A double-pole double-throw radio frequency relay fault real-time detection method and system

Abstract

The application provides a kind of double-pole double-throw radio frequency relay fault real-time detection method and system, it is related to radio frequency relay fault detection technical field.The method includes the following steps: respectively to each independent signal path applies test excitation signal and obtains response signal;By frequency domain feature extraction to response signal, obtain the nonlinear distortion characteristic value of the physical state of the internal contact of each independent signal path;The nonlinear distortion characteristic value of each independent signal path is compared with the corresponding path independent health standard in real time respectively, and the deviation is obtained;According to deviation, whether physical deterioration occurs in each independent signal path is determined independently.The method of the application aims to solve the false alarm and the missing report problem caused by the physical characteristic difference of two independent signal paths in the prior art double-pole double-throw radio frequency relay under long-term operation and environmental change, which is conducive to improving the accuracy of fault detection and the efficiency of production test.

Description

Technical Field

[0001] This invention relates to the field of radio frequency relay fault detection technology, and more specifically, to a real-time fault detection method and system for double-pole double-throw radio frequency relays. Background Technology

[0002] In modern industrial production, especially in the automated testing of communication modules, ensuring the reliability of critical components is paramount. Double-pole double-throw (DPDT) RF relays, as core components for signal switching, directly impact the accuracy of test results. However, under prolonged continuous operation and the influence of environmental factors, these relays may exhibit some subtle, early-stage faults that are difficult to detect. Traditional fault detection methods often struggle to accurately identify these subtle and intermittent problems, particularly when the relay's internal characteristics are affected by constantly changing operating conditions. A significant challenge lies in the fact that the two independent signal paths within a single DPDT relay, due to unavoidable minor differences in manufacturing processes, do not respond entirely identically to environmental changes such as temperature. This asymmetry can lead to false alarms in existing detection systems, or more dangerously, to overlook genuine performance degradation, potentially harming product quality and production efficiency.

[0003] In automated production line testing environments for communication modules, to verify the module's transceiver performance, testing equipment needs to frequently switch the module's antenna port between a signal transmitter and a signal receiver analyzer. This switching function is typically performed by a double-pole double-throw (DPDT) RF relay. To ensure the accuracy of test results, it is crucial that the relay is in normal working order after each switching action, as any poor contact or switching failure will directly invalidate the test data and may even lead to misjudgments of the entire batch of products. Therefore, a real-time fault detection system is deployed to monitor the relay's operating status. The most direct implementation of this system is to measure key parameters of the RF signal passing through the relay's signal path, such as signal attenuation, after each relay switching and compare it with a pre-set standard value. If the measured attenuation exceeds the standard range, the system determines that the relay has malfunctioned and immediately suspends the testing process.

[0004] However, in actual continuous production testing, the test equipment operates uninterruptedly for extended periods. The processor, power supply, and the module under test (DUT) itself generate significant heat, causing the internal operating temperature of the chassis to rise continuously. This temperature increase alters the physical characteristics of all components in the RF path, including the relays themselves, connecting RF cables, and microstrip lines on the circuit board. The direct consequence is that even if the relays themselves are intact, the signal attenuation throughout the entire path will naturally and slowly increase with rising temperature. This leads to a problem: after the equipment has been running for a period, this normal attenuation drift caused purely by temperature may exceed the fixed fault detection threshold initially calibrated at room temperature, triggering false alarms. These false alarms frequently interrupt normal production testing, significantly reducing production efficiency.

[0005] To address the false alarm problem caused by temperature drift, technicians improved the detection method. They installed a temperature sensor near the relay and experimentally determined the baseline attenuation value of the signal path along the health relay at different temperatures, establishing a correlation between temperature and attenuation. During actual testing, the system first reads the current real-time temperature, then calculates the normal attenuation baseline value at that temperature based on this correlation, and finally compares the actual measured value with this dynamic, temperature-varying baseline value. This temperature-compensated approach largely eliminates false alarms caused by changes in overall ambient temperature, enabling the detection system to operate stably in varying temperature environments.

[0006] However, as the system's operating time increased, a more subtle problem gradually emerged. The double-pole double-throw relay actually contains two independent signal switching paths. Although they are packaged within the same device, due to unavoidable minor process differences, the metal contacts and internal support structures of these two paths do not respond exactly the same way to thermal expansion and contraction. In other words, the specific curves of signal attenuation versus temperature for each path differ. Existing temperature compensation schemes typically use only a single temperature sensor and a unified compensation model to correct for the temperature effect on the entire relay. This model is essentially an averaging or approximation of the characteristics of the two paths. This approach fails when dealing with the asymmetric thermodynamic characteristics of the two paths. For example, when the temperature rises, the compensation model may accurately predict the attenuation change of path one, but for the more thermally sensitive path two, the actual increase in attenuation will be greater than the model's prediction. This will cause the system to generate false alarms again, incorrectly judging that path two has failed. A more serious scenario is that if the performance of path one begins to show genuine, minor degradation—for example, slight oxidation of the contacts leading to increased attenuation—but at the same time, the thermal sensitivity of path one happens to be lower than the compensation model's setting, then the overcompensation amount given by the model may exactly offset the increase in attenuation caused by the actual fault, thus perfectly "hiding" the unfolding fault. Ultimately, the detection system reports everything as normal, but in reality, one of the test paths is unreliable. This will continuously produce erroneous test data, posing a significant risk to product quality.

[0007] There is currently no effective technical solution to the above problems. Summary of the Invention

[0008] The purpose of this invention is to provide a real-time fault detection method and system for double-pole double-throw radio frequency relays, which aims to solve the problems of false alarms and missed alarms caused by the differences in physical characteristics of the two independent signal paths in the existing double-pole double-throw radio frequency relays under long-term operation and environmental changes, thereby improving the accuracy of fault detection and the efficiency of production testing.

[0009] In a first aspect, the present invention provides a method for real-time fault detection of a double-pole double-throw radio frequency relay, comprising the following steps:

[0010] S1. When the two independent signal paths of the double-pole double-throw radio frequency relay are in a preset trigger state, a test excitation signal is applied to each of the independent signal paths respectively;

[0011] S2. Obtain the response signal of each independent signal path under the action of the test excitation signal;

[0012] S3. By performing frequency domain feature extraction on the response signal, nonlinear distortion feature values ​​characterizing the physical state of the internal contacts of each independent signal path are obtained;

[0013] S4. Obtain the current environmental status parameters, and match the preset independent health benchmark of each independent signal path according to the environmental status parameters;

[0014] S5. The nonlinear distortion characteristic value of each independent signal path is compared with the corresponding independent health benchmark of the path in real time to obtain the deviation.

[0015] S6. Based on the deviation, independently determine whether each of the independent signal paths has experienced physical degradation.

[0016] The real-time fault detection method for double-pole double-throw radio frequency relays provided by this invention can detect and determine the two independent signal paths inside the double-pole double-throw radio frequency relay separately, effectively avoiding false alarms and missed alarms caused by the difference in characteristics between the two paths in traditional methods, and significantly improving the accuracy and reliability of fault detection.

[0017] Secondly, the present invention provides a real-time fault detection system for a double-pole double-throw radio frequency relay, comprising:

[0018] The control module is used to apply test excitation signals to each of the two independent signal paths of the double-pole double-throw radio frequency relay when the two independent signal paths are in a preset trigger state.

[0019] The acquisition module is used to acquire the response signals of each of the independent signal paths under the action of the test excitation signal;

[0020] The extraction module is used to extract frequency domain features from the response signal to obtain nonlinear distortion feature values ​​that characterize the physical state of the internal contacts of each independent signal path.

[0021] The matching module is used to obtain the current environmental state parameters and match the preset independent health benchmarks of each independent signal path according to the environmental state parameters.

[0022] The comparison module is used to compare the nonlinear distortion characteristic value of each independent signal path with the corresponding independent health benchmark of the path in real time to obtain the deviation.

[0023] The determination module is used to independently determine whether each of the independent signal paths has undergone physical degradation based on the deviation amount.

[0024] As can be seen from the above, the real-time fault detection method for double-pole double-throw RF relays provided by this invention effectively solves the problems of false alarms and missed alarms caused by the differences in physical characteristics (such as inconsistent thermal sensitivity) of the two independent signal paths of a double-pole double-throw relay in the prior art. Traditional methods usually use a uniform compensation model or fixed threshold, which cannot accurately reflect the independent health status of each path. This application achieves independent determination of the physical degradation state of each path by separately exciting, acquiring, extracting nonlinear distortion features, and comparing with the independent health benchmark of the two independent signal paths in real time. This "independent detection, independent determination" strategy can accurately identify minor degradation in a single path, avoiding the situation where the fault is masked by the normality of the other path, and also avoiding misjudgment caused by inconsistent effects of environmental changes on the two paths. Therefore, this application significantly improves the accuracy and reliability of fault detection for double-pole double-throw RF relays, ensures the quality and efficiency of automated testing of communication modules, avoids invalid test data and product misjudgment caused by relay faults, and has significant practical value and technological progress.

[0025] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description

[0026] Figure 1 This is a flowchart of a real-time fault detection method for a double-pole double-throw radio frequency relay provided in an embodiment of the present invention.

[0027] Figure 2 This is a schematic diagram of a real-time fault detection system for a double-pole double-throw radio frequency relay provided in an embodiment of the present invention.

[0028] Label Explanation:

[0029] 100. Control module; 200. Acquisition module; 300. Extraction module; 400. Matching module; 500. Comparison module; 600. Judgment module. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0031] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0032] In fault detection applications of double-pole double-throw RF relays, due to differences in manufacturing processes, the two independent signal paths exhibit different thermodynamic response characteristics when ambient temperature changes. This difference makes it difficult for detection methods based on a single temperature compensation model to distinguish between signal parameter changes caused by temperature drift and actual faults caused by contact degradation. Consequently, during the judgment process, the system may misjudge normal temperature drift as a fault, or fail to identify actual physical degradation, thus negatively impacting the accuracy of test results and the continuity of the production process.

[0033] For example, in automated testing environments for communication modules, double-pole double-throw (DPDT) RF relays are configured to switch antenna ports between signal transmitters and signal receivers / analyzers. When the test equipment operates continuously, the internal temperature rises. Due to differences in material properties, the attenuation curves of the RF signals in the two signal paths within the relay vary with temperature. In this scenario, as the temperature rises, the attenuation of one signal path may exceed the threshold set by the averaging compensation model, triggering a false alarm and causing unplanned interruptions to the test process. Simultaneously, if the other signal path has contact oxidation, the increase in attenuation may be masked by over-correction in the compensation model, preventing timely fault detection and resulting in invalid test data output. Furthermore, this problem is particularly prominent in long-running test production lines, directly impacting the reliability of the testing process and the validity of the data.

[0034] If the above problems are not addressed, false alarms from the fault detection system will frequently cause production testing interruptions, resulting in idle testing resources and decreased production efficiency. False alarms may cause physically degraded signal paths to continue participating in the testing process, outputting incorrect test results and thus triggering quality risks for the entire batch of products. These quality risks not only affect the market reliability of the product but may also damage the reputation of the manufacturing company. Therefore, there is an urgent need for a detection mechanism that can adapt to the differences in thermal response within the internal circuits of a double-pole double-throw RF relay.

[0035] For reference, see the appendix. Figure 1 This invention provides a real-time fault detection method for a double-pole double-throw radio frequency relay, comprising the following steps:

[0036] S1. When the two independent signal paths of the double-pole double-throw RF relay are in the preset trigger state, apply test excitation signals to each independent signal path respectively;

[0037] S2. Obtain the response signals of each independent signal path under the action of the test excitation signal;

[0038] S3. By extracting frequency domain features from the response signal, nonlinear distortion characteristic values ​​that characterize the physical state of the contacts within each independent signal path are obtained;

[0039] S4. Obtain the current environmental status parameters and match the preset independent health benchmarks of each independent signal path according to the environmental status parameters;

[0040] S5. The nonlinear distortion characteristic values ​​of each independent signal path are compared with the corresponding independent health benchmark of the path in real time to obtain the deviation.

[0041] S6. Based on the deviation, independently determine whether physical degradation has occurred in each independent signal path.

[0042] For ease of understanding, the following explains some key terms in this embodiment:

[0043] Double-pole double-throw (DPDT) RF relay: This is an RF relay with two independent switching contacts (pole) and two independent switching positions (throw). It can switch an RF signal from one input to two different outputs, or from two different inputs to one output. In automated testing of communication modules, this type of relay is often used to switch RF signals between different test paths, such as switching between transmitter and receiver tests. Internally, it contains two independent signal paths, each consisting of independent contacts and transmission structures.

[0044] Nonlinear distortion eigenvalues: These are parameters characterizing the physical state of the internal contacts of an RF relay. When an RF signal passes through the relay, if the contacts exhibit physical degradation such as oxidation, carbonization, or fretting wear, it will cause nonlinear distortion in the signal. This means that in addition to the original signal frequency, harmonic components (such as the second harmonic 2f and the third harmonic 3f) and intermodulation distortion products will be generated. Nonlinear distortion eigenvalues ​​quantify the intensity or relative intensity of these nonlinear distortion products, such as harmonic distortion degree or intermodulation distortion degree. These eigenvalues ​​are highly sensitive to microscopic physical changes in the contacts and can reflect early signs of degradation.

[0045] Independent Health Baseline for Each Path: This is a pre-defined reference standard for each independent signal path of a double-pole double-throw RF relay, used to determine its health status. Because the two paths within the relay may have manufacturing differences and different thermal response characteristics, a separate health baseline needs to be established for each path. This baseline can be a range of nonlinear distortion characteristics, a curve or model varying with environmental parameters, used to compare with currently measured nonlinear distortion characteristics during real-time monitoring to assess whether physical degradation has occurred in the path.

[0046] This application proposes a real-time fault detection method for a double-pole double-throw radio frequency relay, aiming to solve the problem of false alarms or missed alarms caused by inconsistent thermal responses of the two independent signal paths inside the double-pole double-throw radio frequency relay.

[0047] In practical implementation, when the two independent signal paths of the double-pole double-throw RF relay are in a preset trigger state, test excitation signals are applied to each independent signal path respectively. The preset trigger state can occur during automated production line testing of the communication module, after the double-pole double-throw RF relay completes a signal path switching action, or within a very short silent period reserved in the testing process. The test excitation signal can be applied via a high-purity signal generator, such as the Keysight N5183B MXG X-series signal generator, connected to the relay's input port through an RF switch matrix. The signal generator can be set to output a sine wave signal with a frequency of 1GHz and a power of -10dBm. If dual-frequency signals are required, it can be set to output signals at both 1GHz and 1.001GHz, each with a power of -13dBm. The system precisely controls the timing and power of the signal injection to ensure no interference with the normal testing process. This rapid injection and control is designed to meet the high efficiency and high throughput requirements of automated production lines, avoiding prolonged occupation of test resources.

[0048] Subsequently, the response signals of each independent signal path under the test excitation signal are acquired. The response signals can be acquired in real time and the output RF signal of each path can be digitized using a high-sensitivity spectrum analyzer. This spectrum analyzer has a sufficiently high sampling rate and dynamic range to capture weak nonlinear distortion components in the signal.

[0049] Next, by extracting frequency domain features from the response signal, nonlinear distortion characteristic values ​​representing the physical state of the contacts within each independent signal path are obtained. After receiving the signal, the spectrum analyzer uses its built-in Fast Fourier Transform (FFT) function to convert the time-domain signal into a frequency-domain spectrum. The system is configured with the spectrum analyzer's marking function to automatically identify and measure the power P_fundamental of the fundamental frequency signal, as well as the power P_distortion at preset harmonic frequencies (e.g., 2f, 3f) or intermodulation frequencies (e.g., f1+f2, f1-f2). Then, the system calculates the relative intensity of these distortion products relative to the fundamental frequency signal, typically expressed in dBc (dB value relative to the carrier). The calculation formula is: Distortion = 10log10(P_distortion / P_fundamental). For example, if the fundamental frequency signal power is -10dBm and the second harmonic power is -70dBm, then the second harmonic distortion is -60dBc. The system records these calculated distortion values ​​as the nonlinear distortion characteristics of each path. This precise quantification of weak distortion signals can capture early signs of microscopic physical changes within the relay.

[0050] Simultaneously, the system acquires current environmental state parameters and matches them with preset independent health benchmarks for each independent signal path. Environmental state parameters may include temperature, humidity, etc. The independent health benchmark can be a pre-established database or model that stores the range of nonlinear distortion characteristic values ​​for each independent signal path under different environmental conditions in a healthy state. When the relay is in a healthy state (e.g., during the initial calibration of a batch of validated healthy relays on a production line), the system independently establishes a unique nonlinear distortion benchmark value or benchmark spectrum for each path. This benchmark reflects the normal fluctuation range of the nonlinear distortion characteristics of the healthy relay at different temperatures.

[0051] Then, the nonlinear distortion characteristic values ​​of each independent signal path are compared in real time with the corresponding independent health benchmark for that path to obtain the deviation. The system continuously tracks the nonlinear distortion characteristics of each path. The comparison process can be either directly comparing whether the current measurement value falls within the health benchmark range, or calculating the difference between the current value and the benchmark center value.

[0052] Finally, based on the deviation, the system independently determines whether physical degradation has occurred in each independent signal path. If the intensity of harmonic or intermodulation distortion products in a certain path shows a significant and sustained increasing trend relative to its healthy baseline, even if its conventional linear attenuation parameters (such as insertion loss) are still within acceptable limits, the system will immediately determine that early physical degradation has occurred in that specific path. By independently monitoring and analyzing the nonlinear distortion characteristics of the two paths, the system can accurately distinguish and attribute the fault to the specific path (e.g., clearly indicating whether path one or path two is deteriorating), rather than making a general judgment that the entire relay is faulty.

[0053] The following example will provide a more detailed explanation of the above technical solution:

[0054] In an automated testing line for communication modules, a double-pole double-throw (DPDT) RF relay is used to switch the RF port of the module under test between a test signal source and a spectrum analyzer. This relay contains two independent signal paths, path one and path two. Traditional methods might determine the relay's health status simply by measuring signal attenuation, but the inconsistent thermal responses of the two paths can lead to false alarms or missed alarms.

[0055] The method of this application first applies test excitation signals to path one and path two separately by a high-purity signal generator after the relay completes a switching action or during the quiet period of the test procedure. For example, the signal generator injects a 1 GHz, -10 dBm sine wave signal into path one, and simultaneously injects a 1.001 GHz, -10 dBm sine wave signal into path two. This method of independently applying excitation signals ensures that each path is not affected by the other path when being tested, thereby enabling independent evaluation of its performance.

[0056] Next, a highly sensitive spectrum analyzer acquires the output response signals of path one and path two in real time. For the response signal of path one, the spectrum analyzer analyzes its spectrum, paying particular attention to the power at 2 GHz (second harmonic). For the response signal of path two, the spectrum analyzer focuses on the power at 2.002 GHz (second harmonic). By calculating the relative strength (e.g., dBc) of these harmonic powers relative to the fundamental frequency signal power, the nonlinear distortion characteristic values ​​of path one and path two are obtained. These nonlinear distortion characteristic values ​​can directly reflect whether there is microscopic physical degradation such as oxidation or carbonization in the internal contacts of the relay.

[0057] Simultaneously, the system acquires the current ambient temperature parameters. Before the relays are put into use, the system has established independent path health benchmarks for Path 1 and Path 2 respectively. For example, the health benchmark for Path 1 might be a range of -60dBc±3dB second harmonic distortion at 25℃, and this range will be dynamically adjusted according to temperature changes. The health benchmark for Path 2 might be -62dBc±4dB second harmonic distortion at 25℃, and its temperature compensation model is also different from that of Path 1 to adapt to its unique physical characteristics.

[0058] During real-time monitoring, the system compares the currently measured nonlinear distortion characteristic value of path one with the corresponding health benchmark to obtain the deviation. Similarly, the nonlinear distortion characteristic value of path two is also compared with the corresponding health benchmark to obtain its deviation. For example, if the current ambient temperature is 40℃, the system calculates the health benchmark range for path one and path two at 40℃ based on a preset model. If the current nonlinear distortion characteristic value of path one is -55dBc, and its health benchmark range at 40℃ is [-58dBc, -52dBc], then path one is considered normal. However, if the current nonlinear distortion characteristic value of path two is -50dBc, and its health benchmark range at 40℃ is [-57dBc, -51dBc], then the nonlinear distortion characteristic value of path two is considered to exceed the healthy range, indicating physical degradation.

[0059] Finally, based on these independent deviations, the system independently determines whether physical degradation has occurred in path one and path two. In this way, even if the thermal response characteristics of the two paths differ, the system can accurately identify which path has deteriorated, avoiding false alarms or missed alarms caused by the overall averaging process in traditional methods. For example, if the contacts in path one begin to oxidize slightly, its nonlinear distortion characteristic value will increase, and the system will immediately detect this change and issue an alarm, without being masked by the normal performance of path two.

[0060] The above technical solution addresses the problem of inaccurate fault detection caused by inconsistent thermal responses of the two signal paths in a double-pole double-throw (DPDT) RF relay by independently processing the two signal paths and utilizing frequency domain feature extraction and independent health benchmark matching. Traditional fault detection methods often struggle to accurately identify subtle and intermittent problems, especially when the relay's internal characteristics are affected by constantly changing operating conditions. A significant challenge lies in the fact that the two independent signal paths within a single DPDT relay do not respond entirely to environmental changes such as temperature due to unavoidable minor differences in manufacturing processes. This asymmetry can lead to false alarms in existing detection systems, or more dangerously, miss genuine performance degradation.

[0061] This application ensures that each independent signal path is independently excited by applying test excitation signals under preset trigger states, avoiding errors caused by overall averaging. The response signals of each path under the test excitation signals are acquired, providing basic data for subsequent analysis. Nonlinear distortion characteristic values ​​are obtained by frequency domain feature extraction of the response signals. These characteristic values ​​effectively characterize changes in the physical state of internal contacts because frequency domain feature extraction can capture the sensitivity of nonlinear distortion to contact degradation. Current environmental state parameters are acquired, and preset independent health benchmarks for each path are matched based on these parameters. This allows for independent adjustment of the benchmarks for the thermal response characteristics of each path, adapting to the impact of environmental changes. The nonlinear distortion characteristic values ​​of each path are compared with their corresponding health benchmarks in real time to obtain the deviation, ensuring that the evaluation of each path is conducted independently and avoiding mutual interference. The deviation values ​​are used to independently determine whether physical degradation has occurred in each path, thereby accurately detecting faults and reducing false alarms and missed alarms. This method is insensitive to temperature-induced linear decay drift because the appearance of nonlinear distortion products indicates physical degradation at the contact material level, rather than a simple temperature effect, thus effectively avoiding false alarms and missed alarms caused by mismatch of a single temperature compensation scheme.

[0062] In some embodiments, the specific steps in step S1 include:

[0063] S11. Determine the frequency of the test excitation signal based on the rated operating frequency range of the double-pole double-throw RF relay and the expected spectral position of the nonlinear distortion products;

[0064] S12. Set the power of the test excitation signal; the power is such that nonlinear distortion products can be detected on the noise floor of the spectrum analyzer, and the power is lower than the rated power threshold for long-term operation of the relay.

[0065] S13. Based on the determined frequency and set power, configure the test excitation signal and apply it to each independent signal path.

[0066] In the above scheme, the rated operating frequency range refers to the frequency range within which the double-pole double-throw RF relay can operate normally and stably, as specified in its design and manufacturing. This range is usually clearly given by the manufacturer in its product specifications, for example, from DC to 6 GHz, or from 10 MHz to 18 GHz. This range can be obtained by consulting the relay's datasheet or conducting preliminary performance tests. The expected spectral position of nonlinear distortion products refers to the signal components that deviate from the fundamental frequency or intermodulation frequency in the spectrum due to the nonlinear characteristics of components such as the internal contacts of the relay under the action of a test excitation signal. These products typically manifest as harmonics (such as second harmonic 2f, third harmonic 3f, etc.) or intermodulation products (such as third-order intermodulation products 2f1-f2, 2f2-f1, etc.). The prediction of their spectral position can be based on the study of the relay's nonlinear model, the accumulation of empirical data, or identification through broadband frequency scanning during the calibration phase. Determining the frequency of the test excitation signal aims to select one or a set of frequencies such that the resulting nonlinear distortion products can be clearly detected without aliasing with other signals or environmental noise during normal relay operation. One approach is to select a frequency point within the relay's rated operating frequency range where its nonlinear response is known to be significant. Another approach is to select one or more frequencies such that the nonlinear distortion products they produce fall within the low-noise region of the spectrum analyzer, or avoid frequencies from other known interference sources.

[0067] Setting the power of the test excitation signal aims to balance detection sensitivity and relay protection. The noise floor of a spectrum analyzer refers to the power level of random noise generated by its internal circuitry when there is no input signal. Any signal below this noise floor cannot be effectively detected. Therefore, the power of the test excitation signal must be high enough to ensure that the resulting weak nonlinear distortion products can be clearly identified by the spectrum analyzer, i.e., its power level must be above the noise floor. The rated power threshold refers to the maximum input power that the relay can withstand under long-term continuous operation. Exceeding this threshold may cause overheating, burning, or accelerated aging of the relay's internal contacts, thus shortening its lifespan. Therefore, the power of the test excitation signal must be strictly controlled below this threshold to ensure that the detection process is non-destructive to the relay. Configuring the test excitation signal refers to generating an RF signal with specific parameters using a signal generator or similar device based on a determined frequency and set power. This may include setting parameters such as the signal generator's output frequency, output power, modulation type (if needed), and signal waveform. Applying the test excitation signal to each independent signal path refers to injecting the configured test excitation signal into each independent signal path of the double-pole double-throw RF relay through appropriate RF connectors, cables, and switch matrices. This ensures that each path can independently receive the test signal, allowing for independent acquisition of its response.

[0068] This solution first precisely determines the frequency of the test excitation signal based on the rated operating frequency range of the double-pole double-throw RF relay and the expected spectral position of the nonlinear distortion products before applying the test excitation signal. This step ensures that the selected frequency can effectively excite the nonlinear effects inside the relay and that the generated nonlinear distortion products fall within an easily detectable spectral region, thus avoiding detection failure or inefficiency that may result from blindly selecting a frequency. Subsequently, the power of the test excitation signal is set, following a dual constraint: on the one hand, it must be high enough to ensure that the generated nonlinear distortion products are clearly above the noise floor of the spectrum analyzer, thereby guaranteeing detection sensitivity; on the other hand, it must be strictly below the rated power threshold for long-term operation of the relay to avoid any form of damage or accelerated aging of the relay during testing. This precise power control is crucial for achieving non-destructive testing. Finally, based on these precisely determined frequencies and set power levels, the test excitation signals are configured and generated, and applied to the individual signal paths of the double-pole double-throw RF relay. Through these steps, this solution ensures that the applied test excitation signal can both maximize the excitation of the weak nonlinear distortion inside the relay and simultaneously protect the relay from unnecessary stress. This optimized excitation signal makes the nonlinear distortion characteristics in the response signal acquired in subsequent steps more significant and reliable, thereby greatly improving the sensitivity and accuracy of detecting early physical degradation of relays. Compared to basic fault detection methods, this scheme intelligently optimizes the source of the applied test excitation signal, enabling subsequent response signal acquisition, frequency domain feature extraction, and comparison with health benchmarks to be performed under more optimized input conditions. This improves the reliability and effectiveness of the entire detection system, especially in scenarios involving weak degradation signals and requiring long-term non-destructive monitoring, where its advantages are even more pronounced.

[0069] The following is a concrete example. During automated production line testing of communication modules, after a double-pole double-throw RF relay completes a signal path switching action, or during a very short silent period reserved in the testing process, a high-purity signal generator, such as a Keysight N5183B MXG X-series signal generator, is connected to the relay's input port via an RF switch matrix, such as a Mini-Circuits ZFSWA2-63DR+. When determining the frequency of the test excitation signal, one can first consult the relay's datasheet to understand its rated operating frequency range, for example, DC to 6GHz. Then, based on experience or pre-calibration tests, determine the spectral location where the expected nonlinear distortion products (such as second harmonics or third-order intermodulation products) may appear. For example, if single-frequency testing is selected, it can be determined that the second harmonic (2GHz) or third harmonic (3GHz) response is more pronounced and easily detected around 1GHz. If dual-frequency testing is chosen, it can be determined that the responses of third-order intermodulation products (such as 2f1-f2 or 2f2-f1) are more significant near 1 GHz and 1.001 GHz. A trade-off needs to be made when setting the power of the test excitation signal. For example, experiments have shown that when the signal power is -10 dBm, the resulting nonlinear distortion products are clearly detectable on the noise floor of the spectrum analyzer. Meanwhile, consulting the relay datasheet confirms that its long-term rated power threshold is +20 dBm. Therefore, setting the power of the test excitation signal to -10 dBm, far below the rated power threshold, ensures the non-destructive nature of the testing process. Based on the determined frequency and set power, the signal generator can be configured to output a sine wave signal with a frequency of 1 GHz and a power of -10 dBm. If a dual-frequency signal is required, it can be configured to output signals at both 1 GHz and 1.001 GHz, each with a power of -13 dBm. The timing of this signal injection is precisely controlled by the central controller of the production line testing system, such as an NI PXIe-8880-based controller, ensuring that the signal injection is completed within 100 microseconds after the relay switching stabilizes, thus avoiding interference with the transient response during the relay switching process. This rapid injection and control is designed to meet the high efficiency and high throughput requirements of automated production lines, avoiding prolonged occupation of testing resources.

[0070] Through the above technical solution, this application effectively solves the problems in traditional detection methods, such as the difficulty in capturing weak nonlinear distortion products caused by early physical degradation of relay internal contacts, and the potential for improper test excitation signal parameters to accelerate relay aging, leading to insufficient detection sensitivity and accuracy. Specifically, by determining the frequency of the test excitation signal based on the rated operating frequency range of the double-pole double-throw RF relay and the expected spectral position of the nonlinear distortion products, the test signal is ensured to efficiently excite the nonlinear effects inside the relay, making weak nonlinear distortion products easier to detect. Simultaneously, by setting the power of the test excitation signal, it is ensured that the nonlinear distortion products are detectable above the noise floor of the spectrum analyzer, while remaining strictly below the rated power threshold for long-term relay operation. This maximizes detection sensitivity and accuracy without compromising relay lifespan. This optimized test excitation signal provides high-quality input for subsequent response signal acquisition and nonlinear distortion feature extraction, enabling the entire real-time fault detection method to identify relay physical degradation earlier and more accurately, avoiding false alarms and missed alarms, and significantly improving the reliability and efficiency of automated production line testing.

[0071] In some embodiments, the specific steps in step S11 include:

[0072] S111. During the relay calibration phase, frequency scanning is performed within the rated operating frequency range of the double-pole double-throw RF relay. During the frequency scanning process, the spectral characteristics of the nonlinear distortion products at each scanning frequency point are monitored and analyzed to obtain the power level and signal-to-noise ratio of the nonlinear distortion products.

[0073] S113. Select and determine the frequency of the test excitation signal based on the power level and signal-to-noise ratio of the nonlinear distortion products.

[0074] The relay calibration phase refers to the initial performance evaluation and parameter setting process for a double-pole double-throw (DPDT) RF relay before its actual use, or after maintenance or upgrades. Its purpose is to establish baseline data for the relay in a healthy state, providing a reliable reference for subsequent real-time fault detection. This phase is typically conducted in a controlled environment to ensure data accuracy and repeatability. Frequency scanning within the rated operating frequency range involves gradually or continuously changing the frequency of the test excitation signal within the frequency range where the DPDT RF relay is designed or specified to operate normally. For example, it can start from the lower limit of the frequency range and increase to the upper limit in preset steps (e.g., 1MHz, 10MHz), or use a logarithmic scanning method. This aims to comprehensively probe the relay's nonlinear response characteristics within this frequency range and identify the most suitable frequency points for detecting nonlinear distortion products. When monitoring and analyzing the spectral characteristics of nonlinear distortion products at each scanned frequency point, frequency domain analysis of the relay's output response signal is performed using professional spectrum analysis equipment (e.g., the spectrum analysis function of a spectrum analyzer or vector network analyzer). This analysis aims to identify and quantify harmonic components (such as second and third harmonics) or intermodulation distortion products (such as third-order intermodulation products) other than the fundamental frequency signal. The presence and intensity of these nonlinear distortion products directly reflect the physical state of the relay's internal contacts (such as oxidation and poor contact). The power level and signal-to-noise ratio (SNR) of these nonlinear distortion products are obtained, where the power level refers to the absolute power value of the nonlinear distortion product at a specific frequency (e.g., dBm), and the SNR is the ratio of the power of the nonlinear distortion product to the noise power near that frequency. These parameters are obtained to evaluate the detectability of the nonlinear distortion products, i.e., whether they are strong enough to be clearly distinguished from background noise, and whether their intensity is sufficient as a sensitive indicator of relay physical degradation. Finally, based on the power level and SNR of the nonlinear distortion products, the frequency of the test excitation signal is selected and determined, aiming to screen the most effective test excitation signal frequency for exciting and detecting nonlinear distortion products from the large amount of data acquired during the frequency scanning process. Selection criteria typically include prioritizing frequencies with high power levels and excellent signal-to-noise ratios for nonlinear distortion products. This ensures that the nonlinear distortion products can be reliably captured and quantified during actual testing, thereby improving the sensitivity and accuracy of fault detection. Additionally, frequencies that exhibit good stability or sensitivity to nonlinear distortion products under different environmental conditions (such as temperature variations) can also be considered.

[0075] This application's solution involves a comprehensive frequency scan of a double-pole double-throw RF relay during the relay calibration phase. During this process, the spectral characteristics of nonlinear distortion products at each scanned frequency are monitored and analyzed in real time to obtain their power level and signal-to-noise ratio (SNR). This frequency selection mechanism based on actual measurement data effectively avoids frequency selection biases caused by theoretical models or empirical settings. Specifically, by performing a detailed frequency scan within the relay's rated operating frequency range, the system can comprehensively understand the relay's nonlinear response characteristics at different frequencies, which is crucial for identifying subtle changes in the microscopic physical state of the relay's internal contacts. Monitoring and analyzing the power level and SNR of nonlinear distortion products during the scan allows for direct evaluation of their detectability and reliability. Distortion products with high power levels are easier to detect, while a high SNR ensures the accuracy of the detection results and reduces background noise interference. Finally, based on these measured power level and SNR data, the system can intelligently select and determine an optimal test excitation signal frequency. This frequency maximizes the excitation of the relay's nonlinear characteristics and generates easily detectable nonlinear distortion products with a high SNR. This data-driven frequency selection method, combined with the steps of applying test excitation signals to each independent signal path and obtaining response signals, ensures that the nonlinear distortion characteristic values ​​obtained during subsequent frequency domain feature extraction of the response signals can accurately and truthfully characterize the physical state of the contacts within each independent signal path. By selecting the optimal test frequency, the detection sensitivity of nonlinear distortion products can be significantly improved, enabling even early and slight physical degradation of the relay to be detected in a timely manner. This avoids the false alarms or missed alarms caused by improper frequency selection in traditional methods, greatly improving the accuracy and reliability of fault detection.

[0076] As a specific implementation, the frequency of the test excitation signal can be determined during the calibration phase before the first use of a double-pole double-throw (DPDT) RF relay. First, the DPDT RF relay to be calibrated is connected to a high-purity RF signal generator and a high-sensitivity spectrum analyzer. The signal generator is configured to scan the frequency range of the relay's rated operating frequency (e.g., from 100 MHz to 6 GHz). For example, it can be set to scan in 10 MHz steps, or a logarithmic stepping method can be used to more finely cover key frequency bands. At each scan frequency point, the signal generator applies a test excitation signal with a preset power (e.g., -10 dBm) to one independent signal path of the relay. Simultaneously, the spectrum analyzer monitors the response signal output by that path in real time. For each scanned frequency point, the spectrum analyzer performs a Fast Fourier Transform (FFT) to obtain the frequency domain spectrum and automatically identifies and measures harmonics (such as the second harmonic 2f and the third harmonic 3f) or intermodulation products generated by the relay's nonlinear characteristics (such as 2f1-f2, 2f2-f1, etc. for dual-frequency excitation f1, f2). The spectrum analyzer records the power level (e.g., -70dBm) of these nonlinear distortion products and their signal-to-noise ratio (SNR) relative to the background noise. After completing the scan of the entire frequency range, the system collects nonlinear distortion product data for all scanned frequency points. For example, the system might find that at an excitation frequency of 1.2 GHz, the power level of the second harmonic is -65dBm with an SNR of 30dB; while at an excitation frequency of 2.5 GHz, the power level of the third harmonic is -75dBm with an SNR of 15dB. By analyzing this data, the system selects one or more optimal frequencies as the test excitation signal frequencies for subsequent real-time detection. For example, the system might preferentially choose to excite at 1.2 GHz because it generates a second harmonic with a higher power level and better signal-to-noise ratio, making nonlinear distortion products easier to detect and quantify. This choice ensures that even minute signs of physical degradation of the relay's internal contacts can be captured with the highest sensitivity and reliability during actual fault detection.

[0077] Through the above technical solution, this application effectively solves the problem of unreliable detection of nonlinear distortion products caused by improper frequency selection in traditional methods. A comprehensive frequency scan is performed during the relay calibration stage, and the frequency of the test excitation signal is selected based on the power level and signal-to-noise ratio of the nonlinear distortion products. This ensures that the selected frequency can maximize the excitation of the relay's nonlinear characteristics and generate easily detectable nonlinear distortion products with a high signal-to-noise ratio. This significantly improves the accuracy and sensitivity of nonlinear distortion feature value extraction, enabling timely detection even of early and slight physical degradation of the relay's internal contacts. Therefore, this solution effectively avoids false alarms or missed alarms caused by inaccurate frequency selection, thereby improving the overall accuracy and reliability of real-time fault detection in double-pole double-throw RF relays and providing a more robust guarantee for the automated testing of communication modules.

[0078] In some embodiments, the specific steps in step S3 include:

[0079] S31. Perform digital down-conversion processing on the response signal to shift the frequency of the nonlinear distortion product to the target frequency, thereby obtaining the shifted response signal; specific steps include:

[0080] S311. The response signal is digitally processed to obtain a digital signal;

[0081] S312. Mix the digital signal with a digital local oscillator signal of a preset frequency to obtain a difference frequency signal, which is used as the response signal after the shift.

[0082] S32. Filter the response signal after relocation to obtain the filtered response signal;

[0083] S33. Measure the power of the filtered response signal and use it as a characteristic value of nonlinear distortion.

[0084] The process involves extracting frequency domain features from the response signal to obtain nonlinear distortion characteristic values ​​that characterize the physical state of the contacts within each independent signal path. This process converts the time-domain signal into frequency-domain information, identifying and quantifying nonlinear distortion products that directly reflect the microscopic changes in the physical state of the contacts. This process can be implemented in several ways. For example, one method is to use Fast Fourier Transform (FFT) to directly perform spectral analysis on the digitized response signal to identify and measure the power of harmonics or intermodulation products. Another method is to use digital down-conversion combined with narrowband filtering to shift the target distortion products to the baseband or low-IF frequency for precise measurement.

[0085] Digital down-conversion of the response signal shifts the frequency of the nonlinear distortion products to the target frequency, resulting in a shifted response signal. Its purpose is to convert the high-frequency nonlinear distortion product signal to a lower frequency, facilitating subsequent filtering and accurate measurement, while effectively suppressing high-frequency noise and interference. This processing can be achieved by multiplying the digitized signal with the digital local oscillator signal using a digital mixer and then performing a low-pass filter; or by employing a multi-stage down-conversion structure to gradually reduce the signal frequency.

[0086] The response signal is digitized to obtain a digital signal. This process converts the continuous analog response signal into a discrete digital signal, laying the foundation for subsequent digital signal processing and improving its accuracy and anti-interference capability. This processing can be achieved by using an analog-to-digital converter (ADC) to sample and quantize the analog response signal; or by using oversampling and digital decimation techniques to improve the effective resolution.

[0087] The digitized signal is mixed with a digital local oscillator signal of a preset frequency to obtain a difference frequency signal, which serves as the shifted response signal. Its function is to shift the frequency by multiplying the digitized signal with a digitally generated local oscillator signal using a digital multiplier, thus shifting the target frequency component to a new frequency position. This mixing process can be achieved by multiplying the digitized signal with the digital local oscillator signal (e.g., generated by a direct digital frequency synthesizer, DDS); or by using a complex mixer to generate two signals, in-phase (I) and quadrature (Q), to preserve phase information.

[0088] The response signal after transfer is filtered to obtain a filtered response signal. The purpose of this filtering is to remove unwanted frequency components (such as image frequencies, other distortion products, or noise) from the transferred response signal, highlight the nonlinear distortion products of the target, and improve measurement accuracy. This filtering can be implemented using digital low-pass or band-pass filters, such as finite impulse response (FIR) filters or infinite impulse response (IIR) filters; or an adaptive filter can be used, dynamically adjusting the filtering parameters according to the signal characteristics.

[0089] The power of the filtered response signal is measured and used as a nonlinear distortion characteristic value. Its function is to quantify the intensity of the processed target nonlinear distortion product. This power value directly reflects the degree of physical degradation of the internal contacts of the relay. This measurement can be achieved by performing a sum-of-squares and averaging operation on the filtered digital signal to obtain its average power; or by using a digital power meter or root mean square (RMS) detector to directly measure the effective power of the signal.

[0090] The proposed solution involves digitally down-converting the response signal to shift the frequency of the nonlinear distortion products to the target frequency, filtering the shifted response signal, and finally measuring its power to accurately extract the nonlinear distortion characteristic values. Specifically, after acquiring the response signals of each independent signal path of the double-pole double-throw RF relay under the test excitation signal, these analog response signals are first digitized, converting them into discrete digital signals. Then, these digital signals are mixed with a digital local oscillator signal of a preset frequency. Through digital multiplication, the high-frequency components of the nonlinear distortion products are precisely shifted to a lower target frequency, generating a difference frequency signal as the shifted response signal. This step effectively shifts the target signal from the high-frequency band to a more easily processed and analyzed frequency band while avoiding high-frequency noise interference. Next, the shifted response signal is digitally filtered to remove image frequencies, other irrelevant frequency components, and background noise generated during the mixing process, resulting in a pure, filtered response signal containing only the target nonlinear distortion products. Finally, the power of the filtered response signal is precisely measured, and this power value is used as a nonlinear distortion characteristic value characterizing the physical state of the relay's internal contacts. This refined frequency domain feature extraction method, combined with the steps of applying a test excitation signal and obtaining the response signal, ensures high-precision and high-reliability detection of weak nonlinear distortion signals inside the relay in complex electromagnetic environments, providing accurate data support for subsequent fault diagnosis.

[0091] The following is a concrete example to illustrate this. In practical applications, when a double-pole double-throw RF relay receives a test excitation signal and generates a response signal, this response signal is first digitized by a high-speed analog-to-digital converter (ADC), for example, converting the analog signal into a 14-bit or 16-bit digital signal at a sampling rate of hundreds of megahertz. Subsequently, these digitized signals are fed into a digital signal processor (DSP) or a field-programmable gate array (FPGA). Inside the DSP / FPGA, a digital mixer multiplies the digitized signal with a digital local oscillator signal generated by a direct digital frequency synthesizer (DDS). For example, if the nonlinear distortion products are located at 2 GHz, and we want to shift them to a target intermediate frequency of 10 MHz, the frequency of the digital local oscillator signal will be precisely set to 1.99 GHz or 2.01 GHz. After mixing, sum and difference frequency components are generated, where the difference frequency signal (e.g., 10 MHz) is the shifted response signal. Next, the shifted response signal is passed through a digital bandpass filter, such as an FIR filter designed with a center frequency of 10MHz and a bandwidth of 100kHz, to filter out other frequency components and noise, resulting in a filtered response signal. Finally, the power value is obtained by performing a root mean square (RMS) calculation on the filtered digital signal. For example, if the fundamental frequency signal power is -10dBm and the second harmonic power is -70dBm, then the second harmonic distortion is -60dBc. The system records these calculated distortion values ​​as the nonlinear distortion characteristics of each path. This precise quantification of weak distortion signals can capture early signs of microscopic physical changes within the relay.

[0092] Through the above technical solution, this application can accurately shift the frequency of nonlinear distortion products to the target frequency and perform efficient filtering, thereby accurately measuring their power as a characteristic value. This significantly improves the accuracy and reliability of nonlinear distortion characteristic value extraction and effectively avoids the influence of noise interference and environmental changes on the detection results in traditional methods. Therefore, this application can more accurately characterize the physical state of the internal contacts of a double-pole double-throw RF relay, providing a solid foundation for real-time, high-reliability fault detection, thus effectively avoiding false alarms and missed alarms caused by mismatched single temperature compensation schemes, and improving the robustness and accuracy of the detection system.

[0093] In some embodiments, the independent health benchmark of the pathway includes an adaptive health operating range; the adaptive health operating range is dynamically constructed based on the nonlinear distortion characteristic value of the historical health status of the corresponding independent signal pathway; the adaptive health operating range is used to compare with the nonlinear distortion characteristic value of the corresponding independent signal pathway in real time when executing step S5 to obtain the deviation.

[0094] The real-time fault detection method for double-pole double-throw radio frequency relays also includes the following steps:

[0095] A1. When the current nonlinear distortion characteristic value is within the adaptive healthy operating range and the independent signal path meets the preset health conditions, the adaptive healthy operating range is updated at a preset rate.

[0096] A2. When the current nonlinear distortion characteristic value continuously exceeds the adaptive healthy operating range, or when it shows a preset abnormal change trend within the adaptive healthy operating range, pause or slow down the update of the adaptive healthy operating range.

[0097] The independent health baseline for each signal path is designed as an adaptive health operating range. This adaptive health operating range is dynamically constructed based on the nonlinear distortion characteristic values ​​of the historical health status of the corresponding independent signal path. This means that for each independent signal path within the double-pole double-throw RF relay, the system establishes a dedicated, time-adjustable health baseline interval based on its past nonlinear distortion characteristic value data under healthy conditions. For example, this dynamic construction process can employ statistical methods, such as calculating the mean and standard deviation of historical health data, and then defining an interval encompassing the vast majority of health data as the upper and lower limits of the operating range; alternatively, machine learning algorithms can be used to predict the normal range of nonlinear distortion characteristic values ​​under current environmental conditions by performing pattern recognition and modeling on historical health data. During step S5, this adaptive health operating range is used to compare in real time with the current nonlinear distortion characteristic value of the corresponding independent signal path to obtain the deviation between the two.

[0098] Building upon this, this application further proposes an intelligent update mechanism for the adaptive healthy operating range. Specifically, when the current nonlinear distortion characteristic value is within the adaptive healthy operating range, and the independent signal path simultaneously meets preset health conditions, the system updates the adaptive healthy operating range at a preset rate. The preset health conditions may include, but are not limited to, the signal insertion loss remaining within a broad normal range, or the absence of catastrophic fault signs such as signal interruption. The preset rate can be a fixed, small adjustment amount; for example, adjusting the center value of the range by a fixed percentage towards the current measurement value each time an update is performed. Alternatively, algorithms such as moving averages or exponential smoothing can be used to allow the range to slowly track the gradual drift of the nonlinear distortion characteristics of the relay caused by natural aging or environmental factors during long-term operation.

[0099] Conversely, when the current nonlinear distortion characteristic value continuously exceeds the adaptive health operating range, or even if it is within the adaptive health operating range but exhibits a preset abnormal change trend, the system will pause or slow down the update of the adaptive health operating range. "Continuously exceeding" can be defined as N consecutive measurements exceeding the range, or more than P% of the measurements exceeding the range in the most recent M measurements. The "preset abnormal change trend" can refer to measurements repeatedly being at the edge of the range and showing an accelerating deterioration trend, or a rate of change exceeding a preset threshold discovered through trend analysis (such as linear regression slope). By pausing or slowing down updates, abnormal data can be effectively prevented from contaminating the health status reference, thereby ensuring that the system can promptly detect early physical degradation that may occur in the internal contacts of the relay.

[0100] This application's solution, by introducing an adaptive healthy operating range and its intelligent update mechanism, combined with fundamental nonlinear distortion feature extraction and independent path detection methods, forms a more robust and accurate real-time fault detection system. First, the system can acquire the response signals of the two independent signal paths of a double-pole double-throw RF relay under the action of a test excitation signal in real time, and obtain nonlinear distortion feature values ​​characterizing the physical state of the internal contacts through frequency domain feature extraction. These nonlinear distortion feature values ​​are highly sensitive to microscopic physical changes in the contacts (such as oxidation and wear) and are relatively unaffected by linear decay drift caused by temperature. Based on this, an adaptive healthy operating range is constructed for each independent signal path as its independent health benchmark. This range is dynamically generated based on the historical health status data of that path, thus effectively addressing the problem of thermal response asymmetry caused by manufacturing differences between the two paths within the relay.

[0101] Building upon this, the proposed solution further enhances the accuracy and adaptability of detection by introducing an intelligent update strategy for the adaptive healthy operating range. When the system determines that the current nonlinear distortion characteristic value is within the adaptive healthy operating range, and the independent signal path meets other preset health conditions, the system will fine-tune and update the range at a preset, slow rate. This gradual update allows the health benchmark to continuously track the slow and normal drift of the nonlinear distortion characteristics of the relay caused by natural aging, environmental cumulative effects, and other factors during long-term operation, thereby avoiding false alarms caused by outdated benchmarks.

[0102] However, when the current nonlinear distortion characteristic value is detected to continuously exceed the adaptive healthy operating range, or even if it is within the range but exhibits a preset abnormal trend, the system will immediately pause or slow down the update of the adaptive healthy operating range. This key mechanism ensures that when the relay may experience early physical degradation, the health benchmark will not be "contaminated" by abnormal data and incorrectly expanded, thus masking the true fault. In this way, the system can keenly detect the transition of the relay's internal contacts from normal drift to actual degradation and issue timely warnings. This strategy of combining dynamic adjustment and freezing allows the detection system to accurately distinguish between the performance drift of the relay caused by environmental changes or normal aging and the performance degradation caused by early physical degradation, greatly reducing the risk of false and false faults and ensuring the accuracy of test results and production efficiency.

[0103] The following is a specific example to illustrate this. Suppose that in an independent signal path of a double-pole double-throw RF relay, its adaptive healthy operating range is set to a nonlinear distortion characteristic value (e.g., second harmonic distortion) between -65dBc and -55dBc.

[0104] In one specific implementation, when the system obtains the latest nonlinear distortion measurement value of -60dBc, it first determines whether this value falls within the adaptive healthy operating range set for the current path. Since -60dBc is within the range of -65dBc to -55dBc, and the system simultaneously detects that other key parameters such as the insertion loss of this independent signal path are at normal levels, it indicates that the path is currently in a healthy state. At this time, the system will fine-tune and update the adaptive healthy operating range of the path at a preset slow rate. For example, the center value of the range can be slightly adjusted towards -60dBc, and the upper and lower limits of the range can be recalculated based on historical data to adapt to the small, normal performance drift that may occur during long-term operation of the relay.

[0105] However, if the system subsequently obtains three consecutive nonlinear distortion measurements of -54dBc, -53dBc, and -52dBc, while -54dBc and -53dBc may still be at or near the edge of the range, -52dBc has already exceeded the upper limit of -55dBc. More importantly, this set of consecutive measurements shows a clear upward trend, which the system identifies as a pre-defined abnormal trend. In this case, the system will immediately pause or significantly slow down the updates of the adaptive healthy operating range of this independent signal path. The purpose of this is to prevent the range from being erroneously expanded, thereby including existing signs of degradation in the "healthy" category, ensuring that the system can issue potential degradation warnings in a timely manner and preventing faults from being masked.

[0106] Through the above technical solution, this application effectively solves the problem of traditional detection methods in distinguishing between normal performance drift and early physical degradation of relays. By dynamically constructing and intelligently updating an adaptive healthy operating range for each independent signal path, the system can accurately track the normal performance evolution of relays under long-term operation and environmental changes, thereby significantly reducing false alarms caused by reference mismatch. Simultaneously, when nonlinear distortion characteristic values ​​are detected to continuously exceed the healthy range or exhibit abnormal changing trends, the system can promptly pause or slow down the updating of the health reference, effectively preventing early degradation signs from being masked by the adaptiveness of the reference, thus avoiding missed detections. This mechanism ensures the sensitivity and accuracy of fault detection, enabling reliable monitoring of double-pole double-throw RF relays even in complex and variable working environments, greatly improving the quality and efficiency of automated testing of communication modules.

[0107] In some embodiments, the specific steps for dynamically constructing the adaptive health operating range based on the nonlinear distortion characteristic values ​​of the historical health status of the corresponding independent signaling pathways include:

[0108] B1. Collect the nonlinear distortion characteristic values ​​of the historical health status of independent signal pathways within a first preset time window, and calculate statistical parameters based on the collected nonlinear distortion characteristic values;

[0109] B2. Define the upper and lower limits of the adaptive healthy operating range based on statistical parameters.

[0110] The operational logic of this scheme is as follows: First, by collecting the nonlinear distortion characteristic values ​​of the historical health status of independent signal paths within a first preset time window, it ensures that the data used to establish the health benchmark accurately reflects the relay's health status and is timely. Independent signal paths refer to each independent signal transmission path within a double-pole double-throw RF relay, each possessing unique physical characteristics. The nonlinear distortion characteristic values ​​of the historical health status refer to the nonlinear distortion parameters reflecting the physical state of the internal contacts, measured when the relay is confirmed to be in normal operating condition. The first preset time window refers to a pre-defined time period, such as the calibration phase before the relay is put into use, or a periodic testing phase to confirm the health status after a period of stable operation. This technical feature aims to acquire raw data for constructing the health benchmark. This can be achieved by performing multiple nonlinear distortion measurements on each independent signal path under controlled health conditions before the relay leaves the factory or during initial installation, and collecting these measurements as historical health data; alternatively, during normal relay operation, when the system confirms its health status, nonlinear distortion characteristic values ​​can be periodically collected and incorporated into the historical health dataset. This targeted data collection avoids benchmark bias that may result from using outdated or irrelevant data. Secondly, based on these collected nonlinear distortion characteristic values, statistical parameters are calculated. This step provides an objective and repeatable basis for quantifying the health range. This technique aims to quantitatively analyze collected historical health data to extract its core distribution characteristics. Statistical parameters are numerical values ​​that describe the central tendency, dispersion, or distribution pattern of a dataset, such as mean, median, standard deviation, variance, and percentiles. The arithmetic mean (µ) and standard deviation (σ) of the collected nonlinear distortion characteristic values ​​can be calculated to characterize the central location and dispersion of the data; the median and interquartile range (IQR) can also be calculated to reduce the impact of outliers on statistical results and more robustly reflect the data distribution. Through statistical analysis of historical health data, the typical performance and fluctuation characteristics of nonlinear distortion characteristic values ​​under health conditions can be accurately grasped. Finally, based on the statistical parameters, upper and lower limits of the adaptive health operating range are defined, enabling the established health range to dynamically reflect the actual changes of independent signaling pathways under health conditions, rather than fixed thresholds. This technology aims to define a healthy range that can effectively distinguish between normal fluctuations and abnormal degradation based on quantitative analysis results. The upper and lower limits of the adaptive healthy operating range are thresholds used to define the health status of independent signaling pathways. When real-time measurements exceed this range, it may indicate that degradation has occurred.The healthy operating range can be defined by setting the center of the range to the average value µ and the upper and lower limits to µ ± kσ, where k is a preset multiple (e.g., k=3, corresponding to the 3σ principle), to cover data points in most healthy states. Alternatively, the upper and lower limits can be defined based on the percentiles of the data, for example, defining the healthy operating range as between the 5th and 95th percentiles, or by determining it based on a specific confidence interval. Defining the upper and lower limits through statistical parameters allows the healthy range to be data-supported and adjustable.

[0111] This construction process, combined with the aforementioned real-time detection method, significantly improves the stability and reliability of fault detection. Specifically, when the two independent signal paths of the double-pole double-throw RF relay are in a preset triggered state, a test excitation signal is applied and the response signal is acquired, thereby extracting nonlinear distortion characteristic values. These characteristic values ​​are then compared in real time with the path-independent health benchmark (i.e., adaptive healthy operating range) dynamically constructed using the above method. Since the health benchmark is dynamically generated based on the historical health data of the path, it can better adapt to the normal fluctuations of the relay under different environmental conditions, thus more accurately identifying true physical degradation rather than normal drift caused by environmental changes during comparison. This dynamically constructed health benchmark enables the system to effectively reduce false alarms and missed alarms when determining whether physical degradation has occurred in the independent signal paths, especially when facing the asymmetric thermodynamic characteristics of the two internal paths of the double-pole double-throw relay, providing a more refined and accurate independent judgment.

[0112] The following is a specific example to illustrate this. As a concrete implementation, the steps for dynamically constructing the adaptive healthy operating range based on the nonlinear distortion characteristic values ​​of the historical health status of the corresponding independent signal paths can be as follows: First, during the production line calibration phase before the relay is first put into use, or after the relay has been rigorously verified to be in a healthy state, the system tests each independent signal path (e.g., path one and path two) of the double-pole double-throw RF relay at multiple stable ambient temperature points (e.g., 25°C, 40°C, and 55°C). For each path at each temperature point, the system repeatedly measures its second harmonic distortion (in dBc) at least 100 times. These measurements are collected and stored as the historical health status nonlinear distortion characteristic values ​​of that path at that temperature. Second, based on these collected nonlinear distortion characteristic values, the system calculates statistical parameters. Specifically, for the 100 second harmonic distortion measurements collected for each path at each temperature point, the system calculates its arithmetic mean (µ) and standard deviation (σ). For example, for the measurement data of path one at 25 degrees Celsius, the calculated average value µ is -60 dBc, and the standard deviation σ is 1 dBc. Finally, based on these statistical parameters, the upper and lower limits of the adaptive healthy operating range are defined. The center value of the healthy operating range can be set as the calculated average value µ, while its upper and lower limits can be set as [µ-3σ, µ+3σ]. For example, for the initial healthy operating range of path one at 25 degrees Celsius, its center value is -60 dBc, and the range is defined as [-63 dBc, -57 dBc]. These initial, statistically parameter-based healthy operating range data are stored in a non-volatile memory, such as EEPROM or a solid-state drive, for use as initial path-independent health benchmarks in subsequent real-time monitoring.

[0113] Through the above technical solution, this application solves the problem of lack of systematicity and accuracy in establishing health benchmarks by systematically constructing an adaptive healthy operating range. Specifically, by collecting historical nonlinear distortion characteristic values ​​of the health status of independent signal paths within a first preset time window, the data used for benchmark construction is ensured to have high timeliness and relevance, avoiding benchmark deviations caused by the use of inaccurate or outdated data. Statistical parameters are calculated based on these collected characteristic values, providing an objective and repeatable basis for the quantification of the health range, making the establishment of the health benchmark more scientific and rigorous. Finally, the upper and lower limits of the adaptive healthy operating range are defined according to the statistical parameters, enabling the health benchmark to dynamically adapt to the normal fluctuations of the relay in a healthy state. This allows for a more accurate distinction between normal performance drift and actual physical degradation of the relay during real-time comparison, significantly improving the accuracy and reliability of fault detection and effectively reducing false alarms and missed alarms.

[0114] In some embodiments, the specific steps in step A1 include:

[0115] A11. Obtain the sequence of nonlinear distortion characteristic values ​​of the independent signal path within the second preset time window;

[0116] A12. Analyze the degree of change in the nonlinear distortion eigenvalue sequence;

[0117] A13. Adjust the preset rate according to the degree of change to obtain the adjusted rate;

[0118] A14. Update the adaptive health operating range based on the adjusted rate.

[0119] Specifically, when acquiring the sequence of nonlinear distortion characteristic values ​​of independent signal paths within a second preset time window, this process aims to collect historical data on the nonlinear distortion characteristics of independent signal paths over a period of time, providing a basis for subsequent analysis of the degree of change. This sequence can be acquired in various ways. For example, the system can continuously monitor and record the nonlinear distortion characteristic values, and when it needs to update the adaptive healthy operating range, extract data within a specified time window from the stored historical database; alternatively, the system can dynamically collect the latest nonlinear distortion characteristic values ​​in the real-time data stream through a sliding time window mechanism to ensure the timeliness of the sequence.

[0120] In analyzing the degree of variation of the nonlinear distortion characteristic value sequence, this step aims to quantify the volatility, trend, or dispersion of the nonlinear distortion characteristic values ​​within a second preset time window. The analysis of the degree of variation can employ statistical methods, such as calculating the standard deviation, variance, or mean absolute deviation of the sequence to directly reflect the dispersion of the data; alternatively, time series analysis methods, such as moving averages, exponential smoothing, or trendline fitting, can be used to identify the volatility, trend, or slope of the sequence, thereby providing a more comprehensive assessment of its variation characteristics.

[0121] When adjusting the preset rate according to the degree of change to obtain the adjusted rate, this step aims to dynamically adjust the update speed of the adaptive healthy operating range based on the actual changes in the nonlinear distortion feature value sequence. The adjustment logic can be implemented based on preset rules or algorithms. For example, a lookup table or piecewise function can be established, with different adjustment factors corresponding to different intervals of change, or a new update rate value can be directly set. This ensures that when the degree of change is small, the update rate is slower to avoid oversensitivity; when the degree of change is large, the update rate is faster to quickly respond to actual degradation. Alternatively, an adaptive algorithm can be used to dynamically calculate the adjustment factor based on the real-time calculated degree of change, thereby enabling the update rate to respond more smoothly and accurately to changes in the nonlinear distortion feature values.

[0122] When updating the adaptive healthy operating range based on the adjusted rate, this step aims to correct the center value and / or boundary of the adaptive healthy operating range using the dynamically adjusted update rate. The update mechanism can employ a weighted average method, weighting the current measurement value with historical data using the adjusted rate as the weight, thereby updating the center value and / or boundary of the healthy operating range; alternatively, a step-size adjustment method can be used, determining the step size for the movement of the healthy operating range boundary or center value at each update based on the adjusted rate, ensuring that the update process both tracks real changes and maintains stability.

[0123] This application's solution addresses the problem of fixed-rate updates potentially failing to adapt to changes in eigenvalues ​​by intelligently adjusting the update rate of the adaptive healthy operating range. The solution first acquires a sequence of nonlinear distortion eigenvalues ​​from independent signal paths within a second preset time window. This provides a set of current and historical data to capture real-time trends in eigenvalue changes, laying the foundation for subsequent analysis. Next, the degree of change in this sequence is analyzed to identify the amplitude or direction of eigenvalue fluctuations, thereby assessing the urgency of the update requirement. Based on this degree of change, the preset update rate is dynamically adjusted to match the actual changes, avoiding lag or over-updates caused by a fixed rate. Finally, the adaptive healthy operating range is updated according to the adjusted rate, ensuring it accurately reflects the current state. This process emphasizes a mechanism of dynamic adjustment based on actual changes, enhancing the system's intelligence and robustness. In the aforementioned real-time fault detection method for double-pole double-throw RF relays, when independent signal paths meet preset health conditions, the update of the adaptive healthy operating range can be more accurate and timely, thereby improving the reliability and adaptability of fault detection.

[0124] The following is a concrete example to illustrate this. Suppose that during the real-time fault detection of a double-pole double-throw RF relay, the system needs to update the adaptive healthy operating range of independent signal path one.

[0125] First, in step A11, the system acquires a sequence of nonlinear distortion characteristic values ​​for an independent signal path within a second preset time window. For example, the system can extract data points of nonlinear distortion characteristic values ​​(such as second harmonic distortion) from the past 100 consecutive measurements, forming a sequence containing 100 values. These data points are stored in a circular buffer to ensure that the latest historical data is always included.

[0126] Next, in step A12, the system analyzes the degree of variation in the sequence of nonlinear distortion features. For example, the system can calculate the standard deviation of these 100 data points. If the calculated standard deviation is small (e.g., less than 0.5 dBc), it indicates that the nonlinear distortion features fluctuate gently; if the standard deviation is large (e.g., greater than 2 dBc), it indicates that the fluctuations are severe.

[0127] Then, in step A13, the system adjusts the preset update rate based on the analyzed degree of change. For example, the system can set a rule: when the standard deviation is less than 0.5 dBc, the update rate is adjusted to 0.01 (i.e., with each update, the center value of the healthy operating range moves 1% towards the latest measurement); when the standard deviation is between 0.5 dBc and 2 dBc, the update rate is adjusted to 0.05; and when the standard deviation is greater than 2 dBc, the update rate is adjusted to 0.1. In this way, when the nonlinear distortion characteristic value changes gradually, the update rate is slower, avoiding overreaction to normal fluctuations; when the change is drastic, the update rate is faster, allowing the healthy operating range to adapt to the actual trend of change more quickly.

[0128] Finally, in step A14, the system updates the adaptive healthy operating range of independent signal path one based on the adjusted rate. For example, if the current center value of the healthy operating range is -60dBc, the latest measured value is -59dBc, and the adjusted rate is 0.05, then the new center value will be updated to -60*(1-0.05)+(-59)*0.05=-59.95dBc. Simultaneously, the upper and lower limits of the healthy operating range will also be adjusted accordingly based on the new center value and the preset range width.

[0129] Through the above technical solution, this application can dynamically adjust the update rate of the adaptive healthy operating range according to the actual degree of change of the nonlinear distortion characteristic value sequence. This effectively solves the problem of untimely or inaccurate updates that may be caused by traditional fixed-rate updates, enabling the healthy operating range to more accurately reflect the true physical state of independent signal paths. Therefore, this application significantly improves the reliability and adaptability of fault detection for double-pole double-throw RF relays, reduces the risk of false alarms and missed alarms, and is particularly effective in more accurately identifying early physical degradation when the relay characteristics drift slowly over time or when operating in complex environments, thereby ensuring the stable operation of the test system and product quality.

[0130] Reference Appendix Figure 2 This invention provides a real-time fault detection system for a double-pole double-throw radio frequency relay (this system adopts the real-time fault detection method for double-pole double-throw radio frequency relays described in the above embodiment; for specific details, please refer to the corresponding steps above), comprising:

[0131] The control module 100 is used to apply test excitation signals to each independent signal path when the two independent signal paths of the double-pole double-throw radio frequency relay are in a preset trigger state.

[0132] The acquisition module 200 is used to acquire the response signals of each independent signal path under the action of the test excitation signal;

[0133] The extraction module 300 is used to extract the frequency domain features of the response signal to obtain nonlinear distortion feature values ​​that characterize the physical state of the contacts inside each independent signal path.

[0134] The matching module 400 is used to obtain the current environmental status parameters and match the preset independent health benchmarks of each independent signal path according to the environmental status parameters.

[0135] The comparison module 500 is used to compare the nonlinear distortion characteristic values ​​of each independent signal path with the corresponding independent health benchmark in real time to obtain the deviation.

[0136] The determination module 600 is used to independently determine whether physical degradation has occurred in each independent signal path based on the deviation amount.

[0137] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0138] The above description is merely an embodiment of the present invention and is not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for real-time fault detection of a double-pole double-throw radio frequency relay, characterized in that, Including the following steps: S1. When the two independent signal paths of the double-pole double-throw radio frequency relay are in a preset trigger state, a test excitation signal is applied to each of the independent signal paths respectively; S2. Obtain the response signal of each independent signal path under the action of the test excitation signal; S3. By performing frequency domain feature extraction on the response signal, nonlinear distortion feature values ​​characterizing the physical state of the internal contacts of each independent signal path are obtained; S4. Obtain the current environmental status parameters, and match the preset independent health benchmark of each independent signal path according to the environmental status parameters; S5. The nonlinear distortion characteristic value of each independent signal path is compared with the corresponding independent health benchmark of the path in real time to obtain the deviation. S6. Based on the deviation, independently determine whether each of the independent signal paths has experienced physical degradation.

2. The real-time fault detection method for a double-pole double-throw radio frequency relay according to claim 1, characterized in that, The specific steps in step S1 include: S11. Determine the frequency of the test excitation signal based on the rated operating frequency range of the double-pole double-throw radio frequency relay and the spectral position of the expected nonlinear distortion products; S12. Set the power of the test excitation signal; the power is such that the nonlinear distortion product can be detected on the noise floor of the spectrum analyzer, and the power is lower than the rated power threshold for long-term operation of the relay; S13. Based on the determined frequency and set power, configure the test excitation signal and apply it to each of the independent signal paths.

3. The real-time fault detection method for a double-pole double-throw radio frequency relay according to claim 2, characterized in that, The specific steps in step S11 include: S111. During the relay calibration phase, frequency scanning is performed within the rated operating frequency range of the double-pole double-throw radio frequency relay. During the frequency scanning process, the spectral characteristics of the nonlinear distortion products at each scanning frequency point are monitored and analyzed to obtain the power level and signal-to-noise ratio of the nonlinear distortion products. S112. Select and determine the frequency of the test excitation signal based on the power level and signal-to-noise ratio of the nonlinear distortion product.

4. The real-time fault detection method for a double-pole double-throw radio frequency relay according to claim 2 or 3, characterized in that, The specific steps in step S3 include: S31. Perform digital down-conversion processing on the response signal to shift the frequency of the nonlinear distortion product to the target frequency, thereby obtaining the shifted response signal; S32. Filter the response signal after relocation to obtain the filtered response signal; S33. Measure the power of the filtered response signal and use it as the nonlinear distortion characteristic value.

5. The real-time fault detection method for a double-pole double-throw radio frequency relay according to claim 4, characterized in that, The specific steps in step S31 include: S311. The response signal is digitally processed to obtain a digital signal; S312. Mix the digitized signal with a digital local oscillator signal of a preset frequency to obtain a difference frequency signal, which is used as the response signal after relocation.

6. The real-time fault detection method for a double-pole double-throw radio frequency relay according to claim 1, characterized in that, The independent health benchmark of the pathway includes an adaptive health operating range; the adaptive health operating range is dynamically constructed based on the nonlinear distortion characteristic value of the historical health status of the corresponding independent signal pathway; the adaptive health operating range is used to compare with the nonlinear distortion characteristic value of the corresponding independent signal pathway in real time when executing step S5 to obtain the deviation amount.

7. The real-time fault detection method for a double-pole double-throw radio frequency relay according to claim 6, characterized in that, It also includes the following steps: A1. When the current nonlinear distortion characteristic value is within the adaptive healthy operating range and the independent signal path meets the preset health conditions, the adaptive healthy operating range is updated at a preset rate. A2. When the current nonlinear distortion characteristic value continuously exceeds the adaptive healthy operating range, or when a preset abnormal change trend is observed within the adaptive healthy operating range, the update of the adaptive healthy operating range is paused or slowed down.

8. The real-time fault detection method for a double-pole double-throw radio frequency relay according to claim 7, characterized in that, The specific steps for dynamically constructing the adaptive health operating range based on the nonlinear distortion characteristic values ​​of the historical health status of the corresponding independent signaling pathways include: B1. Collect the nonlinear distortion characteristic values ​​of the historical health status of the independent signal pathway within a first preset time window, and calculate statistical parameters based on the collected nonlinear distortion characteristic values; B2. Define the upper and lower limits of the adaptive healthy operating range based on the statistical parameters.

9. The real-time fault detection method for a double-pole double-throw radio frequency relay according to claim 8, characterized in that, The specific steps in step A1 include: A11. Obtain the nonlinear distortion feature value sequence of the independent signal path within a second preset time window; A12. Analyze the degree of change in the nonlinear distortion eigenvalue sequence; A13. Adjust the preset rate according to the degree of change to obtain the adjusted rate; A14. Update the adaptive healthy operating range based on the adjusted rate.

10. A real-time fault detection system for a double-pole double-throw radio frequency relay, characterized in that, include: The control module is used to apply test excitation signals to each of the two independent signal paths of the double-pole double-throw radio frequency relay when the two independent signal paths are in a preset trigger state. The acquisition module is used to acquire the response signals of each of the independent signal paths under the action of the test excitation signal; The extraction module is used to extract frequency domain features from the response signal to obtain nonlinear distortion feature values ​​that characterize the physical state of the internal contacts of each independent signal path. The matching module is used to obtain the current environmental state parameters and match the preset independent health benchmarks of each independent signal path according to the environmental state parameters. The comparison module is used to compare the nonlinear distortion characteristic value of each independent signal path with the corresponding independent health benchmark of the path in real time to obtain the deviation. The determination module is used to independently determine whether each of the independent signal paths has undergone physical degradation based on the deviation amount.

Images