A method for controlling sudden current changes in a frequency converter for a compressed air energy storage system

Abstract

The present invention provides a method for controlling the current mutation of the frequency converter of the compressed air energy storage system, which belongs to the field of mutation control technology. It collects the output current data of the frequency converter of the compressed air energy storage system in real time to obtain the current change rate; compares the current change rate with a preset dynamic threshold to determine whether there is a current mutation trend; if it is determined that there is a mutation trend, a closed-loop control mode is triggered; and at the same time, a feedforward compensation mechanism is introduced to adjust the control parameters in advance according to the load change trend to improve the system response speed and control accuracy. Therefore, by dynamically detecting the current change trend of the frequency converter of the compressed air energy storage system and combining closed-loop and feedforward control strategies, it is possible to quickly respond before or in the early stage of the current mutation of the frequency converter of the compressed air energy storage system, effectively suppress the current mutation of the frequency converter of the compressed air energy storage system, improve the control accuracy and system stability, and is suitable for compressed air energy storage systems under complex working conditions.

Description

Technical Field

[0001] The present invention belongs to the technical field of sudden change control, and in particular relates to a method for controlling sudden change in current of a frequency converter used in a compressed air energy storage system. Background Art

[0002] The compressed air energy storage system converts electrical energy into compressed air potential energy through a compressor and stores it. When releasing energy, the compressed air is mixed with a heat storage medium through an expander to generate work and generate electricity. As mentioned in the prior art solution in patent publication number "CN119995180A", the compressed air energy storage system has a component such as a frequency converter.

[0003] In practical applications, in compressed air energy storage systems, the inverter is a key device for driving the compressor or turbine, and the stability of its output current directly affects the operating efficiency and equipment life of the compressed air energy storage system. However, during the operation of the compressed air energy storage system, factors such as sudden load changes, control delays or external disturbances can easily cause sudden changes in the inverter output current, leading to overload of power devices, malfunction of compressed air energy storage system protection, and even equipment damage. In the existing technology, methods such as fixed threshold comparison or hysteresis control are usually used to control current mutations, but these methods have problems such as slow response speed, low control accuracy, and inability to adapt to complex working conditions. It is difficult to meet the requirements of compressed air energy storage systems for high-precision and high-dynamic response control. Summary of the Invention

[0004] In order to address the defects in the existing technology, the present invention provides a method for controlling current mutations of an inverter for a compressed air energy storage system, which can realize rapid identification and high-precision control of current mutations of the compressed air energy storage system, improve the control efficiency and stability of the compressed air energy storage system, and avoid damage to the equipment of the compressed air energy storage system caused by current mutations.

[0005] The present invention utilizes the following technical solutions.

[0006] A method for controlling sudden current changes in a frequency converter of a compressed air energy storage system, comprising:

[0007] Step 1: Real-time acquisition and pre-processing of the inverter current data of the compressed air energy storage system;

[0008] Step 2: Perform dynamic slope detection and mutation trend identification;

[0009] Step 3: By optimizing the voltage adjustment , minimize the future The closed-loop control response is performed by taking the weighted sum of the “square of current deviation” and the “square of current change rate” within each prediction step.

[0010] Furthermore, in step 1, the PLC controller collects the three-phase current data output by the inverter of the compressed air energy storage system in real time through the current sensor connected to it. The three-phase current data collected by the PLC controller is denoised by a bandpass filter. Subsequently, the denoised three-phase current data is differentiated using a sliding window difference algorithm to obtain a time series of the current change rate.

[0011] Furthermore, in step 1, a method for performing differential processing on the three-phase current data after denoising using a sliding window difference algorithm to obtain a time series of current change rate includes:

[0012] Set up the first The three-phase current data after denoising at each acquisition moment are , , , then the first phase current data resultant vector magnitude for:

[0013] ;

[0014] Then, the PLC controller calculates the Current change rate :

[0015] ;

[0016] in , , For the The resultant vector magnitude, For the The resultant vector magnitude.

[0017] Furthermore, in step 2, based on the sliding window mechanism, a dynamic threshold comparison is performed on the current change rate. When it is detected that the current change rate exceeds the dynamic threshold for multiple consecutive sampling cycles, it is determined that there is a current mutation trend, and a mutation warning signal is sent to the LCD screen connected to the PLC controller for display.

[0018] Furthermore, in step 2, Dynamic threshold The calculation formula is:

[0019] ;

[0020] in , , The first The motor speed data is collected by the speed sensor connected to the PLC controller and transmitted to the PLC controller;

[0021] When detected and When the state exceeds three sampling windows continuously, the PLC controller determines that there is a current mutation trend and sends a mutation warning signal to the LCD screen connected to the PLC controller for display.

[0022] Furthermore, in step 3, while sending a sudden change warning signal to a liquid crystal screen connected to the PLC controller for display, the PLC controller starts a closed-loop control mode and adjusts the output frequency and voltage control parameters of the inverter.

[0023] Furthermore, in step 3, the closed-loop control mode adopts the model predictive control strategy, and its control objective is:

[0024] ;

[0025] in is the preset number of prediction steps, 、 are weight coefficients, is the set reference current;

[0026] In the closed-loop control mode, in each set control cycle, based on the three-phase current data of the current set period and the motor speed data, the least squares method is used to predict the three-phase current data of the next five sampling cycles. In this way, the synthetic vector amplitude and current change rate corresponding to the three-phase current data of the next five sampling cycles are further obtained, and the optimal voltage adjustment amount is calculated according to the control target. And it is transmitted to the SVPWM module of the inverter to suppress the current rising rate.

[0027] Furthermore, the inverter current mutation control method for the compressed air energy storage system further includes:

[0028] Step 4: Perform feedforward compensation and load prediction.

[0029] Furthermore, in step 4, the feedforward compensation collects the load pressure, temperature and air flow velocity parameters of the compressed air energy storage system, combines the neural network prediction model to predict the load current change trend of the motor, and adjusts the control parameters of the inverter in advance based on the prediction results, thereby realizing the control strategy of advance control.

[0030] Furthermore, in step 4, the PLC controller collects the load pressure of the compressed air energy storage system through the pressure sensor, temperature sensor, and speed sensor connected to it. ,temperature , air flow velocity Parameters, the PLC controller builds an LSTM neural network prediction model to predict the trend of motor load current changes. The input of the LSTM neural network prediction model for:

[0031] ;

[0032] in For the collection of The load pressure of a compressed air energy storage system, For the collection of The temperature of the compressed air energy storage system, For the collection of The air flow velocity of a compressed air energy storage system, For the collection of The load pressure of a compressed air energy storage system, For the collection of The temperature of the compressed air energy storage system, For the collection of Air flow velocity of a compressed air energy storage system;

[0033] The output of the LSTM neural network prediction model is the predicted motor load current change trend .

[0034] Furthermore, in step 4, based on the prediction result, the control parameters of the frequency converter are adjusted in advance, thereby implementing the control strategy of advance control, including:

[0035] Step 4-1: Dynamically adjust the voltage / frequency ratio of the inverter;

[0036] Step 4-2: Dynamically adjust the dead zone compensation coefficient of the inverter.

[0037] Further, in step 4-1, based on the predicted motor load current change trend as the prediction result , the PLC controller dynamically adjusts the voltage / frequency ratio of the inverter. The adjustment formula for the voltage / frequency ratio of the inverter dynamically adjusted by the PLC controller is:

[0038] ;

[0039] in is the voltage / frequency ratio of the inverter after dynamic adjustment, is the current voltage / frequency ratio of the inverter before dynamic adjustment, is the proportional coefficient.

[0040] Further, in step 4-2, based on the predicted motor load current change trend as the prediction result , the PLC controller dynamically adjusts the dead zone compensation coefficient of the inverter. The adjustment formula for the dead zone compensation coefficient of the inverter dynamically adjusted by the PLC controller is:

[0041] ;

[0042] in is the dead zone compensation coefficient of the inverter after dynamic adjustment, is the dead zone compensation coefficient of the inverter before dynamic adjustment, is the integration coefficient.

[0043] The beneficial effects of the present invention are as follows:

[0044] The present invention collects the output current data of the inverter of the compressed air energy storage system in real time; filters and differentiates the current data to obtain the current change rate; compares the current change rate with a preset dynamic threshold to determine whether there is a current mutation trend; if it is determined that there is a mutation trend, the closed-loop control mode is triggered to dynamically adjust the output frequency and voltage of the inverter to limit the current rise rate; at the same time, a feedforward compensation mechanism is introduced to adjust the control parameters in advance according to the load change trend, thereby improving the system response speed and control accuracy. The method of the present invention dynamically detects the current change trend of the inverter of the compressed air energy storage system and combines closed-loop and feedforward control strategies. It can respond quickly before or in the early stage of the current mutation of the inverter of the compressed air energy storage system, effectively suppress the current mutation of the inverter of the compressed air energy storage system, improve the control accuracy and system stability, and is suitable for compressed air energy storage systems under complex working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flow chart of the inverter current mutation control method for a compressed air energy storage system in the present invention. DETAILED DESCRIPTION

[0046] To make the objectives, technical solutions, and advantages of the present invention more clear, the following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely express the technical solutions of the present invention. The embodiments expressed in this application are only part of the embodiments of the present invention, not all of the embodiments. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without making creative work are all within the scope of protection of the present invention.

[0047] like Figure 1 As shown, a method for controlling a sudden change in current of an inverter for a compressed air energy storage system includes:

[0048] Step 1: Real-time acquisition and pre-processing of the inverter current data of the compressed air energy storage system;

[0049] In a preferred but non-limiting embodiment of the present invention, in step 1, a PLC controller uses a connected high-precision Hall effect current sensor to collect three-phase current data from the compressed air energy storage system's inverter output in real time, with a sampling frequency of no less than 20kHz to ensure that transient current changes are captured. The three-phase current data collected by the PLC controller is then de-noised using a bandpass filter to eliminate high-frequency interference and DC offset, thereby improving the quality of the three-phase current data. Subsequently, a sliding window difference algorithm is used to differentiate the de-noised three-phase current data to obtain a time series of the current rate of change (di / dk), providing basic data for subsequent mutation trend identification.

[0050] For example, the PLC controller uses a connected three-phase high-precision Hall effect current sensor (model: LEM LTS 25-NP) to collect real-time three-phase current data from the compressed air energy storage system's inverter output to the motor. The sampling frequency is set to 20kHz to capture rapidly changing current data. The collected raw current data first passes through a second-order bandpass filter (with cutoff frequencies of 100Hz and 5kHz, respectively) to remove DC offset and high-frequency noise. The PLC controller is also connected to the compressed air energy storage system's inverter.

[0051] In a preferred but non-limiting embodiment of the present invention, in step 1, the method of performing differential processing on the denoised three-phase current data using a sliding window difference algorithm to obtain a time series of the current change rate (di / dk) includes:

[0052] In order to extract the current mutation characteristics, the PLC controller uses the sliding window difference algorithm to process the three-phase current data after denoising. The three-phase current data after denoising at each acquisition moment are , , , then the first phase current data resultant vector magnitude for:

[0053] ;

[0054] Then, the PLC controller uses a sliding window with a length of 10 acquisition points to calculate the Current change rate :

[0055] ;

[0056] in , , For the The resultant vector magnitude, For the The window time is 0.5ms, which is the time it takes to collect ten current data when the current sensor sampling frequency is 20kHz.

[0057] Step 2: Perform dynamic slope detection and mutation trend identification;

[0058] In a preferred but non-limiting embodiment of the present invention, in step 2, a dynamic threshold comparison is performed on the current rate of change based on a sliding window mechanism. This step employs an adaptive threshold algorithm to adjust the upper threshold of the current rate of change in real time based on current operating conditions (e.g., load level, speed, and temperature), thereby avoiding the problem of false or missed detections by fixed thresholds under different operating conditions. If the current rate of change exceeds the dynamic threshold for multiple consecutive sampling periods, a sudden current change trend is determined, and a sudden change warning signal is transmitted to an LCD screen connected to the PLC controller for display.

[0059] In a preferred but non-limiting embodiment of the present invention, in step 2, in order to accurately identify the current mutation, the system adopts a dynamic threshold comparison mechanism. Dynamic threshold Determined by the current load status and operating frequency, Dynamic threshold The calculation formula is:

[0060] ;

[0061] in , , 、 It can also be designed according to specific requirements. The current value of the motor connected to the inverter The motor speed data (unit: rad / s) is collected by a speed sensor connected to the PLC controller and transmitted to the PLC controller. The sampling frequency of the speed sensor is consistent with the sampling frequency of the current sensor.

[0062] When detected and When the state lasts for more than three sampling windows (one sampling window is one window time) in a row, the PLC controller determines that there is a current mutation trend and sends a mutation warning signal to the LCD screen connected to the PLC controller for display.

[0063] Step 3: By optimizing the voltage adjustment , minimize the future The weighted sum of the “square of current deviation” and the “square of current change rate” within the prediction step is used to perform closed-loop control response;

[0064] In a preferred but non-limiting embodiment of the present invention, in step 3, while simultaneously transmitting a sudden change warning signal to an LCD screen connected to the PLC controller for display, the PLC controller initiates a closed-loop control mode to adjust the inverter's output frequency and voltage control parameters. This closed-loop control mode utilizes an improved model predictive control (MPC) algorithm, combining the current operating status of the compressed air energy storage system with the motor model to predict future current trends and optimize the output voltage vector in real time to limit the current rise rate. The control cycle is less than 1ms, ensuring fast response and high-precision control.

[0065] In a preferred but non-limiting embodiment of the present invention, in step 3, the closed-loop control mode adopts a model predictive control (MPC) strategy to achieve active limitation of the current change rate. Its control objectives are:

[0066] ;

[0067] in is the preset number of prediction steps (can be set according to specific requirements), 、 Are all weight coefficients (weight coefficients can be set according to specific requirements), is the set reference current;

[0068] In the closed-loop control mode, in each set control cycle (less than 1ms), based on the three-phase current data (dependent variable) of the current set period and the motor speed data (independent variable), the least squares method is used to predict the three-phase current data of the next five sampling cycles (one sampling cycle is one window time). In this way, the synthetic vector amplitude and current change rate corresponding to the three-phase current data of the next five sampling cycles (one sampling cycle is one window time, and the number of prediction steps is the number of three-phase current data sampled in the five sampling cycles) are further obtained, and the optimal voltage adjustment amount is calculated according to the control target. And it is transmitted to the SVPWM module of the inverter to suppress the current rising rate.

[0069] For example, while sending a sudden change warning signal to an LCD screen connected to a PLC controller for display, the closed-loop control mode calculates that the output voltage needs to be reduced by 5V to limit the current rise rate. This voltage adjustment command to reduce the output voltage by 5V is sent to the inverter's SVPWM module, enabling a quick response.

[0070] In a preferred but non-limiting embodiment of the present invention, the inverter current mutation control method for a compressed air energy storage system further includes:

[0071] Step 4: Perform feedforward compensation and load prediction.

[0072] Through the above-mentioned implementation, the present invention realizes high-precision and high-response control of the sudden change of the inverter current of the compressed air energy storage system, which significantly improves the operating stability and safety of the compressed air energy storage system.

[0073] In a preferred but non-limiting embodiment of the present invention, in step 4, a feedforward compensation mechanism is introduced to improve the responsiveness of the compressed air energy storage system's inverter before a sudden current change. Feedforward compensation collects parameters such as the compressed air energy storage system's load pressure, temperature, and airflow velocity, and uses a neural network prediction model (such as an LSTM) to predict the motor's load current trends. Based on the predictions, inverter control parameters, such as the voltage / frequency ratio and dead-band compensation coefficient, are adjusted in advance, thereby implementing a proactive control strategy and effectively reducing the probability of sudden current changes.

[0074] In a preferred but non-limiting embodiment of the present invention, in step 4, the PLC controller collects the load pressure of the compressed air energy storage system through the pressure sensor, temperature sensor, and speed sensor connected thereto. ,temperature , air flow velocity Parameters, the sampling frequency of the pressure sensor, temperature sensor, speed sensor and current sensor are consistent. The PLC controller builds an LSTM neural network prediction model to predict the trend of motor load current changes. The input of the LSTM neural network prediction model is for:

[0075] ;

[0076] in For the collection of The load pressure of a compressed air energy storage system, For the collection of The temperature of the compressed air energy storage system, For the collection of The air flow velocity of a compressed air energy storage system, For the collection of The load pressure of a compressed air energy storage system, For the collection of The temperature of the compressed air energy storage system, For the collection of Air flow velocity of a compressed air energy storage system;

[0077] The output of the LSTM neural network prediction model is the predicted motor load current change trend .

[0078] In a preferred but non-limiting embodiment of the present invention, in step 4, based on the prediction results, the control parameters of the inverter, such as the voltage / frequency ratio, the dead zone compensation coefficient, etc., are adjusted in advance to implement the control strategy of advance control, including:

[0079] Step 4-1: Dynamically adjust the voltage / frequency ratio of the inverter;

[0080] In a preferred but non-limiting embodiment of the present invention, in step 4-1, based on the predicted motor load current change trend as the prediction result, , the PLC controller dynamically adjusts the voltage / frequency ratio of the inverter, and the voltage / frequency ratio is recorded as If it is predicted that the motor load current will increase in the future, the PLC controller will increase the ratio to improve the motor torque response capability; if it is predicted that the load will decrease, the Ratio, to avoid current mutation caused by voltage overshoot, thereby ultimately improving the motor's responsiveness and avoiding current mutation caused by load mutation. The adjustment formula for the PLC controller to dynamically adjust the voltage / frequency ratio of the inverter is:

[0081] ;

[0082] in is the voltage / frequency ratio of the inverter after dynamic adjustment, is the current voltage / frequency ratio of the inverter before dynamic adjustment, It is the proportional coefficient, which can be set according to the specific requirements of the inverter's inertia and response time.

[0083] Step 4-2: Dynamically adjust the dead zone compensation coefficient of the inverter.

[0084] In a preferred but non-limiting embodiment of the present invention, in step 4-2, based on the predicted motor load current change trend as the prediction result , the PLC controller dynamically adjusts the dead zone compensation coefficient of the inverter, and the dead zone compensation coefficient is recorded as If the current is predicted to rise, the PLC controller increases the dead zone compensation coefficient in advance to reduce the current distortion caused by the dead zone effect. If the current is predicted to fall, the dead zone compensation coefficient is appropriately reduced to prevent current fluctuations caused by overcompensation, thereby ultimately improving the accuracy of the PWM waveform and reducing current distortion. The adjustment formula for the PLC controller to dynamically adjust the dead zone compensation coefficient of the inverter is:

[0085] ;

[0086] in is the dead zone compensation coefficient of the inverter after dynamic adjustment, is the dead zone compensation coefficient of the inverter before dynamic adjustment, It is the integral coefficient, which can be used to adjust the compensation intensity and can be set according to specific requirements.

[0087] For example, in a certain compressed air energy storage system, a frequency converter drives a high-power asynchronous motor for air compression. During operation, sudden changes in load pressure or delayed control response often cause the asynchronous motor's current to fluctuate dramatically, leading to problems such as power device overload, increased electromagnetic interference, and decreased system efficiency. To address these issues, a high-precision, high-response control method for current fluctuations has been developed based on the aforementioned technical solution, achieving promising results in actual operation.

[0088] The beneficial effects of the present invention are as follows:

[0089] The present invention collects the output current data of the inverter of the compressed air energy storage system in real time; filters and differentiates the current data to obtain the current change rate; compares the current change rate with a preset dynamic threshold to determine whether there is a current mutation trend; if it is determined that there is a mutation trend, the closed-loop control mode is triggered to dynamically adjust the output frequency and voltage of the inverter to limit the current rise rate; at the same time, a feedforward compensation mechanism is introduced to adjust the control parameters in advance according to the load change trend, thereby improving the system response speed and control accuracy. The method of the present invention dynamically detects the current change trend of the inverter of the compressed air energy storage system and combines closed-loop and feedforward control strategies. It can respond quickly before or in the early stage of the current mutation of the inverter of the compressed air energy storage system, effectively suppress the current mutation of the inverter of the compressed air energy storage system, improve the control accuracy and system stability, and is suitable for compressed air energy storage systems under complex working conditions.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not deviate from the spirit and scope of the present invention should be covered within the protection space of the claims of the present invention.

Claims

1. A method for controlling sudden current changes in a frequency converter of a compressed air energy storage system, characterized in that: include: Step 1: Real-time acquisition and pre-processing of the inverter current data of the compressed air energy storage system; Step 2: Perform dynamic slope detection and mutation trend identification; Step 3: By optimizing the voltage adjustment , minimize the future The weighted sum of the "square of current deviation" and the "square of current change rate" within the prediction step is used to perform closed-loop control response; The method for controlling sudden current changes in a frequency converter of a compressed air energy storage system further includes: Step 4: Perform feedforward compensation and load prediction; In step 4, feedforward compensation collects the load pressure, temperature, and airflow velocity parameters of the compressed air energy storage system and uses a neural network prediction model to predict the load current change trend of the motor. Based on the prediction results, the control parameters of the inverter are adjusted in advance, thereby realizing the control strategy of advance control. In step 4, based on the prediction result, the control parameters of the inverter are adjusted in advance, thereby implementing the control strategy of advance control, including: Step 4-1: Dynamically adjust the voltage / frequency ratio of the inverter; Step 4-2: Dynamically adjust the dead zone compensation coefficient of the inverter.

2. The method for controlling a sudden change in current of a frequency converter for a compressed air energy storage system according to claim 1, wherein: In step 1, the PLC controller collects the three-phase current data output by the inverter of the compressed air energy storage system in real time through the current sensor connected to it. The three-phase current data collected by the PLC controller is denoised by a bandpass filter. Subsequently, the denoised three-phase current data is differentiated using a sliding window difference algorithm to obtain a time series of the current change rate.

3. The inverter current sudden change control method for a compressed air energy storage system according to claim 2, characterized in that: In step 1, a method for performing differential processing on the three-phase current data after denoising using a sliding window difference algorithm to obtain a time series of current change rate includes: Set up the first The three-phase current data after denoising at each acquisition moment are , , , then the first phase current data resultant vector magnitude for: ； Then, the PLC controller calculates the Current change rate : ； in , , For the The resultant vector magnitude, For the The resultant vector magnitude.

4. The method for controlling a sudden change in current of a frequency converter for a compressed air energy storage system according to claim 3, wherein: In step 2, based on the sliding window mechanism, a dynamic threshold comparison is performed on the current change rate. When it is detected that the current change rate exceeds the dynamic threshold for multiple consecutive sampling cycles, it is determined that there is a current mutation trend, and a mutation warning signal is sent to the LCD screen connected to the PLC controller for display; In step 2, Dynamic threshold The calculation formula is: ； in , , The first The motor speed data is collected by the speed sensor connected to the PLC controller and transmitted to the PLC controller; When detected and When the state exceeds three sampling windows continuously, the PLC controller determines that there is a current mutation trend and sends a mutation warning signal to the LCD screen connected to the PLC controller for display.

5. The method for controlling a sudden change in current of a frequency converter for a compressed air energy storage system according to claim 4, wherein: In step 3, while sending a sudden change warning signal to the LCD screen connected to the PLC controller for display, the PLC controller starts the closed-loop control mode and adjusts the output frequency and voltage control parameters of the inverter; In step 3, the closed-loop control mode adopts the model predictive control strategy, and its control objective is: ； in is the preset number of prediction steps, 、 are weight coefficients, is the set reference current; In the closed-loop control mode, in each set control cycle, based on the three-phase current data of the current set period and the motor speed data, the least squares method is used to predict the three-phase current data of the next five sampling cycles. In this way, the synthetic vector amplitude and current change rate corresponding to the three-phase current data of the next five sampling cycles are further obtained, and the optimal voltage adjustment amount is calculated according to the control target. And it is transmitted to the SVPWM module of the inverter to suppress the current rising rate.

6. The method for controlling a sudden change in current of a frequency converter for a compressed air energy storage system according to claim 5, characterized in that: In step 4, the PLC controller collects the load pressure of the compressed air energy storage system through the pressure sensor, temperature sensor, and speed sensor connected to it. ,temperature , air flow velocity Parameters, the PLC controller builds an LSTM neural network prediction model to predict the trend of motor load current changes. The input of the LSTM neural network prediction model for: ； in For the collection of The load pressure of a compressed air energy storage system, For the collection of The temperature of the compressed air energy storage system, For the collection of The air flow velocity of a compressed air energy storage system, For the collection of The load pressure of a compressed air energy storage system, For the collection of The temperature of the compressed air energy storage system, For the collection of Air flow velocity of a compressed air energy storage system; The output of the LSTM neural network prediction model is the predicted motor load current change trend .

7. The inverter current sudden change control method for a compressed air energy storage system according to claim 6, characterized in that: In step 4-1, based on the predicted motor load current change trend as the prediction result , the PLC controller dynamically adjusts the voltage / frequency ratio of the inverter. The adjustment formula for the voltage / frequency ratio of the inverter dynamically adjusted by the PLC controller is: ； in is the voltage / frequency ratio of the inverter after dynamic adjustment, is the current voltage / frequency ratio of the inverter before dynamic adjustment, is the proportional coefficient.

8. The method for controlling sudden current changes of a frequency converter for a compressed air energy storage system according to claim 7, wherein: In step 4-2, based on the predicted motor load current change trend as the prediction result , the PLC controller dynamically adjusts the dead zone compensation coefficient of the inverter. The adjustment formula for the dead zone compensation coefficient of the inverter dynamically adjusted by the PLC controller is: ； in is the dead zone compensation coefficient of the inverter after dynamic adjustment, k is the dead zone compensation coefficient of the inverter before dynamic adjustment, i is the integration coefficient.

Images