A method and system for full-cycle out-of-step detection of electronic expansion valves based on high-frequency signal injection

By using high-frequency signal injection and dynamic bandpass filtering technology, full-cycle step loss detection of the electronic expansion valve is achieved, which solves the problem of inaccurate refrigerant flow regulation caused by stepper motor step loss, improves the stability and control accuracy of the refrigeration system, and reduces hardware costs.

CN123065732APending Publication Date: 2026-11-10XINLEI COMPRESSOR CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202611137960.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-29
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

In existing electronic expansion valve control, the stepper motor may lose steps, stall, or passively shift its position due to factors such as changes in refrigerant pressure differential, sudden load changes, and fluctuations in drive voltage. This can lead to inaccurate refrigerant flow regulation, affecting the stability of the refrigeration system. Furthermore, existing high-frequency signal injection methods are insufficient to accurately determine the actual mechanical position of the valve needle.

Method used

By employing a high-frequency signal injection method, a high-frequency AC voltage signal is injected into the stepper motor winding. Combined with dynamic bandpass filtering and synchronous demodulation, dual-frequency complex impedance characteristics are extracted to generate electrical position and perform mechanical backlash compensation. This determines the actual mechanical position of the valve needle, enabling full-cycle step loss detection without the need for additional position sensors.

Benefits of technology

Accurately detects valve needle position under static, uniform speed, and acceleration/deceleration conditions, suppresses the effects of load disturbances and winding parameter changes, improves the control accuracy and reliability of the refrigeration system, and reduces hardware costs and structural complexity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123065732A_ABST
    Figure CN123065732A_ABST
Patent Text Reader

Abstract

This invention relates to the field of electronic expansion valve control and stepper motor sensorless detection technology for refrigeration equipment, and particularly to a method and system for full-cycle out-of-step detection of electronic expansion valves based on high-frequency signal injection. The method injects a high-frequency AC signal with a frequency misalignment from the PWM carrier into the stepper motor windings, acquires the winding current response, and performs dynamic bandpass filtering with center frequency tracking. Amplitude and phase features are extracted through locked-amplification synchronous demodulation or dual-frequency injection ratio method. The electrical position is inverted based on a pre-calibrated feature-rotor position mapping relationship, and the actual mechanical position of the valve needle is obtained through mechanical backlash compensation. The actual mechanical position is compared with the commanded position, and a dynamic threshold is set according to the valve body operating state to determine the degree of out-of-step detection and execute deviation compensation, reset, or shutdown protection. This invention reuses existing drive and current sampling circuits, eliminating the need for additional position sensors, and can achieve millisecond-level out-of-step detection with resistance to temperature drift and load disturbances under static, uniform speed, and acceleration / deceleration states.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of electronic expansion valve control for refrigeration equipment and sensorless detection technology for stepper motors, and particularly to a method and system for full-cycle out-of-step detection of electronic expansion valves based on high-frequency signal injection. Background Technology

[0002] Electronic expansion valves are throttling elements used in air conditioners, heat pumps, and other refrigeration equipment to regulate refrigerant flow. They are typically driven by a stepper motor, which rotates the rotor and, through a threaded transmission mechanism, moves the valve needle axially to change the flow area of ​​the valve orifice. Most existing electronic expansion valves use open-loop control; the controller determines the commanded opening degree based on the number of output drive pulses, but it usually cannot directly obtain the actual mechanical position of the valve needle.

[0003] During operation, the electronic expansion valve is susceptible to various factors such as refrigerant pressure differential changes, sudden load changes, drive voltage fluctuations, electromagnetic interference, mechanical friction, and abnormal drive pulses. These factors can cause the stepper motor to lose steps, stall, or experience passive position deviation, resulting in a discrepancy between the actual and commanded valve needle positions. Such positional deviations can lead to inaccurate refrigerant flow regulation, consequently affecting the evaporator outlet superheat and the operational stability of the refrigeration system. Determining the electronic expansion valve opening solely based on the cumulative value of the drive pulses is insufficient to promptly identify actual positional deviations.

[0004] Chinese invention patent application CN103673426A discloses an opening feedback electronic expansion valve and its control system. This design incorporates a rotary potentiometer within the electronic expansion valve, which meshes with the expansion valve gear. The potentiometer's resistance changes with the valve needle position, thus obtaining the actual opening degree of the electronic expansion valve and achieving closed-loop control. While this design can directly feedback the valve needle position, it requires adding a potentiometer and gear transmission structure within the valve body, placing additional demands on the valve body structure, sealing assembly, and electrical interfaces, and increasing the number of components and manufacturing costs.

[0005] On the other hand, high-frequency signal injection technology has been applied to sensorless position detection of motors. Chinese invention patent application CN109617487A discloses a method for observing the rotor position of a permanent magnet synchronous motor based on high-frequency voltage signal injection. This method injects a high-frequency square wave voltage signal into the direct axis of a synchronous rotating coordinate system, extracts the high-frequency current response containing rotor position information based on the sampled current, and obtains rotor position and speed information using a phase-locked loop. This method can be used for position observation of permanent magnet synchronous motors at zero speed and low speed.

[0006] However, the aforementioned high-frequency signal injection methods are primarily designed for permanent magnet synchronous motors using vector control, obtaining only the rotor's electrical position. The stepper motors used in electronic expansion valves typically exhibit characteristics such as a large number of pole pairs, significant magnetic circuit nonlinearity, and periodic repetition of electrical positions. Furthermore, the valve needle movement is also affected by refrigerant pressure differential and mechanical transmission idle distance. When general high-frequency signal injection position observation methods are directly applied to electronic expansion valves, problems may still arise, such as low-frequency load disturbances affecting high-frequency response extraction, the difficulty in uniquely corresponding the electrical position with the valve needle's mechanical position, and temperature changes causing winding parameter drift. These issues make it difficult to directly determine the step-out deviation between the actual mechanical position and the commanded position of the valve needle.

[0007] Therefore, it is still necessary to provide an electronic expansion valve out-of-step detection technology that does not require the addition of a position sensor inside the electronic expansion valve, can obtain the actual mechanical position of the valve needle when the valve is stationary, moving at a constant speed, or accelerating or decelerating, and can suppress the effects of load disturbances and changes in winding parameters. Summary of the Invention

[0008] This invention aims to provide a method and system for detecting out-of-step movement of an electronic expansion valve throughout its entire lifecycle based on high-frequency signal injection. Without adding a position sensor or changing the mechanical structure of the electronic expansion valve, it enables the detection of the actual mechanical position and judgment of out-of-step movement of the valve in static, uniform motion, and acceleration / deceleration states.

[0009] Firstly, in order to achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0010] A method for detecting full-cycle step loss in an electronic expansion valve based on high-frequency signal injection includes:

[0011] S1. When the electronic expansion valve is in a stationary, uniform speed or acceleration / deceleration state, a candidate position range is established based on the most recent effective position or reset zeroing position, the cumulative command steps and the direction of movement. The first injection frequency and the second injection frequency are selected from the calibration feature library, and the corresponding high-frequency AC voltage signal is injected into the stepper motor winding.

[0012] S2. Acquire the winding current response, pre-filter it using a dynamic bandpass filter whose center frequency follows the corresponding injection frequency, and determine the response confidence level based on the high-frequency response energy and the low-frequency disturbance energy.

[0013] S3. When the response confidence reaches the preset threshold, the pre-filtered current response is synchronously demodulated to determine the dual-frequency complex impedance, equivalent resistance, position-dependent inductance and impedance phase, and the electrical position within a magnetic pole cycle is obtained according to the calibration feature library.

[0014] S4. Generate full-stroke candidate positions based on the electrical position and the number of pole pairs of the stepper motor, determine the unique candidate position that falls within the candidate position range, and perform mechanical backlash compensation according to the direction of movement to obtain the actual mechanical position of the valve needle; if the candidate position is not unique, mark it as an invalid position.

[0015] S5. Compare the actual mechanical position of the valve needle with the commanded position, and determine the step loss result based on the step loss judgment threshold corresponding to the operating state; if the response confidence is insufficient or the position is invalid, maintain the most recent valid position and trigger re-detection or reset zeroing.

[0016] As a further improvement, the calibration feature library is established as follows: the electronic expansion valve is controlled to traverse the entire stroke along both the opening and closing directions. Under multiple calibration temperatures, the actual mechanical position of the valve needle is determined using an external position measuring device. High-frequency AC voltage signals of multiple candidate frequencies are injected at each calibration position, and the corresponding complex voltage, complex current, position-related inductance, and impedance phase are recorded. Calibration points are densified in intervals where the characteristic distance between adjacent positions is less than 60% of the average characteristic distance between adjacent positions throughout the entire stroke. Calibration data for the opening direction, calibration data for the closing direction, mechanical idle distance for the opening direction, and mechanical idle distance for the closing direction are stored respectively.

[0017] Preferably, the first injection frequency and the second injection frequency are both between 1kHz and 20kHz, and the frequency difference between them is no more than one-fifth of the lower injection frequency; the first injection frequency and the second injection frequency are selected based on the minimum adjacent feature distance of the entire stroke, and avoid the PWM carrier frequency, integer multiples of the PWM carrier frequency and the corresponding sideband frequency; the high-frequency response current generated by the high-frequency AC voltage signal does not exceed 10% of the rated drive current of the stepper motor.

[0018] As a further improvement, in step S2, the response confidence level is determined according to the following formula:

[0019] ;

[0020] In the formula, Indicates the confidence level of the response; This represents the total energy of the high-frequency response at the first and second injection frequencies. This indicates low-frequency disturbance energy with a frequency below 100Hz; This indicates a positive number that prevents the denominator from being zero.

[0021] The sampling frequency of the winding current response is not less than 10 times the second injection frequency; when the response confidence is lower than the preset threshold corresponding to the current operating state, or when current sampling is clipped, the current detection is marked as a low confidence detection.

[0022] As a further improvement, in step S3, the first The complex impedance corresponding to each injection frequency is determined by the following formula:

[0023] ;

[0024] In the formula, Indicates the first The complex impedance corresponding to each injection frequency; Indicates the first The complex voltage corresponding to each injection frequency; Indicates the first The complex current corresponding to each injection frequency; The injection frequency number is either 1 or 2;

[0025] The equivalent resistance is determined according to the following formula:

[0026] ;

[0027] In the formula, This represents the equivalent resistance of the stepper motor windings; and These represent the complex impedances corresponding to the first injection frequency and the second injection frequency, respectively. This indicates taking the real part of a complex number;

[0028] The location-dependent inductance is determined according to the following formula:

[0029] ;

[0030] In the formula, This represents the position-dependent inductance after dual-frequency fusion; This indicates taking the imaginary part of a complex number; and These represent the first injection frequency and the second injection frequency, respectively. Represents pi;

[0031] Preferably, in step S3, the position-related inductance and the phases of the first complex impedance and the second complex impedance are used to form a position feature vector:

[0032] ;

[0033] In the formula, Represents the location feature vector; This represents the position-dependent inductance after dual-frequency fusion; and These represent the phase angles of the first complex impedance and the second complex impedance, respectively.

[0034] Select the valve opening direction calibration data or valve closing direction calibration data according to the current direction of motion, and query or interpolate the corresponding calibration data to obtain the electrical position within a magnetic pole cycle based on the current temperature or the winding temperature determined by the equivalent resistance.

[0035] As a further improvement, in step S4, the first The candidate positions corresponding to each electrical cycle are determined according to the following formula:

[0036] ;

[0037] In the formula, Indicates the first Candidate step positions corresponding to each electrical cycle; This indicates the number of drive steps corresponding to each mechanical revolution of the stepper motor rotor; This indicates the number of pole pairs in a stepper motor; Indicates the electrical position within one magnetic pole cycle; Indicates the electrical cycle number, rounded to the nearest integer; This indicates the step offset corresponding to the fully closed zero position; Represents pi;

[0038] Retain candidate positions within the effective full stroke of the electronic expansion valve, and determine the only candidate position that falls within the candidate position range as the rotor step position; if there is no candidate position or two or more candidate positions fall within the candidate position range, mark this detection as an invalid position;

[0039] Preferably, with the valve opening direction as the direction of step increase, the mechanical backlash compensation in step S4 includes:

[0040] During valve opening, the actual mechanical position of the valve needle is determined according to the following formula:

[0041] ;

[0042] In the formula, Indicates the actual mechanical position of the valve needle; This indicates a uniquely determined rotor step position; Indicates the mechanical backlash compensation value for the valve opening direction;

[0043] During valve closing, the actual mechanical position of the valve needle is determined according to the following formula:

[0044] ;

[0045] In the formula, Indicates the actual mechanical position of the valve needle; This indicates a uniquely determined rotor step position; This indicates the mechanical backlash compensation value for the valve closing direction;

[0046] After the direction of motion changes, the most recent valid mechanical position before the change is maintained until the cumulative number of drive steps in the new direction of motion reaches the corresponding mechanical idle compensation value.

[0047] As a further improvement, in step S5, the step loss judgment threshold for the stationary state is 1 step, and the step loss judgment threshold for the moving state is the larger of the value of 2 steps and the nominal positioning error plus 1 step; when the position deviation reaches the step loss judgment threshold but is less than 8 steps, the step loss event is recorded and an early warning and position compensation are executed; when the position deviation reaches or exceeds 8 steps, multiple consecutive detections are judged as step loss, or multiple consecutive position detections are invalid, the current drive is stopped and a reset and zeroing are executed; if a valid position cannot be obtained after the reset and zeroing, a serious fault signal is output.

[0048] Secondly, the present invention also provides a full-cycle out-of-step detection system for an electronic expansion valve based on high-frequency signal injection. This system is used to implement the method described above and includes a controller, a stepper drive circuit, a current sampling circuit, a memory, and an electronic expansion valve. The controller includes:

[0049] An injection frequency selection module is used to select a first injection frequency and a second injection frequency based on the candidate position range, position separability, and PWM carrier spectrum.

[0050] The high-frequency signal injection module is used to inject a corresponding high-frequency AC voltage signal into the stepper motor winding of the electronic expansion valve through the stepper drive circuit.

[0051] The response confidence determination module is used to perform dynamic bandpass filtering on the winding current response acquired by the current sampling circuit, and determine the response confidence based on the high-frequency response energy and the low-frequency disturbance energy.

[0052] The impedance feature extraction module is used to perform synchronous demodulation when the response confidence reaches a preset threshold, determine the dual-frequency complex impedance, equivalent resistance, position-dependent inductance and impedance phase, and obtain the electrical position according to the calibration feature library.

[0053] The candidate position disambiguation module is used to generate full-stroke candidate positions based on the electrical position and the number of pole pairs of the stepper motor, determine the unique candidate position that falls within the candidate position range, and perform mechanical backlash compensation according to the direction of movement to obtain the actual mechanical position of the valve needle.

[0054] The out-of-step detection module is used to compare the actual mechanical position of the valve needle with the commanded position, and output the result of normal, warning, position compensation, reset zeroing or serious fault according to the out-of-step detection threshold corresponding to the operating status.

[0055] As a further improvement, the current sampling circuit reuses the original current sampling resistor, amplifier circuit and analog-to-digital conversion channel of the stepper drive circuit; the electronic expansion valve does not have a potentiometer, Hall sensor or encoder for feedback valve needle position.

[0056] And / or, the memory is used to store the candidate injection frequency set, the position separability corresponding to the candidate injection frequency, valve opening direction calibration data, valve closing direction calibration data, valve opening direction mechanical idle distance, valve closing direction mechanical idle distance, the most recent valid position, response confidence threshold, out-of-step judgment threshold, and out-of-step event records; the controller triggers detection when the electronic expansion valve completes the stepping action, an overheating abnormality occurs, the compressor starts or stops, the system is powered on, or the operating mode is switched, and extends the detection interval when the electronic expansion valve is stationary and continuous detection is normal.

[0057] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the method described above.

[0058] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described thereon.

[0059] This invention injects a high-frequency AC signal into the stepper motor windings and acquires the corresponding current response. Combined with dynamic bandpass filtering, synchronous demodulation, and dual-frequency impedance feature extraction, it can suppress the influence of low-frequency load disturbances caused by refrigerant pressure differential changes, PWM switching noise, and winding resistance temperature drift on the detection results. The rotor electrical position is obtained through a pre-established calibration relationship between frequency domain features and electrical position. Periodic position disambiguation is performed using a candidate position range, and combined with mechanical backlash compensation corresponding to the opening and closing direction of the electronic expansion valve, the periodic electrical position is converted into the actual mechanical position of the valve needle, thereby improving the accuracy of full-stroke position detection. By comparing the actual mechanical position with the commanded position and setting the step loss judgment threshold and response confidence conditions according to the valve's operating state, the invention can promptly identify step loss, stall, and passive position deviation in static, uniform, and accelerated / decelerated motion states. Simultaneously, the invention can reuse the current sampling resistor and analog-to-digital conversion channel in the original drive circuit of the electronic expansion valve, eliminating the need for additional potentiometers, Hall sensors, or encoders. This reduces hardware costs and structural complexity. Furthermore, it can perform recording and warning, deviation compensation, re-detection, reset and zero-finding, or shutdown protection based on the degree of position deviation, improving the control accuracy of the electronic expansion valve and the operational reliability of the refrigeration system. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of the overall structure of the electronic expansion valve full-cycle out-of-step detection system of the present invention.

[0061] Figure 2 This is a schematic diagram illustrating the principle of dual-frequency high-frequency signal injection, dynamic bandpass filtering, and synchronous demodulation of the present invention.

[0062] Figure 3 This is a flowchart of the electronic expansion valve full-cycle out-of-step detection method of the present invention.

[0063] Figure 4 This is a flowchart illustrating the establishment of the calibration feature library and the selection of injection frequency in this invention.

[0064] Figure 5 This is a schematic diagram illustrating the principles of electrical position candidate position generation, periodic position disambiguation, and mechanical idle distance compensation in this invention.

[0065] Figure 6 This is a schematic diagram of the process for determining the position validity, determining the loss of synchronization, and handling fault classification based on the operating status of the present invention.

[0066] Figure 7 This is a comparison chart of the average absolute position error of the present invention under different temperatures and different motion states.

[0067] Figure 8 This is a comparison chart of the out-of-step detection rate, false alarm rate, and response time of the present invention and different out-of-step detection methods. Detailed Implementation

[0068] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings. It should be noted that the following embodiments are used to illustrate the technical concept and possible implementation methods of the present invention, and are not intended to limit the scope of protection of the present invention. Equivalent substitutions made by those skilled in the art to the signal form, sampling parameters, filtering methods, data storage forms, and controller hardware without departing from the technical concept of the present invention should all fall within the scope of protection of the present invention.

[0069] I. Terminology Explanation

[0070] In this embodiment, the electronic expansion valve refers to an electric throttling element that is driven by a stepper motor to rotate a rotor and converts the rotational motion of the rotor into the axial motion of the valve needle through a threaded transmission mechanism to adjust the refrigerant flow area.

[0071] The command position refers to the target step position of the electronic expansion valve obtained by the controller based on the accumulated output step drive pulses. The fully closed position of the electronic expansion valve is taken as the zero position, and the direction in which the valve needle moves away from the valve port is taken as the valve opening direction. The command step number increases for the valve opening direction and decreases for the valve closing direction.

[0072] Electrical position refers to the position of the stepper motor rotor within one magnetic pole cycle. Since a stepper motor has multiple magnetic pole cycles, different full-stroke mechanical positions may correspond to the same or similar electrical positions. Therefore, the unique mechanical position of the valve needle within the full stroke cannot be directly determined based solely on the electrical position within one magnetic pole cycle.

[0073] The candidate position range refers to the mechanical position search interval established based on the most recent valid position or reset zeroing position, the cumulative number of signed command steps during the detection period, the direction of movement, and the maximum permissible positioning error.

[0074] The actual mechanical position refers to the actual number of steps of the valve needle obtained after obtaining the electrical position based on the high-frequency response characteristics and after periodic position disambiguation and mechanical backlash compensation.

[0075] Mechanical backlash refers to the transmission gap when the electronic expansion valve changes direction, during which the stepper motor rotor has already rotated but the valve needle has not yet produced a corresponding axial displacement. The mechanical backlash in the valve opening direction and the mechanical backlash in the valve closing direction are calibrated and stored separately.

[0076] A valid location is defined as the location where the response confidence level reaches a preset confidence threshold, and only one of the multiple candidate locations generated based on the electrical location falls within the candidate location range. If the response confidence level is insufficient, the candidate location is empty, or the candidate location is not unique, the detection result is marked as an invalid location.

[0077] The response confidence level is used to characterize the reliability of the high-frequency current response at the injection frequency relative to low-frequency load disturbances. The current high-frequency response signal is only used for location inversion when the response confidence level reaches a preset confidence threshold.

[0078] Step loss refers to the deviation between the actual mechanical position of the valve needle and the command position recorded by the controller reaching or exceeding the step loss judgment threshold under the corresponding operating state. This includes situations such as the stepper motor failing to complete the expected rotation according to the drive pulse, jamming of the mechanical transmission mechanism, and passive displacement of the valve needle or rotor caused by external force.

[0079] Full-cycle detection refers to the ability to perform position detection and step loss judgment when the electronic expansion valve is in a stationary state, a uniform motion state, and an acceleration / deceleration motion state, rather than being limited to a single detection time after the electronic expansion valve stops moving.

[0080] II. System Overall Structure

[0081] like Figure 1 As shown, the electronic expansion valve full-cycle out-of-step detection system provided by the present invention includes a controller, a stepper drive circuit, a current sampling circuit, a memory, and an electronic expansion valve.

[0082] The controller is connected to the stepper drive circuit, the current sampling circuit, and the memory. The stepper drive circuit is connected to the stepper motor windings of the electronic expansion valve to generate a normal stepper drive signal and superimpose a high-frequency AC voltage signal onto it. The current sampling circuit is used to acquire the current response generated by the stepper motor windings under the excitation of the high-frequency AC voltage signal.

[0083] The controller can be a microcontroller, a digital signal processor, or a main control chip for refrigeration equipment with digital signal processing capabilities. In one embodiment, the controller has a main frequency of 120MHz, and has at least two PWM output channels, one analog-to-digital conversion channel, and non-volatile memory space for storing a calibration feature library.

[0084] The stepper motor drive circuit can employ a bipolar H-bridge drive circuit or a multiphase switching drive circuit suitable for unipolar stepper motors. The high-frequency AC voltage signal can be a sinusoidal signal, a square wave signal, or an equivalent AC signal generated by PWM modulation.

[0085] The current sampling circuit preferably reuses the existing current sampling resistor, amplifier circuit, and analog-to-digital conversion channel of the stepper drive circuit. Therefore, there is no need to add a potentiometer, Hall sensor, or encoder inside the electronic expansion valve, nor is it necessary to change the valve body structure or existing electrical interfaces of the electronic expansion valve. The analog-to-digital conversion resolution of the current sampling circuit is preferably no less than 12 bits, and its range covers the superposition range of normal drive current and high-frequency response current.

[0086] The memory is used to store the set of candidate injection frequencies, the position separability corresponding to different candidate injection frequencies, valve opening direction calibration feature data, valve closing direction calibration feature data, valve opening direction mechanical idle distance, valve closing direction mechanical idle distance, the most recent valid position, the step loss judgment threshold, the response confidence threshold, and the step loss event record corresponding to each operating state.

[0087] The controller includes an injection frequency selection module, a high-frequency signal injection module, a current response processing module, a response confidence determination module, an impedance feature extraction module, an electrical position inversion module, a candidate position disambiguation module, a mechanical backlash compensation module, a step-out judgment module, and a fault classification processing module. These modules can be implemented through a program running on the same controller, or they can be set up in multiple interconnected processing units.

[0088] like Figure 2As shown, the high-frequency signal injection module is connected to the stepper drive circuit, and the current sampling circuit is sequentially connected to the dynamic bandpass filter unit, the synchronous demodulation unit, the complex impedance determination unit, and the electrical position inversion unit. The dynamic bandpass filter unit is located before the synchronous demodulation unit, and its center frequency follows the current injection frequency to attenuate low-frequency load disturbances and PWM carrier noise caused by refrigerant pressure difference changes before synchronous demodulation.

[0089] III. Specific Implementation Methods

[0090] like Figure 3 As shown, the method of the present invention includes, in sequence: establishing a candidate position range and injecting dual-frequency high-frequency signals; acquiring current response and determining response confidence; extracting dual-frequency impedance features and inverting electrical position; generating candidate positions, determining unique positions and performing mechanical backlash compensation; and judging and classifying out-of-step based on the deviation between the actual mechanical position and the commanded position.

[0091] (a) Establishment of the calibration feature library

[0092] like Figure 4 As shown, a calibration feature library between frequency domain characteristics and electrical position is pre-established before the electronic expansion valve is put into operation. The calibration feature library can be established during the factory testing stage of the electronic expansion valve, or a model-level calibration feature library can be established for electronic expansion valves of the same model, with the same number of magnetic pole pairs and the same transmission structure.

[0093] During calibration, the electronic expansion valve is installed on the calibration fixture, and the actual mechanical position of the valve needle is detected by an external position measuring device. The external position measuring device can be a laser displacement sensor, a grating displacement sensor, or a precision encoder, and its position resolution is preferably no greater than 0.1 steps corresponding to the mechanical displacement.

[0094] First, the electronic expansion valve is controlled to move gradually from the fully closed position to the fully open position. At each calibration position, a high-frequency AC voltage signal from the candidate frequency set is injected sequentially, and the corresponding complex voltage, complex current, equivalent resistance, position-related inductance, and impedance phase are recorded to form valve opening direction calibration data.

[0095] Subsequently, the electronic expansion valve is controlled to move from the fully open position to the fully closed position, and valve closing direction calibration data is generated in the same manner. The valve opening direction calibration data and the valve closing direction calibration data are stored separately to preserve the influence of mechanical travel, friction direction, and hysteresis on the response characteristics.

[0096] The candidate injection frequencies are set in the range of 1 kHz to 20 kHz. In one embodiment, the candidate injection frequencies are set sequentially to 3.2 kHz, 4.0 kHz, 4.8 kHz, 5.6 kHz, 6.4 kHz, 8.0 kHz, and 9.6 kHz. The number and interval of the candidate injection frequencies are determined based on the stepper drive circuit bandwidth, PWM carrier frequency, and stepper motor winding parameters.

[0097] Calibration can be performed at multiple temperature points. In one embodiment, the calibration temperature points are set to -30°C, 0°C, 25°C, 60°C, and 85°C. When the actual operating temperature of the electronic expansion valve is between two adjacent calibration temperature points, the controller obtains the calibration characteristics corresponding to the current temperature through linear interpolation.

[0098] The response amplitude and impedance phase at each calibration location are normalized, and adjacent calibration locations at the candidate injection frequency are normalized. The characteristic distance below is determined by the following formula:

[0099] ;

[0100] In the formula, Indicates the first The first calibration position and the first Each calibrated location at the candidate injection frequency Normalized feature distance; and These represent the normalized response amplitudes at two adjacent calibration locations; and These represent the normalized impedance phase at two adjacent calibration positions, respectively; 0.6 is the response amplitude weight; and 0.4 is the impedance phase weight. The sum of the two weights is 1.

[0101] The positional separability of the candidate injection frequency over the entire stroke range is determined by the following formula:

[0102] ;

[0103] In the formula, Indicates candidate injection frequency Corresponding positional separability; Indicates the first Adjacent locations, temperature points and direction of movement Normalized feature distance; direction of motion This includes the valve opening direction and the valve closing direction.

[0104] Using the minimum adjacent feature distance over the entire stroke range as the evaluation index can avoid selecting injection frequencies that have high resolution only in some locations but are difficult to distinguish from adjacent locations in others.

[0105] When the distance between adjacent features within a certain interval is less than 60% of the average distance between adjacent features over the entire stroke, that interval is defined as the magnetic circuit nonlinear distortion region. Within the magnetic circuit nonlinear distortion region, the calibration interval is reduced to one-half or one-quarter of the normal calibration interval by micro-stepping drive or external calibration fixtures, and the newly added calibration points are interpolated and densified.

[0106] The mechanical idle distance is determined through a motion direction switching test. The rotor is controlled to switch from the valve opening direction to the valve closing direction, and the number of drive steps between the start of the change in the rotor's electrical position and the start of the change in the valve needle's actual mechanical position is recorded as the mechanical idle distance in the valve closing direction. The same method is used to obtain the mechanical idle distance in the valve opening direction when the rotor is controlled to switch from the valve closing direction to the valve opening direction. Each direction is measured at least five times, and the median value is taken as the mechanical idle distance compensation value for that direction. The mechanical idle distance of an electronic expansion valve is typically 1 to 3 steps; the specific value should be determined based on the calibration results of the corresponding valve body.

[0107] (II) Step S1: Establish candidate location range and implement dual-frequency high-frequency signal injection

[0108] The controller first identifies the current operating state of the electronic expansion valve. If no stepping drive pulse is output within two consecutive detection cycles, the electronic expansion valve is determined to be stationary; if the rate of change of the stepping pulse frequency within the detection window does not exceed 10%, the electronic expansion valve is determined to be in a constant speed motion state; if the rate of change of the stepping pulse frequency within the detection window exceeds 10%, the electronic expansion valve is determined to be in an acceleration / deceleration motion state.

[0109] The controller reads the most recent valid position or the most recent reset zeroing position, and determines the current predicted position based on the cumulative number of signed instruction steps output since that position:

[0110] ;

[0111] In the formula, Indicates the current predicted position; Indicates the most recent valid position or the most recent reset zero-finding position; This indicates the number of step actions that have occurred since the most recent valid position or the most recent reset zero-finding position; Indicates the first The number of driving steps included in each step movement; Indicates the first The direction coefficient for each step action is 1 for the valve opening direction and -1 for the valve closing direction.

[0112] Establish a range of candidate locations centered on the current predicted location:

[0113] ;

[0114] In the formula, Indicates running status The corresponding range of candidate positions; Indicates the current predicted position; Indicates running status The corresponding maximum permissible positioning error half-width.

[0115] In one embodiment, the maximum permissible positioning error half-width corresponding to the stationary state is 3 steps, the maximum permissible positioning error half-width corresponding to the uniform motion state is 5 steps, and the maximum permissible positioning error half-width corresponding to the acceleration / deceleration motion state is 8 steps. When the nominal positioning error of the electronic expansion valve is greater than the above values, the maximum permissible positioning error half-width can be the sum of the nominal positioning error and the state-additional error, wherein the state-additional errors corresponding to the stationary state, uniform motion state, and acceleration / deceleration motion state are 1 step, 2 steps, and 4 steps, respectively.

[0116] The candidate position range is limited to the effective full stroke range of the electronic expansion valve, and the portion exceeding the full stroke endpoint is truncated. The candidate position range is continuously calculated from the most recent effective position and the number of signed command steps thereafter, thus preserving the motion history of the electronic expansion valve and providing constraints for disambiguation of subsequent cycle repeating positions.

[0117] The controller selects the first injection frequency from the calibration feature library. Second injection frequency The first and second injection frequencies are both in the range of 1kHz to 20kHz, and the frequency difference between them is no more than one-fifth of the lower injection frequency. They also avoid the PWM carrier frequency, integer multiples of the PWM carrier frequency, and corresponding sideband frequencies.

[0118] The injection frequency and the PWM carrier harmonics satisfy the following:

[0119] ;

[0120] In the formula, Indicates the first Injection frequency, Choose 1 or 2; Indicates the PWM carrier harmonic sequence number, which is a positive integer; Indicates the PWM carrier frequency; Indicates the spectrum protection interval.

[0121] The preferred spectral protection interval is 2kHz. When the bandwidth of the stepper drive circuit is limited, the spectral protection interval should be at least greater than the sum of the passband width of the dynamic bandpass filter and the sideband width of the PWM.

[0122] For each available frequency combination, determine the single-frequency position separability, the dual-frequency joint position separability, and the spectral security, and determine the frequency combination evaluation value according to the following formula:

[0123] ;

[0124] In the formula, Indicates the first injection frequency With the second injection frequency The frequency combination evaluation value; This represents the normalized value of the single-frequency position separability between the two injection frequencies; This represents the normalized value of the minimum neighboring feature distance corresponding to the joint feature of two injection frequencies; This represents the normalized value of the spectral security between the frequency combination and the PWM carrier, harmonics, and sidebands; 0.4, 0.4, and 0.2 are the weights of single-frequency position separability, dual-frequency joint position separability, and spectral security, respectively, and the sum of the three weights is 1.

[0125] The controller selects the frequency combination with the highest evaluation value as the current injection frequency combination. In one embodiment, the PWM carrier frequency is 24kHz, the first injection frequency is 4.8kHz, the second injection frequency is 5.6kHz, and the frequency difference between the two injection frequencies is 0.8kHz.

[0126] The first high-frequency AC voltage signal and the second high-frequency AC voltage signal can be injected sequentially or simultaneously. To reduce the intermodulation effect between the two frequencies, this embodiment adopts a sequential injection method, with each injection frequency lasting for no less than 8 complete signal cycles.

[0127] In a stationary state, a first high-frequency AC voltage signal is injected first, followed by a second high-frequency AC voltage signal. In uniform motion and acceleration / deceleration motion, the first and second high-frequency AC voltage signals can be injected alternately in adjacent detection windows, or the dual-frequency injection can be performed sequentially within the same detection window.

[0128] The injection amplitude should ensure that the generated high-frequency response current does not exceed 10% of the stepper motor's rated drive current. The initial injection voltage amplitude can be set to 3% of the rated drive voltage and adjusted based on the high-frequency response current obtained from the previous detection window.

[0129] ;

[0130] In the formula, This indicates the injection voltage amplitude for the next detection window; This indicates the injection voltage amplitude in the current detection window; Indicates the amplitude of the target high-frequency response current; This indicates the amplitude of the high-frequency response current measured in the current detection window; and These represent the minimum and maximum allowable injection voltage amplitudes, respectively. This indicates a range-limiting operation that restricts the calculation results to between the minimum and maximum values.

[0131] The target high-frequency response current amplitude is preferably 6% of the rated drive current, the minimum injection voltage amplitude is 1% of the rated drive voltage, and the maximum injection voltage amplitude is determined based on the condition that the high-frequency response current does not exceed 10% of the rated drive current.

[0132] (III) Step S2: Acquire the current response and determine the response confidence level

[0133] like Figure 2 As shown, during the injection of a high-frequency AC voltage signal, the current sampling circuit acquires the current response of the stepper motor windings. The sampling frequency is at least 10 times the higher injection frequency. In the embodiment where the first injection frequency is 4.8 kHz and the second injection frequency is 5.6 kHz, the sampling frequency is set to 80 kHz.

[0134] The sampling time is set synchronously with the PWM carrier and avoids the switching instant of the power switching devices. The controller acquires no fewer than 16 current sample values ​​during the no-high-frequency signal interval before each injection, and takes their average value as the current sampling zero point. The current sample values ​​during the injection period are subtracted from the current sampling zero point, and then converted into winding current based on the current sampling resistor value and the amplifier circuit gain.

[0135] The corrected current response enters a dynamic bandpass filter. A corresponding dynamic bandpass filter is set for each injection frequency, and the filter's center frequency is updated according to the current injection frequency. The dynamic bandpass filter can be a second-order infinite impulse response filter, whose transfer function is:

[0136] ;

[0137] In the formula, Indicates the first The dynamic bandpass filter transfer function corresponding to each injection frequency; Represents discrete-domain variables; Indicates the filter gain coefficient; Indicates the radius of the filter poles; Indicates the first The normalized angular frequency corresponding to each injection frequency; Choose 1 or 2.

[0138] The normalized angular frequency is determined by the following formula:

[0139] ;

[0140] In the formula, Indicates the first The normalized angular frequency of a dynamic bandpass filter; Indicates the first One injection frequency; Indicates the sampling frequency; It represents pi (π).

[0141] The filter gain coefficient is set according to the following formula:

[0142] ;

[0143] In the formula, Indicates the first Gain coefficients of a dynamic bandpass filter; This represents the pole radius of the corresponding dynamic bandpass filter.

[0144] In the stationary state, the pole radius is set to 0.96 to obtain a narrower passband; in the uniform motion state, the pole radius is set to 0.95; in the acceleration and deceleration motion state, the pole radius is set to 0.93 to appropriately increase the passband width and reduce the impact of the response frequency spread caused by speed changes on the filtering results.

[0145] The controller also utilizes a low-pass filter with a cutoff frequency no higher than 100Hz to extract low-frequency load disturbance components. The low-pass filter can be a second-order Butterworth digital filter, whose coefficients are determined by bilinear transformation based on the sampling frequency and the cutoff frequency.

[0146] In length Within the detection window, the total energy of the high-frequency response at the two injection frequencies is determined according to the following formula:

[0147] ;

[0148] In the formula, This represents the total energy of the high-frequency response at the two injection frequencies; Indicates the first The output of the first dynamic bandpass filter is the... One high-frequency current sampling value; Indicates the number of sampling points within the detection window; This indicates the injection frequency sequence number.

[0149] The low-frequency load disturbance energy is determined by the following formula:

[0150] ;

[0151] In the formula, This indicates the low-frequency load disturbance energy within the detection window; The output of the low-pass filter represents the first... One low-frequency current sample value; This indicates the number of sampling points within the detection window.

[0152] The confidence level of the response is determined by the following formula:

[0153] ;

[0154] In the formula, Indicates the confidence level of the response; This represents the total energy of the high-frequency response at the two injection frequencies; This indicates the energy of low-frequency load disturbance; This indicates a positive number that prevents the denominator from being zero.

[0155] With the current already normalized to the rated current... Pick The confidence threshold is set to 4 for a stationary state, 3 for a uniform motion state, and 2.5 for an accelerated / decelerated motion state.

[0156] In addition to response confidence, the controller also detects whether clipping has occurred in the analog-to-digital converter, whether the detection window contains a complete injection cycle, and whether the number of valid sampling points corresponding to the two injection frequencies meets the requirements. When the current sampling value reaches the upper or lower limit of the analog-to-digital converter range more than 3 times consecutively, the number of complete cycles corresponding to any injection frequency is less than 8, or the response confidence does not reach the corresponding confidence threshold, the current detection is marked as a low-confidence detection, and no new position update is performed.

[0157] The dynamic bandpass filter is set before synchronous demodulation, so that the low-frequency current component generated by the sudden change in refrigerant pressure difference is attenuated before entering the synchronous demodulation unit, thereby reducing the situation where strong low-frequency disturbances occupy the dynamic range of synchronous demodulation or cause distortion of demodulation results.

[0158] (iv) Step S3: Extract dual-frequency impedance characteristics and invert electrical position

[0159] Once the response confidence level reaches the corresponding confidence threshold, the controller synchronously demodulates the current response after dynamic bandpass filtering. Synchronous demodulation uses the current injected voltage signal as a reference signal to obtain the complex voltage and complex current corresponding to the first and second injection frequencies, respectively.

[0160] No. The complex current corresponding to each injection frequency is determined by the following formula:

[0161] ;

[0162] In the formula, Indicates the first The complex current corresponding to each injection frequency; Indicates the first The injection frequency corresponding to the first One filtered current sample value; Indicates the number of sampling points within the synchronous demodulation window; Indicates the first One injection frequency; Indicates the sampling frequency; Represents the imaginary unit; It represents pi (π).

[0163] No. The complex voltage corresponding to each injection frequency is determined by the following formula:

[0164] ;

[0165] In the formula, Indicates the first The complex voltage corresponding to each injection frequency; Indicates the first The injection frequency corresponding to the first One injection voltage sample value or injection voltage reconstructed value; Indicates the number of sampling points within the synchronous demodulation window; Indicates the first One injection frequency; Indicates the sampling frequency; Represents the imaginary unit; It represents pi (π).

[0166] When the stepper drive circuit does not have an independent voltage sampling channel, the injected voltage is reconstructed based on the DC bus voltage, PWM duty cycle, and power device on-state voltage drop.

[0167] No. The complex impedance corresponding to each injection frequency is determined by the following formula:

[0168] ;

[0169] In the formula, Indicates the first The complex impedance corresponding to each injection frequency; Indicates the first The complex voltage corresponding to each injection frequency; Indicates the first The complex current corresponding to each injection frequency; Choose 1 or 2.

[0170] The equivalent resistance is determined based on the complex impedance corresponding to the two injection frequencies:

[0171] ;

[0172] In the formula, This represents the equivalent resistance of the stepper motor windings; and These represent the complex impedances corresponding to the first injection frequency and the second injection frequency, respectively. This indicates taking the real part of a complex number.

[0173] Position-dependent inductance is determined by the following formula:

[0174] ;

[0175] In the formula, This represents the position-dependent inductance after dual-frequency fusion; and These represent the complex impedances corresponding to the first injection frequency and the second injection frequency, respectively. This indicates taking the imaginary part of a complex number; and These represent the first injection frequency and the second injection frequency, respectively. It represents pi (π).

[0176] The position-dependent inductance and the impedance phases corresponding to the two injection frequencies are used to form the position eigenvector:

[0177] ;

[0178] In the formula, This represents the position feature vector of the current detection window; This represents the position-dependent inductance after dual-frequency fusion; and These represent the phase angles of the first complex impedance and the second complex impedance, respectively.

[0179] The controller queries valve opening direction calibration data or valve closing direction calibration data based on the current direction of movement. When a temperature sensor is configured, it interpolates between adjacent temperature calibration tables based on the current temperature; when no temperature sensor is configured, it uses equivalent resistance... The winding temperature is estimated using the pre-calibrated resistance-temperature relationship.

[0180] For the current position feature vector In the corresponding calibration data, find the two adjacent calibration points with the smallest feature distance, and use an inverse distance weighting method to determine the electrical position within one magnetic pole cycle:

[0181] ;

[0182] In the formula, This indicates the electrical position within one magnetic pole cycle obtained through inversion; and Indicates the electrical position corresponding to two adjacent calibration points; and These represent the normalized distances between the current position feature vector and the feature vectors of two adjacent calibration points, respectively. This represents a positive number that prevents the feature distance from being zero.

[0183] When the calibration features have been normalized Pick When the normalized distance between the current location feature vector and a certain calibration point is less than... When the calibration point is used, the electrical position corresponding to that calibration point is used directly.

[0184] (v) Step S4: Generate candidate positions and perform periodic position disambiguation and mechanical idle compensation.

[0185] like Figure 5 As shown, the electrical position obtained in step S3 only represents the relative position of the rotor within one magnetic pole cycle. To obtain a unique position throughout the entire stroke of the electronic expansion valve, the controller generates multiple candidate positions based on the electrical position, the number of magnetic pole pairs, the number of drive steps corresponding to each mechanical revolution, and the candidate position range.

[0186] No. The candidate step position corresponding to each electrical cycle is determined by the following formula:

[0187] ;

[0188] In the formula, Indicates the first Candidate step positions corresponding to each electrical cycle; This indicates the number of controller drive steps corresponding to each mechanical revolution of the stepper motor rotor; This indicates the number of pole pairs in a stepper motor; Indicates the electrical position within one magnetic pole cycle; Indicates the electrical cycle number, rounded to the nearest integer; This indicates the step offset corresponding to the fully closed zero position; It represents pi (π).

[0189] When using segmentation-driven... Determined based on the equivalent number of driving steps after subdivision. Number of pole pairs. The parameters are determined by the specifications of the electronic expansion valve stepper motor or the results of offline calibration, and are pre-stored in the memory.

[0190] Generate a set of candidate positions within the effective full stroke of the electronic expansion valve:

[0191] ;

[0192] In the formula, This represents the set of candidate locations for the entire journey; Indicates the first Candidate step positions corresponding to each electrical cycle; This indicates the maximum effective stroke number of the electronic expansion valve.

[0193] Intersect the set of candidate locations for the entire journey with the range of candidate locations established in step S1:

[0194] ;

[0195] In the formula, This represents the set of valid candidate positions that fall within the range of candidate positions. This represents the set of candidate locations for the entire journey; This indicates the range of candidate locations corresponding to the current running state.

[0196] When the set of valid candidate positions contains only one candidate position, that candidate position is determined as the unique rotor step position. When the set of valid candidate positions is empty, it indicates that the current electrical position is inconsistent with the motion history, and the detection result is marked as invalid. When the set of valid candidate positions contains more than two candidate positions, it indicates that the current candidate position range is insufficient to exclude periodically repeating positions, and the detection result is also marked as invalid.

[0197] To increase the probability of obtaining a unique candidate position, the total width of the candidate position range is preferably smaller than the distance between two adjacent candidate positions in a cycle. When it is necessary to expand the candidate position range during acceleration or deceleration, the detection frequency can be increased to reduce the cumulative instruction steps between the most recent valid position and the current detection time.

[0198] Obtain the unique rotor step position Then, mechanical backlash compensation is performed based on the current direction of motion. The step count increases in the valve opening direction, and the actual mechanical position of the valve needle during the valve opening process is determined by the following formula:

[0199] ;

[0200] In the formula, This indicates the actual mechanical position of the valve needle after compensation; Indicates the unique rotor step position; This indicates the mechanical backlash compensation value for the valve opening direction.

[0201] The actual mechanical position of the valve needle during valve closing is determined by the following formula:

[0202] ;

[0203] In the formula, This indicates the actual mechanical position of the valve needle after compensation; Indicates the unique rotor step position; This indicates the mechanical backlash compensation value for the valve closing direction.

[0204] Mechanical backlash compensation value for valve opening direction Mechanical backlash compensation value for valve directional movement The values ​​are obtained through the calibration process, typically taking 1 to 3 steps. When the mechanical travel distances in the two directions of motion are different, the calibration values ​​for the corresponding directions are used respectively.

[0205] When the direction of motion changes, the controller establishes a residual idle distance. The initial value of the residual idle distance is the mechanical idle distance compensation value corresponding to the new direction of motion. The residual idle distance is decremented by 1 for each drive step output in the new direction. While the residual idle distance is greater than zero, the actual mechanical position of the valve needle remains the last valid mechanical position before the switching; after the residual idle distance is reduced to zero, the actual mechanical position of the valve needle is updated according to the unique rotor step position.

[0206] pass Figure 5 The process of generating periodic candidate positions, constraining the range of candidate positions, determining uniqueness, and compensating for direction-related mechanical backlash, as shown, converts the periodically repeating electrical positions into the actual mechanical positions of the valve needle throughout the full stroke of the electronic expansion valve.

[0207] (vi) Step S5: Loss of synchronization judgment and fault classification handling

[0208] like Figure 6 As shown, the controller compares the actual mechanical position of the valve needle after mechanical backlash compensation with the current commanded position to determine the position deviation:

[0209] ;

[0210] In the formula, This indicates the absolute deviation in steps between the actual mechanical position and the commanded position of the valve needle. Indicates the actual mechanical position of the valve needle; This indicates the current recorded instruction position of the controller.

[0211] The step loss detection threshold is set to 1 step when the device is stationary. The step loss detection threshold when the device is moving is determined according to the following formula:

[0212] ;

[0213] In the formula, This represents the threshold for determining step loss during motion, expressed in steps. This indicates the nominal positioning error of the electronic expansion valve at the current operating speed or acceleration, expressed in steps. This means taking the larger of the two values.

[0214] When the electronic expansion valve is in an acceleration / deceleration state, and the current acceleration reaches more than 80% of the rated acceleration, a one-step transient margin can be added to the motion state step loss judgment threshold. This transient margin is cancelled upon entering a uniform motion state or a stationary state.

[0215] When the response confidence reaches the corresponding confidence threshold, the candidate position is unique, and the position deviation is less than the corresponding out-of-step judgment threshold, the electronic expansion valve is determined to be in normal position, and the current actual mechanical position is updated to the most recent valid position.

[0216] When the response confidence level reaches the corresponding confidence threshold, the candidate position is unique, and the position deviation reaches or exceeds the corresponding out-of-step judgment threshold, the electronic expansion valve is determined to have lost synchronization. When the position deviation is less than 8 steps, a formal out-of-step flag can be output after two consecutive detections meet the out-of-step conditions; when the position deviation reaches or exceeds 8 steps, a severe out-of-step flag is output directly.

[0217] When the response confidence level does not reach the corresponding confidence threshold, the candidate location set is empty, there is more than one candidate location, or clipping occurs during current sampling, the current detection result is marked as an invalid location. Invalid locations are not used to update the most recent valid location, nor is a normal location conclusion output. The controller retains the most recent valid location and re-detects after a preset waiting time.

[0218] The re-detection waiting time in a stationary state can be 20ms; in a uniform motion state, it can wait for one step drive cycle before re-detection; in an acceleration / deceleration motion state, it can re-detection after the acceleration decreases to below 80% of the rated acceleration.

[0219] If three consecutive detections fail, the controller performs a low-speed position re-detection; if five consecutive detections fail, or if the cumulative number of instruction steps since the most recent valid position exceeds one electrical cycle, the controller triggers a reset and zero-finding.

[0220] Fault handling is divided into normal state, mild loss of step state, and severe loss of step state.

[0221] Under normal conditions, if the position deviation is less than the out-of-step judgment threshold, the controller will maintain normal drive and determine the timing of the next detection based on the operating status.

[0222] In a state of mild step loss, if the position deviation reaches the step loss judgment threshold but is less than 8 steps, the controller records the step loss occurrence time, command position, actual mechanical position, position deviation, running direction, response confidence level, and the injection frequency used, and outputs a warning message to the main controller of the refrigeration equipment. When position compensation is allowed, the compensation steps corresponding to the position deviation are added to subsequent drive commands. The number of compensation steps in a single operation is preferably no more than 4 steps; if the position deviation is greater than 4 steps, compensation is completed in two or more control cycles.

[0223] In a severe out-of-step state, if the position deviation reaches or exceeds 8 steps, three consecutive detections are all considered out of step, or five consecutive detections are invalid, the controller stops the current drive output and performs a reset and zero-finding. The reset and zero-finding establishes a fully closed zero position by closing the valve at a low speed to the mechanical limit position, and the reset speed does not exceed 30% of the normal maximum adjustment speed.

[0224] After reset, the high-frequency signal injection detection is re-executed. If a valid position still cannot be obtained, or if the deviation between the actual mechanical position and the fully closed position after reset still reaches the out-of-step judgment threshold, a serious fault signal is output. The refrigeration equipment main controller restricts or stops the compressor operation according to the safety policy.

[0225] IV. Detection Timing and Power Consumption Control

[0226] This invention employs a combination of periodic detection and event-triggered detection.

[0227] During normal adjustment of the electronic expansion valve, the controller performs detection at intervals of no more than 1 second. Event trigger detection is executed when the electronic expansion valve completes a step action, when the refrigeration system experiences abnormal superheat, when the compressor starts or stops, when the operating mode is switched, or when the direction of movement of the electronic expansion valve changes.

[0228] After the system is powered on and initialized, and before the compressor starts, a position self-test is performed. If a unique position cannot be obtained, a reset and zero-finding process is performed first, and then an initial valid position is established.

[0229] To reduce the additional power consumption caused by high-frequency signal injection, an intermittent injection method is adopted. The total duration of dual-frequency injection for each detection is preferably no more than 10ms. When the valve is stationary and continuous detection is normal, the detection interval is extended to 5 to 10 seconds; when overheating abnormalities occur, movement direction changes, acceleration or deceleration occurs, or low-confidence results occur, the detection interval is shortened.

[0230] V. Specific Implementation Examples and Comparative Examples

[0231] (a) Test apparatus and evaluation methods

[0232] The test setup includes an electronic expansion valve, a stepper drive circuit, a main controller, a current sampling circuit, a DC regulated power supply, a high and low temperature test chamber, a refrigerant differential pressure simulation circuit, an external position measuring device, an oscilloscope, and data acquisition equipment.

[0233] The tested electronic expansion valve has a rated drive voltage of 12V, a rated drive current of 0.35A, and a maximum effective stroke of 480 steps. The stepper drive circuit has a PWM carrier frequency of 24kHz, a current sampling resistor of 0.2Ω, an analog-to-digital conversion resolution of 12 bits, and a sampling frequency of 80kHz. The first injection frequency is 4.8kHz, the second injection frequency is 5.6kHz, and the duration of each injection frequency is 4ms.

[0234] The external position measuring device uses a laser displacement sensor with a resolution of 1μm. Before the experiment, the electronic expansion valve was controlled to move gradually throughout its full stroke to establish the correspondence between the valve needle displacement and the number of drive steps. The valve needle position obtained by the external position measuring device was used as the reference actual position.

[0235] The refrigerant differential pressure simulation loop includes a high-pressure receiver tank, a low-pressure receiver tank, a pressure regulating valve, a pressure sensor, and a bypass valve, used to create pressure differentials of 0.2 MPa, 0.8 MPa, and 1.6 MPa across the electronic expansion valve. Low-frequency load surges are generated by rapidly switching the bypass valve to simulate disturbances caused by compressor start-up and shutdown, changes in refrigerant state, and sudden changes in valve flow.

[0236] The test temperatures of the high and low temperature test chamber are set to -20℃, 0℃, 25℃, 50℃, and 70℃, respectively. Each temperature point is held for no less than 30 minutes to allow the temperature of the electronic expansion valve winding to stabilize basically with the test chamber temperature before starting the test.

[0237] The operating states of the electronic expansion valve include:

[0238] In a stationary state, stop and detect at positions of 40, 120, 240, 360, and 440 steps respectively;

[0239] In a state of uniform motion, the motion reciprocates within a range of 40 to 440 steps at a speed of 60 steps per second.

[0240] In the acceleration and deceleration motion state, the speed increases from 0 steps per second to 120 steps per second within 0.5 seconds, and then decreases from 120 steps per second to 0 steps per second within 0.5 seconds.

[0241] Two methods are used to inject out-of-step faults. The first method involves blocking a predetermined number of winding commutation pulses through a fault injection switch while the stepper drive command count remains normal, resulting in position deviations of 1, 2, 4, and 8 steps, respectively. The second method involves increasing the rotor load through an adjustable mechanical braking mechanism during valve needle movement, causing the rotor to fail to complete all command steps. This is used to verify the invention's ability to detect out-of-step faults under actual mechanical overload.

[0242] The mean absolute error of position is determined by the following formula:

[0243] ;

[0244] In the formula, Indicates the average absolute error of the position; Indicates the number of valid tests; Indicates the first The actual mechanical position of the valve needle obtained from the second test; Indicates the first The reference actual position determined by an external position measuring device during the next test.

[0245] The maximum position error is determined by the following formula:

[0246] ;

[0247] In the formula, This represents the maximum positional error among all valid detection results; Indicates the first The actual mechanical position of the valve needle obtained from the second test; Indicates the first The reference actual position corresponding to this detection; Indicates the number of valid tests.

[0248] The out-of-step detection rate is determined by the following formula:

[0249] ;

[0250] In the formula, Indicates the out-of-step detection rate; Indicates the number of tests that correctly detected the out-of-step. This represents the total number of actual injection synchronization failures.

[0251] The false alarm rate is determined by the following formula:

[0252] ;

[0253] In the formula, Indicates the false alarm rate; This indicates the number of times the system was incorrectly identified as having lost step when no step loss fault was injected. This indicates the total number of times a step-out fault was detected without injection.

[0254] The effective position output rate is determined by the following formula:

[0255] ;

[0256] In the formula, Indicates the effective position output rate; This represents the number of detections that achieved a response confidence threshold and yielded a unique candidate location. This indicates the total number of location checks performed.

[0257] The detection response time refers to the time from the actual occurrence of the out-of-step fault to the output of the out-of-step judgment result by the controller.

[0258] (II) Example 1: Position detection under different temperatures and motion conditions

[0259] Establish a calibration feature library for the valve opening direction and a calibration feature library for the valve closing direction according to the aforementioned calibration method. The calibration of the mechanical idle distance in the valve opening direction is 2 steps, and the calibration of the mechanical idle distance in the valve closing direction is 3 steps.

[0260] Position detection was performed at -20℃, 25℃, and 70℃. 100 detections were conducted for each temperature and motion state combination, and the mean absolute position error, maximum position error, and effective position output rate were statistically analyzed. The results are shown in Table 1.

[0261] Table 1. Position detection results under different temperatures and motion conditions.

[0262]

[0263] As shown in Table 1, under static conditions, the average absolute position error of the present invention is no greater than 0.35 steps; under uniform motion conditions, the average absolute position error is no greater than 0.53 steps; and under acceleration / deceleration motion conditions, the average absolute position error is no greater than 0.79 steps. When the temperature changes from -20℃ to 70℃, no significant increase in position error with changes in winding resistance is observed.

[0264] The maximum position error is higher during acceleration / deceleration than during stationary motion because the stepping pulse frequency and refrigerant load change simultaneously within the detection window. By increasing the passband width of the dynamic bandpass filter and using a larger candidate position range during acceleration / deceleration, an effective position output rate of over 94% can still be maintained.

[0265] Based on the data in Table 1, the following can be added to the instruction manual: Figure 7 . Figure 7The average absolute position error under different temperatures and motion conditions is represented by a grouped bar chart to illustrate that the present invention can perform position detection under static, uniform motion, and acceleration / deceleration motion conditions.

[0266] (III) Example 2: Anti-interference test under low-frequency load disturbance

[0267] The electronic expansion valve was set to operate at 25℃, and was subjected to uniform motion and acceleration / deceleration motion under pressure differentials of 0.2MPa, 0.8MPa, and 1.6MPa, respectively. During the test, the bypass valve was rapidly switched to gradually increase the peak value of the low-frequency disturbance current.

[0268] A dynamic bandpass filter was used to process the current response before synchronous demodulation, and the response confidence level was determined based on the total energy of the high-frequency response and the energy of the low-frequency load disturbance. Each combination of pressure difference and motion state was tested 200 times. The results are shown in Table 2.

[0269] Table 2. Anti-interference test results under different pressure differences

[0270]

[0271] As shown in Table 2, as the pressure difference across the electronic expansion valve increases, the low-frequency load disturbance energy increases, and the response confidence level decreases accordingly. Under the conditions of a pressure difference of 1.6 MPa and acceleration / deceleration, the average response confidence level is still higher than the confidence threshold of 2.5 corresponding to the acceleration / deceleration state, and the effective position output rate is 94.0%.

[0272] For detection windows with a response confidence level below the threshold, the controller does not output a new actual mechanical position, but instead retains the most recent valid position and triggers a re-detection. Therefore, low signal-to-noise ratio data is not directly used for out-of-step judgment, which helps reduce the false alarm rate under strong load disturbances.

[0273] (iv) Example 3: Temperature adaptability test of dual-frequency characteristics

[0274] To verify the effect of dual-frequency impedance characteristics on suppressing winding resistance temperature drift, static position detection was performed at -20℃, 0℃, 25℃, 50℃, and 70℃. For each temperature point, five positions were selected: 40 steps, 120 steps, 240 steps, 360 steps, and 440 steps. Each position was repeatedly detected 40 times, resulting in 200 sets of detection data for each temperature point.

[0275] In this embodiment, a first injection frequency of 4.8 kHz and a second injection frequency of 5.6 kHz were used. The position-dependent inductance and the phase of the two impedances were determined using the complex impedances corresponding to the two injection frequencies, and a multi-temperature, multi-directional calibration feature library was consulted. The results are shown in Table 3.

[0276] Table 3. Location detection results of dual-frequency features at different temperatures.

[0277]

[0278] As shown in Table 3, the equivalent resistance of the winding increases with increasing temperature. However, after using dual-frequency impedance characteristics and multi-temperature calibration data, the average absolute error of the position at each temperature point is less than 0.5 steps. This result indicates that by using the position-dependent inductance and the two impedance phases to form a joint position characteristic, the position deviation caused by the temperature drift of the winding resistance in the single response amplitude can be reduced.

[0279] (V) Example 4: Periodic position disambiguation and mechanical backlash compensation test

[0280] Within the maximum effective stroke range of 480 steps, the electronic expansion valve is controlled to continuously complete 100 reciprocating movements from fully open to fully closed and then from fully closed to fully open. During each movement, detection is performed at 20 randomly selected positions, resulting in a total of 2000 sets of position detection data.

[0281] The controller establishes a candidate position range based on the most recent valid position and the cumulative number of command steps. It then generates full-stroke candidate positions based on the electrical position, the number of magnetic pole pairs, and the number of drive steps corresponding to each mechanical revolution, selecting the unique candidate position that falls within the candidate position range. When switching motion directions, a 2-step mechanical backlash compensation value for the valve opening direction and a 3-step mechanical backlash compensation value for the valve closing direction are used, respectively. The results are shown in Table 4.

[0282] Table 4 Results of Periodic Position Disambiguation and Mechanical Idle Stroke Compensation

[0283]

[0284] Periodic mismatch refers to a situation where the electrical position obtained through inversion is basically correct, but the mechanical position throughout the entire stroke is incorrectly assigned to other magnetic pole periods. For detection windows with multiple candidate positions or empty candidate positions, this invention marks them as invalid positions instead of forcing the output position. Therefore, the number of periodic mismatches in Table 4 is relatively small.

[0285] The average position error during the reversing phase is higher than that during the stable motion phase, but by using mechanical backlash compensation values ​​in the valve opening direction and valve closing direction respectively, the average error in the reversing zone is still less than 1 step.

[0286] (vi) Example 5: Detection and grading test under different degrees of step loss

[0287] Under conditions of 25℃ and 0.8MPa pressure difference, 1-step, 2-step, 4-step, and 8-step step loss faults were injected into the stationary, uniform motion, and acceleration / deceleration motion states, respectively. Each combination of step loss number and motion state was repeated 50 times.

[0288] The threshold for step loss in a stationary state is 1 step; the threshold for step loss in uniform motion is 2 steps; and the threshold for step loss in acceleration / deceleration motion is 3 steps. If the position deviation is less than 8 steps, a step loss flag is output after two consecutive detections meet the step loss condition; if the position deviation reaches 8 steps, a severe step loss flag is output directly. The results are shown in Table 5.

[0289] Table 5 Detection results under different degrees of step loss

[0290]

[0291] In Table 5, the detection rate of 1-step loss of step is low in uniform motion and acceleration / deceleration states because the deviation does not reach the loss of step judgment threshold for the corresponding motion state. This is an intentional transient tolerance set by the controller, rather than a missed detection. After the electronic expansion valve enters a stationary state, the loss of step judgment threshold in the stationary state returns to 1 step, which can detect the residual 1-step position deviation.

[0292] For 2-step and 4-step step loss, the controller records the step loss event and allocates the compensation steps to subsequent control cycles. For 8-step step loss, the controller directly stops the current drive and performs a reset and zero-finding. This result demonstrates that the present invention can not only determine whether a step loss has occurred, but also implement graded processing based on the number of position deviation steps.

[0293] (vii) Comparative Example 1: Cancellation of dynamic bandpass filtering and response confidence determination

[0294] Comparative Example 1 uses the same hardware, injection frequency, and sampling frequency as Example 2, but the acquired current response is not subjected to dynamic bandpass filtering and is directly demodulated synchronously; at the same time, the response confidence is not calculated, and all detection results are used for position update and out-of-step judgment.

[0295] Acceleration and deceleration motion tests were performed under pressure differences of 0.8 MPa and 1.6 MPa, respectively. For each pressure difference, 200 normal tests and 100 two-step out-of-step tests were performed. The results are shown in Table 6.

[0296] Table 6. Impact of dynamic filtering and response confidence on detection results

[0297]

[0298] As shown in Table 6, after canceling the dynamic bandpass filtering and response confidence determination, low-frequency load disturbances directly enter the synchronous demodulation process, leading to a significant increase in position error and false alarm rate. Under a pressure difference of 1.6 MPa, the false alarm rate of Comparative Example 1 is 15.5%, higher than the 2.0% of this invention. This comparison demonstrates that performing dynamic bandpass filtering before synchronous demodulation and response confidence determination before position inversion is beneficial for improving detection reliability under strong load disturbance conditions.

[0299] (viii) Comparative Example 2: Injection of a single-frequency high-frequency signal

[0300] Comparative Example 2 uses only a single injection frequency of 4.8kHz. Based on the response amplitude and phase at this frequency, the position calibration table is consulted. A second injection frequency is not used, and the position-dependent inductance after dual-frequency fusion is not determined.

[0301] 200 static position tests were performed at -20℃, 25℃, and 70℃ respectively, with other test conditions the same as in Example 3. The results are shown in Table 7.

[0302] Table 7. Comparison of temperature adaptability between dual-frequency injection and single-frequency injection.

[0303]

[0304] Comparative Example 2 still achieves a certain degree of position discrimination capability near 25℃, but at -20℃ and 70℃, the single-frequency response amplitude is affected by changes in winding resistance, resulting in a significant increase in position error. Example 3 uses the complex impedances corresponding to the two injection frequencies to jointly determine the position-related inductance and impedance phase, which can reduce the influence of single-frequency characteristic drift with temperature changes.

[0305] (ix) Comparative Example 3: Cancellation of candidate position disambiguation

[0306] Comparative Example 3 directly queries the full-stroke mechanical position based on the electrical position within a magnetic pole cycle, without establishing a candidate position range based on the most recent valid position and the cumulative number of instruction steps, and without performing a uniqueness judgment on multiple cycle candidate positions.

[0307] The same reciprocating motion test as in Example 4 was performed over the entire 480-step range. The results are shown in Table 8.

[0308] Table 8. Impact of Candidate Position Disambiguation on Full-Stroke Position Detection

[0309]

[0310] As shown in Table 8, after removing the candidate position range constraint and uniqueness judgment, the same or similar electrical features were assigned to the wrong magnetic pole cycle, resulting in position jumps throughout the entire stroke. The maximum position error in Comparative Example 3 reached 96 steps, which is significantly greater than that in Example 4. This result indicates that obtaining the electrical position within only one magnetic pole cycle is insufficient to determine the full-stroke mechanical position of the electronic expansion valve; it is necessary to combine the motion history to establish the candidate position range and perform periodic position disambiguation.

[0311] (x) Comparative Example 4: Cancellation of Direction-Related Mechanical Idle Travel Compensation

[0312] Comparative Example 4 retains dual-frequency injection, dynamic bandpass filtering, response confidence determination, and candidate position disambiguation, but does not distinguish between the mechanical idle stroke in the valve direction and the mechanical idle stroke in the valve closing direction, and directly uses the unique rotor step position as the actual mechanical position of the valve needle.

[0313] The system underwent 100 reciprocating cycles at 25℃ and a pressure difference of 0.8MPa, with five positions continuously monitored after each cycle. The results are shown in Table 9.

[0314] Table 9. Impact of Mechanical Backlash Compensation on Reversing Position Detection

[0315]

[0316] In Comparative Example 4, after the rotor begins to move in the reverse direction, the controller immediately changes the mechanical position of the valve needle. However, the actual valve needle only begins to move axially after the transmission backlash is eliminated, thus generating a false position report in the reversing zone. Example 4 uses mechanical backlash compensation values ​​for the valve opening direction and the valve closing direction, respectively, reducing the average error in the reversing zone from 2.26 steps to 0.78 steps.

[0317] (xi) Comparison of overall performance

[0318] This invention is compared with indirect detection methods based on thermodynamic parameters of refrigeration systems, stall detection methods based on the root mean square value of winding current, and single-frequency high-frequency injection methods.

[0319] The indirect thermodynamic parameter detection method judges electronic expansion valve malfunctions based on changes in evaporator outlet superheat and suction pressure; the winding current root mean square value detection method judges stall based on the degree of increase in drive current; the single-frequency high-frequency injection method uses a single injection frequency but does not implement dual-frequency temperature compensation and periodic position disambiguation. Each method was subjected to 300 normal tests and fault tests with different degrees of step loss. The results are shown in Table 10.

[0320] Table 10 Comparison of the overall performance of different out-of-step detection methods

[0321]

[0322] As shown in Table 10, indirect detection of thermodynamic parameters requires waiting for the effects of step loss on superheat and pressure, resulting in a long average response time and a low detection rate for minor step loss. Root mean square current detection can identify severe stall, but it is insensitive to minor step loss ranging from 1 to 4 steps. The single-frequency high-frequency injection method improves the response speed, but its detection results are still easily affected by winding temperature and cycle position repetition.

[0323] This invention, through dual-frequency high-frequency signal injection, dynamic bandpass pre-filtering, response confidence control, electrical position period disambiguation, and direction-related mechanical backlash compensation, demonstrates a high rate of detection of minor out-of-step events, a low false alarm rate, and a short response time in experiments.

[0324] Figure 8 The two-step out-of-step detection rate, false alarm rate, and average response time of different detection methods are presented to visually illustrate the technical effects of the present invention relative to each comparative example.

[0325] The results of the above embodiments and comparative examples show that:

[0326] First, the dual-frequency high-frequency signal injection and dual-frequency impedance characteristics can reduce the influence of the winding resistance change with temperature on the position inversion results, so that the electronic expansion valve can maintain stable position detection accuracy over a wide temperature range.

[0327] Second, by setting a dynamic bandpass filter whose center frequency follows the injection frequency before synchronous demodulation, and determining the response confidence based on the high-frequency response energy and low-frequency disturbance energy, the strong low-frequency disturbance caused by the sudden change in refrigerant pressure difference can be suppressed, and false alarms caused by low-confidence detection results participating in the out-of-synchronization judgment can be reduced.

[0328] Third, by establishing a range of candidate positions based on the most recent valid position, the cumulative number of command steps, and the direction of movement, and determining a unique position from multiple candidate positions in a cycle, the problem that the electrical position within a magnetic pole cycle cannot directly correspond to the mechanical position throughout the entire stroke can be solved.

[0329] Fourth, by using mechanical backlash compensation values ​​for the valve opening direction and the valve closing direction respectively, the false position reports caused by the asynchronous position of the rotor and the valve needle during the reversing process of the electronic expansion valve can be reduced.

[0330] Fifth, it adopts dynamic step loss judgment thresholds corresponding to static and dynamic states, and performs recording, early warning, compensation, reset and zero-finding or shutdown protection according to the degree of position deviation, which can balance the sensitivity of slight step loss detection and the ability to resist false alarms in dynamic states.

[0331] The foregoing description of embodiments of the present invention, through which those skilled in the art are able to implement or use the present invention, will be readily apparent to those skilled in the art. Various modifications to these embodiments will be readily apparent to those skilled in the art. The general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novelty disclosed herein.

[0332] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0333] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0334] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0335] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0336] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0337] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0338] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

Claims

1. A method for detecting full-cycle step loss in an electronic expansion valve based on high-frequency signal injection, characterized in that, include: S1. When the electronic expansion valve is in a stationary, uniform speed or acceleration / deceleration state, a candidate position range is established based on the most recent effective position or reset zeroing position, the cumulative command steps and the direction of movement. The first injection frequency and the second injection frequency are selected from the calibration feature library, and the corresponding high-frequency AC voltage signal is injected into the stepper motor winding. S2. Acquire the winding current response, pre-filter it using a dynamic bandpass filter whose center frequency follows the corresponding injection frequency, and determine the response confidence level based on the high-frequency response energy and the low-frequency disturbance energy. S3. When the response confidence reaches the preset threshold, the pre-filtered current response is synchronously demodulated to determine the dual-frequency complex impedance, equivalent resistance, position-dependent inductance and impedance phase, and the electrical position within a magnetic pole cycle is obtained according to the calibration feature library. S4. Generate full-stroke candidate positions based on the electrical position and the number of pole pairs of the stepper motor, determine the unique candidate position that falls within the candidate position range, and perform mechanical backlash compensation according to the direction of motion to obtain the actual mechanical position of the valve needle. If a candidate position is not unique, it is marked as an invalid position. S5. Compare the actual mechanical position of the valve needle with the commanded position, and determine the step loss result based on the step loss judgment threshold corresponding to the operating state; if the response confidence is insufficient or the position is invalid, maintain the most recent valid position and trigger re-detection or reset zeroing.

2. The method for detecting out-of-step operation of an electronic expansion valve throughout its entire lifecycle according to claim 1, characterized in that, The calibration feature library is established as follows: the electronic expansion valve is controlled to traverse its entire stroke along both the opening and closing directions. Under multiple calibration temperatures, the actual mechanical position of the valve needle is determined using an external position measuring device. High-frequency AC voltage signals of multiple candidate frequencies are injected at each calibration position, and the corresponding complex voltage, complex current, position-related inductance, and impedance phase are recorded. Calibration points are densified in intervals where the characteristic distance between adjacent positions is less than 60% of the average characteristic distance between adjacent positions throughout the entire stroke. Calibration data for the opening direction, calibration data for the closing direction, mechanical travel in the opening direction, and mechanical travel in the closing direction are stored respectively. Preferably, the first injection frequency and the second injection frequency are both between 1kHz and 20kHz, and the frequency difference between them is no more than one-fifth of the lower injection frequency; the first injection frequency and the second injection frequency are selected based on the minimum adjacent feature distance of the entire stroke, and avoid the PWM carrier frequency, integer multiples of the PWM carrier frequency and the corresponding sideband frequency; the high-frequency response current generated by the high-frequency AC voltage signal does not exceed 10% of the rated drive current of the stepper motor.

3. The method for detecting out-of-step operation of an electronic expansion valve throughout its entire lifecycle according to claim 1, characterized in that, In step S2, the response confidence level is determined according to the following formula: ; In the formula, Indicates the confidence level of the response; This represents the total energy of the high-frequency response at the first and second injection frequencies. This indicates low-frequency disturbance energy with a frequency below 100Hz; This indicates a positive number that prevents the denominator from being zero; The sampling frequency of the winding current response is not less than 10 times the second injection frequency; when the response confidence is lower than the preset threshold corresponding to the current operating state, or when current sampling is clipped, the current detection is marked as a low confidence detection.

4. The method for detecting out-of-step operation of an electronic expansion valve throughout its entire lifecycle according to claim 3, characterized in that, In step S3, the first The complex impedance corresponding to each injection frequency is determined by the following formula: ; In the formula, Indicates the first The complex impedance corresponding to each injection frequency; Indicates the first The complex voltage corresponding to each injection frequency; Indicates the first The complex current corresponding to each injection frequency; The injection frequency number is either 1 or 2; The equivalent resistance is determined according to the following formula: ; In the formula, This represents the equivalent resistance of the stepper motor windings; and These represent the complex impedances corresponding to the first injection frequency and the second injection frequency, respectively. This indicates taking the real part of a complex number; The location-dependent inductance is determined according to the following formula: ; In the formula, This represents the position-dependent inductance after dual-frequency fusion; This indicates taking the imaginary part of a complex number; and These represent the first injection frequency and the second injection frequency, respectively. Represents pi; Preferably, in step S3, the position-related inductance and the phases of the first complex impedance and the second complex impedance are used to form a position feature vector: ; In the formula, Represents the location feature vector; This represents the position-dependent inductance after dual-frequency fusion; and These represent the phase angles of the first complex impedance and the second complex impedance, respectively. Select the valve opening direction calibration data or valve closing direction calibration data according to the current direction of motion, and query or interpolate the corresponding calibration data to obtain the electrical position within a magnetic pole cycle based on the current temperature or the winding temperature determined by the equivalent resistance.

5. The method for detecting out-of-step operation of an electronic expansion valve throughout its entire lifecycle according to claim 1, characterized in that, In step S4, the first The candidate positions corresponding to each electrical cycle are determined according to the following formula: ; In the formula, Indicates the first Candidate step positions corresponding to each electrical cycle; This indicates the number of drive steps corresponding to each mechanical revolution of the stepper motor rotor; This indicates the number of pole pairs in a stepper motor. Indicates the electrical position within one magnetic pole cycle; Indicates the electrical cycle number, rounded to the nearest integer; This indicates the step offset corresponding to the fully closed zero position; Represents pi; Retain candidate positions within the effective full stroke of the electronic expansion valve, and determine the only candidate position that falls within the candidate position range as the rotor step position; if there is no candidate position or two or more candidate positions fall within the candidate position range, mark this detection as an invalid position; Preferably, with the valve opening direction as the direction of step increase, the mechanical backlash compensation in step S4 includes: During valve opening, the actual mechanical position of the valve needle is determined according to the following formula: ; In the formula, Indicates the actual mechanical position of the valve needle; This indicates a uniquely determined rotor step position; Indicates the mechanical backlash compensation value for the valve opening direction; During valve closing, the actual mechanical position of the valve needle is determined according to the following formula: ; In the formula, Indicates the actual mechanical position of the valve needle; This indicates a uniquely determined rotor step position; This indicates the mechanical backlash compensation value for the valve closing direction; After the direction of motion changes, the most recent valid mechanical position before the change is maintained until the cumulative number of drive steps in the new direction of motion reaches the corresponding mechanical idle compensation value.

6. The method for detecting out-of-step operation of an electronic expansion valve throughout its entire lifecycle according to claim 1, characterized in that, In step S5, the step loss judgment threshold corresponding to the stationary state is 1 step, and the step loss judgment threshold corresponding to the moving state is the larger of the value of 2 steps and the nominal positioning error increased by 1 step. When the position deviation reaches the out-of-step judgment threshold but is less than 8 steps, record the out-of-step event and execute the warning and position compensation; when the position deviation reaches or exceeds 8 steps, multiple consecutive detections determine out-of-step, or multiple consecutive position detections are invalid, stop the current drive and execute reset and zeroing; if a valid position cannot be obtained after reset and zeroing, output a serious fault signal.

7. A full-cycle out-of-step detection system for an electronic expansion valve based on high-frequency signal injection, characterized in that, This system is used to implement the method according to any one of claims 1-6, comprising a controller, a stepper drive circuit, a current sampling circuit, a memory, and an electronic expansion valve, wherein the controller comprises: An injection frequency selection module is used to select a first injection frequency and a second injection frequency based on the candidate position range, position separability, and PWM carrier spectrum. The high-frequency signal injection module is used to inject a corresponding high-frequency AC voltage signal into the stepper motor winding of the electronic expansion valve through the stepper drive circuit. The response confidence determination module is used to perform dynamic bandpass filtering on the winding current response acquired by the current sampling circuit, and determine the response confidence based on the high-frequency response energy and the low-frequency disturbance energy. The impedance feature extraction module is used to perform synchronous demodulation when the response confidence reaches a preset threshold, determine the dual-frequency complex impedance, equivalent resistance, position-dependent inductance and impedance phase, and obtain the electrical position according to the calibration feature library. The candidate position disambiguation module is used to generate full-stroke candidate positions based on the electrical position and the number of pole pairs of the stepper motor, determine the unique candidate position that falls within the candidate position range, and perform mechanical backlash compensation according to the direction of movement to obtain the actual mechanical position of the valve needle. The out-of-step detection module is used to compare the actual mechanical position of the valve needle with the commanded position, and output the result of normal, warning, position compensation, reset zeroing or serious fault according to the out-of-step detection threshold corresponding to the operating status.

8. The electronic expansion valve full-cycle out-of-step detection system according to claim 7, characterized in that, The current sampling circuit reuses the original current sampling resistor, amplifier circuit and analog-to-digital conversion channel of the stepper drive circuit; the electronic expansion valve does not have a potentiometer, Hall sensor or encoder for feedback valve needle position. And / or, the memory is used to store the candidate injection frequency set, the position separability corresponding to the candidate injection frequency, valve opening direction calibration data, valve closing direction calibration data, valve opening direction mechanical idle distance, valve closing direction mechanical idle distance, the most recent valid position, response confidence threshold, out-of-step judgment threshold, and out-of-step event records; the controller triggers detection when the electronic expansion valve completes the stepping action, an overheating abnormality occurs, the compressor starts or stops, the system is powered on, or the operating mode is switched, and extends the detection interval when the electronic expansion valve is stationary and continuous detection is normal.

9. An electronic device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 6.

Citation Information

Patent Citations

  • Opening feedback electronic expansion valve and control system thereof

    CN103673426A

  • High-frequency voltage signal injection-based rotor position observation method of permanent-magnet synchronous motor

    CN109617487A