a fuzzy fusion of multiple linear models (each rule corresponds to a linear model).

[0202] Each linear model is determined by its own... , , , Defined as a matrix group, it is a mathematical model parameter used in the design of TS fuzzy controllers to describe the dynamic characteristics of the subsystem.

[0203] State matrix (i.e.) A matrix describes the dynamic characteristics of the evolution of the system's internal state variables over time.

[0204] Input matrix (i.e.) A matrix defines how external control inputs affect changes in the internal state of a system; m represents the dimension of the input variable.

[0205] Affine term (or offset term) In the system model, represents a constant offset used to describe the static characteristics of the system near its equilibrium or linearization point.

[0206] Output matrix ( A matrix defines how the internal states of a system are mapped to measurable outputs; it is typically an identity matrix, meaning that all state variables can be directly observed as outputs.

[0207] Based on six temperature and humidity deviation variables in the three rooms , , , , , The definition corresponds to the following 6 state variables: , , , , , ,in , , Indicates the air humidity content of the three chambers, and , , They can be converted to each other using formulas. Additionally, the system uses two external control variables for regulation: the airflow driven by the i-th fan. Related to fan speed / duty cycle control quantity Correspondingly; the heating power of the j-th heater .

[0208] Taking the transformer room as an example, its temperature change rate The nonlinear equation for establishing the heat balance of the transformer room is as follows:

[0209] ;

[0210] Taking the transformer room as an example again, its moisture content change rate The nonlinear equation for the wet balance of the transformer room is established as follows:

[0211] ;

[0212] in: and These represent outdoor temperature and outdoor humidity, respectively.

[0213] Heat and humidity balance equations were established for the three rooms (transformer room, high-voltage room, and low-voltage room), resulting in a total of six first-order nonlinear differential equations. Arranging these equations into matrix form yields a rudimentary nonlinear continuous-time state-space equation:

[0214] ;

[0215] in:

[0216] It is the derivative of the state vector;

[0217] It is a non-linear vector function;

[0218] It is the control input vector (fan, heater).

[0219] This equation clearly demonstrates that the dynamic changes in the system state (how temperature and humidity change) depend on the current state itself, control commands, and external environmental disturbances, and that there are complex nonlinear coupling relationships among them.

[0220] To a complex nonlinear system Near a specific operating point (equilibrium point), it approximates a simple linear system. For each TS fuzzy rule L, a representative equilibrium point is selected, and the nonlinear function is applied near this equilibrium point. Performing a first-order Taylor expansion, ignoring higher-order terms in the nonlinear function, and retaining only the first-order term about the deviation from the operating point, we obtain a locally linear model: Thus, the matrix is ​​calculated. , , , This linear model can describe the system dynamics accurately enough within a small range around the operating point, and can be used for analysis and control design using linear system theory.

[0221] This is precisely why linearization is subsequently performed to solve the matrix in the TS fuzzy model. , , , The foundation of the entire TS fuzzy control framework lies in the locally linearized model obtained by linearizing the aforementioned nonlinear model at different operating points (corresponding to different fuzzy rules). , , , Weighted fusion.

[0222] 5. Fuzzy state feedback control:

[0223] For each local linear model corresponding to the aforementioned 27 TS fuzzy rules, an optimal control law is designed, employing the error-free control characteristics of a PI controller to achieve smooth control of ambient temperature and humidity. These local control laws are then weighted and fused according to the real-time state of the system to ultimately generate a global, stable control command that adapts to the nonlinear characteristics of the system, namely, the duty cycle or speed command for each fan. Its design process can be summarized as "decompose first, then design, then integrate, and finally enhance".

[0224] a. Decomposition: Design an LQR controller for each local subsystem;

[0225] Since the temperature and humidity system of a new energy substation is a complex nonlinear system, we have linearized it at 27 different operating points (corresponding to 27 rules) using the TS fuzzy model, resulting in 27 locally linear models. The dynamic characteristics of each model are determined by the state equations. describe.

[0226] To ensure that each local subsystem is not only stable but also has good dynamic performance (such as fast response and low overshoot), this scheme designs a linear quadratic regulator (referred to as LQR) for each rule.

[0227] The goal of LQR is to find the optimal control gain. This results in the following performance indicators Minimize:

[0228] ;

[0229] in, It is a state vector (i.e., the temperature and humidity of the three rooms);

[0230] It is the control input vector (i.e., the command for each wind turbine);

[0231] It is the state weight matrix (positive definite matrix), which determines the system's sensitivity to temperature and humidity deviations; increasing it... This means the system will work harder to reduce deviations and improve control precision, but this may lead to overly aggressive control actions and increased energy consumption.

[0232] The control weight matrix (positive definite matrix) determines the system's emphasis on the control variable (fan energy consumption); increasing... This means the system will be more inclined to use smaller control actions to save energy, but may sacrifice some adjustment speed.

[0233] By solving the famous algebraic Riccati equation:

[0234] ;

[0235] Obtain the positive definite matrix Afterwards, local LQR gain It is given by the following formula:

[0236] ;

[0237] Substituting into the state equations, we obtain the closed-loop system:

[0238] ;

[0239] Through reasonable selection and It can guarantee The eigenvalues ​​(i.e., system poles) are located within the stable region, thus ensuring the stability and excellent dynamic performance of the local subsystem.

[0240] b. Fusion: Weighted summation yields global fuzzy control commands;

[0241] At any given moment, the system's operating state may simultaneously activate multiple TS fuzzy rules (i.e., belong to multiple fuzzy sets). Therefore, a control law using only one rule cannot be simply applied.

[0242] Final global control command It is the weighted average of 27 local LQR control commands:

[0243] ;

[0244] It is the normalized activation of the Lth rule (calculated in the previous section), which reflects the current system state. The degree of matching with the Lth rule;

[0245] It is the control quantity recommended by rule L;

[0246] This weighted fusion process enables the controller to smoothly transition between different operating points and adaptively handle the nonlinear characteristics of the system.

[0247] c. Enhancement: Introduce an integral element to eliminate steady-state error (forming a TS-Fuzzy-PI controller).

[0248] Although the aforementioned fuzzy state feedback controller performs well, it may exhibit steady-state error (static error) in processes requiring high precision, such as temperature and humidity control. This means that after long-term operation, there is a slight deviation between the actual value and the setpoint. To completely eliminate steady-state error, this solution adds an integral term to the fuzzy state feedback, forming a more advanced TS-Fuzzy-PI controller.

[0249] Step 1, Design of the integral term:

[0250] Define the integral state , For tracking error;

[0251] in For setting value, This is the actual output value.

[0252] Define the integral state It is the cumulative sum of all historical errors: ;

[0253] Introduce it into the state vector or design the integral gain separately. This expands the control law to:

[0254] ;

[0255] Step 2: Integration of PI control logic:

[0256] At this time, the controller also has a "proportional term". "and integral terms" This approach retains the adaptability of fuzzy state feedback to nonlinear systems while forcibly eliminating steady-state error through the integral element, ultimately achieving a combined control effect of "TS fuzzy and PI control".

[0257] This new control law includes:

[0258] Proportional Term Provides fast, proportional control actions that are adaptively adjusted by TS fuzzy logic.

[0259] Integral term By accumulating historical deviations, a continuous control effect is generated until the error is eliminated.

[0260] This composite structure of "TS fuzzy and PI regulation" retains the powerful ability of fuzzy control to handle nonlinearity, while introducing the zero steady-state error tracking characteristic of the integral element, ultimately achieving high-precision temperature and humidity control.

[0261] After the above series of steps, the final calculated control command is obtained. It is a vector, where each element directly corresponds to the duty cycle (if it is switch control) or speed command (if it is frequency converter control) of a wind turbine. This command is sent to the frequency converter or contactor corresponding to each wind turbine, thereby realizing intelligent, precise, and fault-tolerant control of temperature and humidity within the substation.

[0262] II. The tiered start-stop control strategy is as follows (please refer to...) Figure 4 ):

[0263] To achieve high-precision, zero-steady-state-error control of various room temperature and humidity levels, a PI controller is added. The preceding steps determine the control parameters of the PI controller; and using these methods, the fan duty cycle is calculated. This enables the control of the fan.

[0264] This step simply utilizes the aforementioned output to control the fan. In terms of control strategy, the duty cycle can be set to a longer period, such as 30 seconds, to avoid frequent starts; the temperature and humidity sensor values ​​are compared with target values ​​to obtain the start-stop temperature difference of the bottom crossflow fan; to avoid "thermal and humidity conflict," the transformer room uses a phased activation method of wall-mounted / top-mounted axial flow fans; the high and low voltage rooms use a phased activation method of cabinet / wall-mounted fans.

[0265] Transformer room:

[0266] Bottom crossflow fan start / stop temperature difference (Target value defaults to 8 K): Duty cycle modulation is used to avoid frequent start-stop; wall / ceiling axial flow fans are designed according to "ambient temperature > "Invest in the project in stages, prioritizing the wall area and then the top."

[0267] High-pressure / low-pressure chamber:

[0268] The cabinet fan is directly started by the cabinet's internal temperature sensor PT100 (i.e., a thermistor), and has higher priority than the wall-mounted fan; humidity > At the same time, the cabinet fan and heater are started simultaneously, while the wall fan starts 5 minutes later to prevent "thermal and humidity conflict".

[0269] Fault diagnosis and isolation of faulty fans are detailed below:

[0270] a. Four-category criteria, as shown in Table 2:

[0271] Table 2. Four-category classification criteria table

[0272]

[0273] Current and voltage values ​​(Ij and Uj) are sampled in real time using current transformers (CT) and voltage transformers (PT); the rated current value (In) and rated voltage value (Un) can be preset in the system.

[0274] The criteria for this table are based on the characteristic combination of current and voltage, used to distinguish different types of fan failures.

[0275] a1. Internal circuit break in the fan:

[0276] When the control contactor is turned on, the detected fan current is ≈0, indicating that there is no current in the circuit; and the detected fan voltage is equal to the rated value (e.g., 380VAC), indicating that the power supply is normal. However, the fan operating current is zero, indicating that there is a disconnection somewhere inside the fan.

[0277] a2. Fan overload:

[0278] When the control contactor is activated, if the detected fan current is greater than 1.5 times the rated current In, it indicates that the current exceeds the normal operating range, but has not yet reached the level of a short circuit. This could be due to conditions such as bearing seizure or impeller jamming. A slight drop in fan voltage is also detected; this is due to current overload, causing a slight decrease in voltage, but not as drastic as a short circuit. Overload can be detected by comparing the actual current I_j with the rated current curve in real time.

[0279] a3. Internal short circuit in the fan winding:

[0280] When the control contactor is turned on, if the detected fan current is greater than twice the rated current In and the fan current rises sharply, it is because the current will rise sharply when the fan circuit is short-circuited, exceeding twice the normal operating current. At the same time, it will cause the fan voltage to drop sharply by more than 10%, because a short circuit will cause the voltage to drop sharply because the resistance of the short circuit path is very low. In this case, the contactor will be immediately disconnected to cut off the circuit and protect the equipment.

[0281] a4. Contactor not engaged or power supply phase missing:

[0282] If the fan current is detected to be approximately 0, the fan voltage is detected to be 0, and the upstream power supply is detected to be in normal condition;

[0283] This indicates that the contactor has issued a connection command, but the main contacts of the contactor have not actually closed; the fan itself is not faulty, but the contactor is malfunctioning, causing the fan to fail to start.

[0284] Summarize:

[0285] This table quickly and accurately identifies the type of fan failure by monitoring the current and voltage of each fan and combining specific characteristic combinations. This monitoring method can detect problems in a timely manner and prevent faulty fans from affecting the normal operation of other equipment.

[0286] b. Disconnect the faulty fan:

[0287] Upon detecting a fault, the edge controller immediately disconnects the dedicated contactor for that fan, while the power supply to the other fans in the same room remains unaffected.

[0288] Update TS fuzzy rules: From Deleting column j yields the input matrix for rule L after wind turbine failure. Recalculate real-time weights With global state feedback control law This enables online weight reconstruction.

[0289] Next step temperature and humidity status values The current global temperature and humidity status output value. Matrix is ​​required in calculations such as these. , etc. Among them. This is the natural drift matrix, which is related to the sensor input. This is the wind turbine gain matrix, related to the wind turbine control output. Each column corresponds to one wind turbine and is used to define the operating duty cycle of each turbine. When a wind turbine fails and is disconnected, in... The corresponding columns of the matrix are deleted, resulting in Then, the global control rules are redefined, and the output is reconstructed online to ensure the original control requirements are met.

[0290] These are the real-time weights in the global output of the fuzzy system, used to synthesize the 27 local matrices into a global matrix output.

[0291] c. Remaining fan running time auto-tuning (please refer to...) Figure 5 ):

[0292] The original required air volume of the faulty fan The operating time increment is calculated based on the "air volume-speed-power" curve and weighted accordingly. = (Available air volume i) / (Total remaining air volume) is allocated to the normal fans in the same room, which is achieved through frequency conversion or duty cycle.

[0293] The adoption of the above technical solution brings the following advantages:

[0294] A. Compared with existing distributed PID controllers, the settling time is reduced by more than 40%, and the overshoot is <0.5°C / 2%RH;

[0295] B. The fan failure is resolved and reconfigured within 1 second, with indoor temperature overshoot <1°C and humidity <3%RH;

[0296] C. The number of start-stop cycles of the crossflow fan at the bottom of the transformer room has been reduced from the traditional 120 times / day to <30 times / day, extending its service life by 2.5 times;

[0297] D. Annual fan energy consumption is reduced by 20-30%, fault location accuracy is >98%, avoiding the risk of high temperature tripping of the protection system caused by "one fault causing a complete shutdown".

[0298] Please refer to Figure 2 and Figure 3 Embodiment two of the present invention is as follows:

[0299] Please refer to Figure 2 A temperature and humidity control system for a new energy substation, used to implement the above-mentioned temperature and humidity control method, the temperature and humidity control system comprising:

[0300] Multiple temperature and humidity sensors 1 are arranged in multiple electrical rooms to collect the temperature and humidity values ​​of each electrical room;

[0301] Multiple fans 2 are respectively arranged in multiple electrical rooms;

[0302] Multiple current transformers 3 and multiple voltage transformers 4 are respectively installed in the power supply circuit of each fan to collect the current and voltage values ​​of each fan.

[0303] Edge controller 5 is used to run the multivariable TS fuzzy control algorithm. It is electrically connected to multiple temperature and humidity sensors 1, multiple fans 2, multiple current transformers 3 and multiple voltage transformers 4 respectively, and executes the steps of the above method.

[0304] Please refer to Figure 3 Temperature and humidity sensor deployment:

[0305] High-pressure chamber: Deploy 3 sets of temperature and humidity sensors, installed inside the cabinet, in the center of the chamber, and on the wall respectively;

[0306] Low-pressure chamber: Deploy 3 sets of temperature and humidity sensors, installed in the cabinet, in the center of the chamber, and on the wall respectively;

[0307] Transformer room: One set of temperature and humidity sensors is installed on the wall and in the center. Three sets of PT100 resistance temperature sensors are installed on the transformer windings to accurately collect the temperature of the transformer windings.

[0308] Wind turbine deployment:

[0309] High-pressure chamber: One cooling fan is installed inside the cabinet, and one axial flow fan is installed on the wall;

[0310] Low-pressure compartment: One cooling fan is installed inside the cabinet, and one axial flow fan is installed on the wall.

[0311] Transformer room: Six cross-flow cooling fans (independently controlled by contactors) are installed at the bottom, and one axial flow fan is installed on the top and one on the wall.

[0312] Current transformers: Each wind turbine's power supply circuit is equipped with current transformers and voltage transformers to collect the wind turbine's current value in real time. and voltage value .

[0313] Edge controller: Adopts a dual-core ARM+DSP architecture, supports IEC-61850 and Modbus-TCP communication interfaces, and is electrically connected to various sensors, fans and transformers for running control algorithms and processing data.

[0314] In summary, the temperature and humidity control method and system for new energy substations provided by this invention achieves deep integration of temperature and humidity control with fault tolerance through a complete control process including data acquisition, fuzzy control prerequisite variable calculation, membership value calculation, control law synthesis, hierarchical start-stop control strategy, fan fault diagnosis, fan fault handling, and closed-loop cycle. This allows for precise adaptation to the strong coupling characteristics of the multi-chamber environment in new energy substations. Furthermore, the global control law synthesized through the TS fuzzy model enables coordinated operation of fans in multiple chambers, significantly improving the accuracy and consistency of temperature and humidity control and effectively avoiding local overheating or condensation problems. The start-stop control strategy can differentiate the operation of fans based on the structural characteristics and temperature and humidity requirements of different electrical rooms, reducing frequent start-stop cycles, thus lowering energy consumption and extending fan lifespan. The fault diagnosis and fault-tolerant control system can quickly identify fan faults and shut down only the faulty fan. Through model reconstruction and adjustment of remaining fan operating time, it avoids the system paralysis risk caused by a single fault-induced shutdown in traditional control methods, ensuring the continuity and stability of temperature and humidity control. This comprehensively improves the system's adaptability and operational reliability, providing strong support for the safe and stable operation of key equipment in new energy substations. Multiple temperature and humidity sensors can comprehensively and accurately collect temperature and humidity data from different areas of each electrical room, providing reliable and comprehensive basic data support for the control algorithm and avoiding control deviations caused by incomplete data collection. Independently configured current and voltage transformers for each fan can collect fan operating data in real time and accurately, providing an accurate data source for fault diagnosis and ensuring the accuracy of fault type identification.

[0315] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for temperature and humidity control in a new energy substation, characterized in that, Includes the following steps: S1. Collect temperature and humidity values ​​from multiple electrical rooms within the substation; S2. Based on the temperature and humidity values ​​of multiple electrical rooms collected in step S1, calculate multiple fuzzy control prerequisite variables for the TS fuzzy model. The multiple fuzzy control prerequisite variables in step S2 include the maximum global temperature deviation, the maximum global humidity deviation, and the transformer room temperature rise rate; S3. Based on the fuzzy control premise variables calculated in step S2, calculate the membership values ​​of each fuzzy control premise variable on each fuzzy set in the TS fuzzy model. S4. Based on the membership values ​​calculated in step S3, synthesize the global control law through the TS fuzzy model to generate control commands for the coordinated control of the fans in each electrical room. Step S4 is as follows: Based on the membership value calculated in step S3, the normalized activation degree of each fuzzy rule in multiple TS fuzzy models is calculated. Based on the normalized activation degree, the local linear control laws corresponding to multiple fuzzy rules are weighted and averaged to synthesize a global control law. The output of the global control law is the duty cycle of each wind turbine; S5. Based on the control instructions generated in step S4, execute the layered start-stop control strategy for fans in different locations in each electrical room; In step S5, a tiered start-stop control strategy is implemented for fans in different locations within each electrical room, specifically including the following steps: S51. Identify the type of the target electrical room based on the control command and the temperature and humidity values ​​of multiple electrical rooms collected in step S1; S52. If the target electrical room is a transformer room, then execute the first layering strategy: Based on the temperature deviation of the transformer room, the bottom crossflow fan of the transformer room is controlled by duty cycle modulation; if the ambient temperature exceeds the temperature setting value of a higher level, the wall axial flow fan and the top axial flow fan of the transformer room are started in sequence according to the preset priority. S53. If the target electrical room is a high-voltage room or a low-voltage room, then execute the second stratification strategy: When the temperature inside the cabinet exceeds the standard, the cabinet fan should be activated first. When the indoor humidity exceeds the standard, the cabinet fan and heater will start simultaneously, and the wall fan will start after a preset delay. S6. Collect the current and voltage values ​​of the fans in each electrical room, and perform fault diagnosis on each fan based on the collected current and voltage values ​​to determine whether each fan has malfunctioned. S7. If the fan does not malfunction, return to step S1 and continue to the next cycle; if the fan malfunctions, proceed to step S8. S8. Disconnect the faulty fan and reconstruct the TS fuzzy model online; S9. Based on the reconstructed TS fuzzy model, readjust the operating time of the remaining normal fans; S10. Return to step S1 and continue executing the next control loop.

2. The temperature and humidity control method for a new energy substation according to claim 1, characterized in that, The formula for calculating the maximum global temperature deviation is as follows: ； in, This represents the maximum global temperature deviation. For transformer room temperature deviation, For the temperature deviation in the high-pressure chamber, Temperature deviation in the low-pressure chamber; The formula for calculating the maximum global humidity deviation is as follows: ； in, This represents the maximum global humidity deviation. For humidity deviation in the transformer room, For humidity deviation in the high-pressure room, Humidity deviation in the low-pressure chamber; The formula for calculating the transformer's room temperature rise rate is as follows: ； in, For the transformer room temperature rise rate, This represents the temperature rise of the transformer room per unit time. Unit of time.

3. The temperature and humidity control method for a new energy substation according to claim 1, characterized in that, In step S6, fault diagnosis of each fan based on the collected current and voltage values ​​includes: comparing the collected current value with a preset rated current value, and comparing the collected voltage value with a preset rated voltage value, and determining the fault type of the fan according to a preset four-category criterion; the four-category criterion includes: like and If so, the diagnosis is an open circuit; like and If the decrease is less than 5%, it is diagnosed as overload; like And it rose rapidly, at the same time If the sudden drop exceeds 10%, it is diagnosed as a short circuit; like and If the upstream power supply is normal, the diagnosis is a fan contactor fault. in, This represents the current value of the fan. This is the voltage value of the fan. This is the rated current value. This is the rated voltage value.

4. The temperature and humidity control method for a new energy substation according to claim 1, characterized in that, In step S3, the membership values ​​of each fuzzy control premise variable on each fuzzy set in the TS fuzzy model are calculated using the triangular membership function.

5. The temperature and humidity control method for a new energy substation according to claim 4, characterized in that, The calculation process of the membership function of the triangle includes: The actual value of each of the fuzzy control premise variables is normalized to a preset standard universe of discourse. Within the predefined standard universe of discourse, three triangular membership functions are defined for each fuzzy control premise variable, successively covering its negative value range, the range near zero, and the positive value range. Substituting the normalized fuzzy control premise variables into their corresponding three triangular membership functions yields three membership values.

6. The temperature and humidity control method for a new energy substation according to claim 5, characterized in that, The normalized standard universe of discourse for the fuzzy control premise variables is [-1, 1]; The vertices of the three membership functions of the triangle are located near the negative boundary, zero, and positive boundary of the preset standard universe of discourse, respectively.

7. The temperature and humidity control method for a new energy substation according to claim 1, characterized in that, In step S8, the TS fuzzy model is reconstructed online, specifically including: Remove the matrix column corresponding to the faulty wind turbine from the input matrix corresponding to each fuzzy rule of the TS fuzzy model to obtain the input matrix after the fault. Based on the obtained input matrix after the fault, the normalized activation degree and global control law of each fuzzy rule in the TS fuzzy model are recalculated.

8. A temperature and humidity control system for a new energy substation, used to implement the temperature and humidity control method according to any one of claims 1-7, characterized in that, The temperature and humidity control system includes: Multiple temperature and humidity sensors are arranged in multiple electrical rooms to collect temperature and humidity values ​​in each electrical room; Multiple fans are located in multiple electrical rooms; Multiple current transformers and multiple voltage transformers are installed in the power supply circuit of each wind turbine to collect the current and voltage values ​​of each wind turbine. An edge controller, used to run a multivariable TS fuzzy control algorithm, is electrically connected to multiple temperature and humidity sensors, multiple fans, multiple current transformers and multiple voltage transformers, and performs the steps of the method described in any one of claims 1-7.

Images