Engine impact detection method and related device

Abstract

The invention discloses an engine impact detection method and a related device, and the method comprises the steps: determining an effective value of a noise signal for a target engine through employing a full-period first vibration signal generated in an operation process under a target working condition in a normal state, and the incidence relation among the full-period first vibration signal, a normal impact signal and the noise signal, and obtaining the effective value of the noise signal; the noise signal effective value is used for indicating the signal strength characteristic of the noise signal in the time domain. And determining a number threshold value of normal impact segments according to the noise signal effective value and the whole-period first vibration signal, and determining the number of actual impact segments according to the noise signal effective value and the whole-period second vibration signal. If the number of the actual impact segments exceeds the number threshold value, it is indicated that impact except normal impact is generated in the actual operation process, and it is determined that abnormal impact exists in the target engine in the whole period. Compared with the prior art, the impact signal threshold value does not need to be set manually, human subjective deviation can be avoided, and the detection accuracy is improved.

Description

Technical Field

[0001] The present application relates to the field of engine technology, and in particular to an engine impact detection method and related devices. Background Art

[0002] The engine is used to provide driving force for vehicles (such as vehicles and ships) to meet the driving needs of the vehicles (such as vehicles and ships). For example, a common four-stroke engine is provided.

[0003] In actual applications, during the normal operation of the engine, the moving parts will also generate excitations, which are transmitted through the engine body. The excitation is expressed in the vibration signal collected on the engine casing, which can be called impact. As the engine is used, internal components of the engine may fail. The existence of such a failure causes knocking failures in the engine during operation, such as cylinder tamping failures and abnormal valve clearance failures. Specifically, knocking failures will also generate excitations, which are also expressed in the vibration signal in the form of a shock. The impact caused by knocking failures is an abnormal shock, so that timely maintenance can be carried out to avoid potential risks caused by the continuous operation of the engine in the presence of a fault (such as engine damage, ship or vehicle driving safety issues, etc.).

[0004] In the related art, the shock signal threshold is set manually based on experience, etc., and the current vibration signal is judged to be a normal shock or an abnormal shock based on the shock signal threshold. However, this method is difficult to accurately distinguish between normal shock and abnormal shock, that is, there is a problem of poor detection accuracy. Summary of the invention

[0005] In order to solve the above technical problems, the present application provides an engine impact detection method and related devices, which no longer require manual setting of impact signal thresholds, can avoid human subjective bias and improve detection accuracy.

[0006] The embodiments of the present application disclose the following technical solutions:

[0007] On the one hand, an embodiment of the present application provides an engine impact detection method, the method comprising:

[0008] Acquire a first vibration signal of a whole period generated during the operation of the target engine in a normal state under the target working condition;

[0009] Determine the effective value of the noise signal according to the first vibration signal of the entire cycle, and the correlation relationship between the first vibration signal of the entire cycle, the normal impact signal, and the noise signal, wherein the effective value of the noise signal is used to indicate the signal strength characteristics of the noise signal in the time domain, and the correlation relationship is used to indicate that the signal value of the first vibration signal of the entire cycle at the tth moment includes the normal impact signal value at the tth moment and the noise signal value at the tth moment, and the tth moment is any moment in the time domain of the entire cycle;

[0010] Determining a threshold value of the number of normal impact segments included in the target engine under the target operating condition according to the effective value of the noise signal and the first vibration signal of the entire period;

[0011] Acquire a full-cycle second vibration signal generated during the operation of the target engine in the to-be-detected state under the target operating condition;

[0012] Determining the number of actual impact fragments included in the target engine under the target operating condition according to the effective value of the noise signal and the full-cycle second vibration signal;

[0013] If the number of the actual impact segments exceeds the number threshold, it is determined that the target engine has abnormal impacts within the entire cycle.

[0014] On the other hand, an embodiment of the present application provides an engine impact detection device, the device comprising an acquisition unit and a determination unit:

[0015] The acquisition unit is used to acquire a first vibration signal of a whole period generated during the operation of the target engine in a normal state under the target working condition;

[0016] The determining unit is used to determine the effective value of the noise signal according to the first vibration signal of the entire cycle, and the correlation relationship between the first vibration signal of the entire cycle, the normal impact signal, and the noise signal, wherein the effective value of the noise signal is used to indicate the signal strength characteristics of the noise signal in the time domain, and the correlation relationship is used to indicate that the signal value of the first vibration signal of the entire cycle at the tth moment includes the normal impact signal value at the tth moment and the noise signal value at the tth moment, and the tth moment is any moment in the time domain of the entire cycle;

[0017] The determination unit is further used to determine a threshold value of the number of normal impact segments included in the target engine under the target operating condition according to the effective value of the noise signal and the first vibration signal of the entire period;

[0018] The acquisition unit is further used to acquire a full-cycle second vibration signal generated during the operation of the target engine in the to-be-detected state under the target operating condition;

[0019] The determination unit is further used to determine the number of actual impact fragments included in the target engine under the target operating condition according to the effective value of the noise signal and the full-cycle second vibration signal;

[0020] The determining unit is further configured to determine that the target engine has abnormal impacts within the entire cycle if the number of the actual impact fragments exceeds the number threshold.

[0021] In another aspect, an embodiment of the present application provides a computer device, the computer device comprising a processor and a memory:

[0022] The memory is used to store program code and transmit the program code to the processor;

[0023] The processor is used to execute the engine impact detection method described in the above aspects according to the instructions in the program code.

[0024] It can be seen from the above technical solution that since the vibration signal generated during the operation of the engine includes a normal impact signal and a certain noise signal, and when running in a stable working condition, the noise signal does not change much in the whole cycle. Therefore, for the target engine, firstly, the first vibration signal of the whole cycle generated during the operation under the target working condition in a normal state, and the correlation between the first vibration signal of the whole cycle, the normal impact signal, and the noise signal are used to determine the effective value of the noise signal. Among them, the correlation is used to indicate that the signal value of the first vibration signal of the whole cycle at the tth moment includes the normal impact signal value at the tth moment and the noise signal value at the tth moment, and the tth moment is any moment in the time domain of the whole cycle. And, the effective value of the noise signal is used to indicate the signal strength characteristics of the noise signal in the time domain, that is, the effective value of the noise signal can indicate the size of the noise signal of the target engine in the target working condition in the whole cycle. Then, according to the effective value of the noise signal and the first vibration signal of the whole cycle, the threshold value of the number of normal impact segments included in the target engine under the target working condition is determined. And, the whole cycle second vibration signal generated during the operation of the target engine in the state to be detected under the target working condition is obtained, and the number of actual impact segments included in the target engine under the target working condition is determined according to the effective value of the noise signal and the whole cycle second vibration signal. It can be seen that the number threshold is used to indicate the normal impact situation included in the whole cycle of the target engine under the normal state and the target working condition, and the number of actual impact segments is used to indicate the actual impact situation included in the whole cycle of the target engine under the state to be detected and the same target working condition. Therefore, if the number of actual impact segments exceeds the number threshold, it indicates that in the actual operation process, impacts other than normal impacts are generated, such as additional impacts caused by knocking failures, so it is determined that the target engine has abnormal impacts in the whole cycle.

[0025] Compared with related technologies, it is no longer necessary to manually set the impact signal threshold, which can avoid human subjective bias and improve detection accuracy. In addition, since the effective value of the noise signal is determined under the target working condition and in the normal state, and in the application stage, the detection is also performed on the state to be detected under the same target working condition, because the working conditions before and after are consistent, it is possible to take into account the impact of the changing working conditions on the generated impact, so that it can be applied to the detection of various working conditions and has higher applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 A flow chart of an engine impact detection method provided in an embodiment of the present application;

[0028] Figure 2 A schematic diagram of determining the number of impact fragments provided in an embodiment of the present application;

[0029] Figure 3 A schematic diagram of a first envelope of a whole cycle provided in an embodiment of the present application;

[0030] Figure 4 A schematic diagram of a second envelope of a full cycle provided in an embodiment of the present application;

[0031] Figure 5 A schematic diagram of determining a threshold value of the number of normal impact fragments provided in an embodiment of the present application;

[0032] Figure 6 A schematic diagram of determining the number of actual impact fragments provided in an embodiment of the present application;

[0033] Figure 7 A schematic diagram of an impact in the late stage of a knocking fault provided by an embodiment of the present application;

[0034] Figure 8 A structural diagram of an engine impact detection device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0036] The engine impact detection method provided in the embodiment of the present application can be implemented by a computer device, which can be a terminal device or a server, wherein the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. Terminal devices include but are not limited to mobile phones, computers, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, ship-mounted terminals, etc. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, and this application does not impose any restrictions on this.

[0037] The specific examples are as follows:

[0038] Figure 1 This is a flow chart of an engine impact detection method provided in an embodiment of the present application, which is described by taking a terminal device as the aforementioned computer device as an example. The method includes S101-S106:

[0039] S101: Acquire a first vibration signal of a whole period generated during the operation of a target engine in a normal state under a target operating condition.

[0040] Wherein, normal state can refer to the state that there is no fault (such as knocking fault) in the target engine, and target operating condition can refer to the operating condition of the target engine (such as can be determined according to the speed, load, etc. of the target engine). Therefore, the impact included in the first vibration signal of the whole cycle can be considered as the normal impact during the operation of the engine. And, the whole cycle can refer to the process of a complete working cycle of the target engine. For example, taking the target engine as a four-stroke engine as an example, a complete working cycle can include intake, compression, work, and exhaust. It can be understood that the time domain length of the whole cycle corresponding to the engine of different strokes is different.

[0041] It should be noted that the present application does not impose any limitation on how to obtain the first vibration signal of the entire period. For ease of understanding, the present application provides the following example description:

[0042] In a possible implementation, the vibration signal generated during the operation of the target engine in a normal state and under the target operating condition for a full cycle can be collected to obtain a first vibration signal of the full cycle.

[0043] In another possible implementation, the vibration signal generated by an engine of the same model as the target engine during a full cycle under normal conditions and the same target operating conditions can be collected, that is, the vibration signal of the same engine and the same operating conditions on the market can be collected, and then the collected data can be used to determine the aforementioned first vibration signal of the full cycle. Based on this, it is beneficial to the existing market data and more efficient. And because the collected market data is still generated by the same model of engine, and is also generated under normal conditions and the same operating conditions, the validity of the first vibration signal of the full cycle determined based on this can be guaranteed, thereby ensuring the accuracy of subsequent detection.

[0044] In practical applications, the vibration signal can be a vibration acceleration signal. Usually, the vibration signal can be collected by a vibration acceleration sensor installed on the engine housing, specifically, the vibration acceleration signal of the engine housing in the time domain collected during a complete working cycle of the engine (such as intake, compression, power, and exhaust), and the full cycle vibration signal can be obtained.

[0045] It is understandable that the collected vibration signal is sensitive to the installation position of the vibration accelerometer sensor on the engine. Therefore, when using market data to determine the aforementioned first vibration signal of the entire cycle, the collected market data can be grouped according to the installation position of the vibration accelerometer sensor, and then the market data in the group with the same installation position as the vibration accelerometer sensor on the target engine is used to determine the aforementioned first vibration signal of the entire cycle, thereby ensuring the accuracy of the first vibration signal of the entire cycle.

[0046] In addition, when collecting market data, you can also collect design-related parameters of engines and key components, such as sensor installation location, number of cylinders, cylinder diameter, stroke, rated speed, rated power, etc. Based on this, from a richer dimension, filter the market data of engines with the same configuration as the target engine to determine the first vibration signal of the entire cycle corresponding to the target engine to ensure accuracy.

[0047] In practical applications, the collected market data can also be preprocessed (such as removing outliers, filtering, etc.) to eliminate interference caused by environmental noise, sensor errors, etc., to ensure that the processed data can be used for subsequent analysis. For example, removing outliers can include removing null values, repeated values, extreme values, values ​​that exceed technical conditions, and sensor hysteresis values, etc., to maintain the stability and reliability of the data. For another example, it is generally believed that the spectral energy of the impact waveform caused by knocking excitation is concentrated in the high frequency band, so the collected vibration signal can be high-pass filtered to eliminate low-frequency interference in the data.

[0048] S102: Determine the effective value of the noise signal according to the first vibration signal of the entire period, and the correlation relationship between the first vibration signal of the entire period, the normal impact signal, and the noise signal.

[0049] Since the vibration signal generated during the operation of the engine includes a normal impact signal and a certain noise signal, and when the engine is running in a stable condition, the noise signal does not change much within the entire cycle. Therefore, for the target engine, the effective value of the noise signal can be determined based on the first vibration signal of the entire cycle, and the correlation between the first vibration signal of the entire cycle, the normal impact signal, and the noise signal.

[0050] The correlation relationship is used to indicate that the signal value of the first vibration signal of the entire cycle at the tth moment includes the normal impact signal value at the tth moment and the noise signal value at the tth moment, and the tth moment is any moment in the time domain of the entire cycle. And the effective value of the noise signal is used to indicate the signal strength characteristics of the noise signal in the time domain, that is, the effective value of the noise signal can indicate the size of the noise signal of the target engine in the entire cycle under the target working condition.

[0051] It should be noted that the present application does not impose any limitation on how to determine the effective value of the noise signal. For ease of understanding, the present application provides the following example description:

[0052] During operation under stable working conditions (such as target working conditions), although the size of the noise signal generated by the engine operation has certain differences, the size of the noise signal is similar because the working conditions have not changed. Therefore, in a possible implementation, the effective value of the noise signal can be used to indicate the mean value of the signal strength of the noise signal in the time domain, that is, the effective value of the noise signal can indicate the mean value corresponding to the size of the noise signal generated by the target engine under the target working conditions in the entire cycle. For example, it can be the root mean square value, the arithmetic mean, etc.

[0053] In order to better understand the method of determining the effective value of the noise signal, the embodiment of the present application provides the following example of solving and determining the effective value of the noise signal:

[0054] The aforementioned association relationship can be expressed by the following formula (represented as Formula 1):

[0055] S(t)=A(t)+N(t)

[0056] In the above formula, S(t) is used to represent the signal value of the first vibration signal of the entire period at the t-th moment, A(t) is used to represent the normal impact signal value at the t-th moment, N(t) is used to represent the noise signal value at the t-th moment, t∈[0, t1], and t1 is used to represent the time length of the time domain.

[0057] Accordingly, the aforementioned S102 may include, when implemented specifically:

[0058] First, according to the first vibration signal of the entire period and the correlation relationship, the following formula (recorded as Formula 2) is constructed in the time domain:

[0059]

[0060] Next, let the effective value of the noise signal M0 = N(t) and construct the following formula in the time domain (recorded as Formula 3):

[0061]

[0062] In summary, the following expression is constructed (recorded as Formula 4):

[0063]

[0064] Solve for the zero point of R(M), and determine the value of M corresponding to the zero point of R(M) as the effective value M0 of the noise signal.

[0065] Because during the operation under stable working conditions (such as target working conditions), the size of the noise signal generated by the engine operation has certain differences, but because the working conditions have not changed, the size of the noise signal is similar. Therefore, the effective value M0 of the noise signal can be used to indicate the size of the noise signal in the time domain, that is, to indicate the aforementioned N(t). And, in practical applications, the noise signal belongs to additive noise in the full-cycle vibration signal and is not related to the impact of the engine. Therefore, the above-mentioned correlation relationship and related formulas can be constructed based on this noise assumption, and the effective value of the noise signal can be solved.

[0066] In order to solve M0, a function R(M) with M as the independent variable is constructed. If R(M) has a zero point, it means that the expression of the constructed R(M) at the zero point is the aforementioned formula 2, which means that the value of M at the zero point is the effective value of the noise signal M0.

[0067] To this end, we first prove the rationality of the constructed function R(M) and the existence of zeros. Specifically:

[0068] First, the rationality of constructing the function R(M) is proved as follows:

[0069] First, compared with Formula 2, Formula 4 quilt replace, quilt Where, when M=M0, it can be known from Formula 3 that:

[0070]

[0071] The above two equations are recorded as Formula 5 and Formula 6 respectively. From Formula 5 and Formula 6, we can know that: quilt The rationality of the substitution only needs to prove the following formula:

[0072]

[0073] Right now:

[0074]

[0075] Since the full-cycle vibration signal is a vibration acceleration signal, its signal is nearly symmetrical. Therefore, qualitative analysis, is close to 0, so the above derivation is valid. That is, the substitution of the above formula 2 to formula 4 is reasonable, that is to say, the construction of function R(M) is reasonable. In the specific implementation, the vibration acceleration signal S(t), after S(t) is high-pass filtered, the low-frequency part is removed, its mean is approximately zero, and the signal is close to a symmetrical form.

[0076] Secondly, prove that the function R(M) has zeros, as follows:

[0077] ① From formula 4, it can be seen that function R(M) is a continuous quadratic function with M as the independent variable;

[0078] ②When M=0, the function value is 0;

[0079] ③When M=max(|S(t)|), It can be obtained that R(M)>0, wherein max(|S(t)|) is used to indicate the maximum value of the vibration signal;

[0080] ④When M→0 + hour, It can be obtained that R(M)<0;

[0081] From these 4 items, we can see that the function R(M) has a zero point between 0 and max(|S(t)|).

[0082] Therefore, by solving the zero point of R(M), the M value corresponding to the zero point of R(M) is determined as the effective value of the noise signal M0. Based on this, the effective value of the noise signal corresponding to this target working condition is determined. In specific implementation, the zero point of R(M) can be solved by the dichotomy method to obtain M0.

[0083] Corresponding to the above example, the noise signal effective value is used to represent the size of the noise signal in the time domain. In a possible implementation, the signal strength root mean square value of the noise signal in the time domain can be used to represent the size of the noise signal in the entire time domain. That is, the noise signal effective value can be used to indicate the signal strength root mean square value of the noise signal in the time domain. In this example, the aforementioned formula 3 is also valid, thus ensuring the rationality of subsequent derivation, proof and solution.

[0084] S103: Determine a threshold value of the number of normal impact segments included in the target engine under the target operating condition according to the effective value of the noise signal and the first vibration signal of the entire period.

[0085] Since the effective value of the noise signal can indicate the signal strength characteristics of the noise signal in the time domain, and the first vibration signal of the entire cycle is in a normal state, the number threshold of the normal impact segments included in the target engine under the target working condition can be determined according to the effective value of the noise signal and the first vibration signal of the entire cycle. The number threshold can be used to indicate the normal impact conditions included in the entire cycle of the target engine under normal conditions and under the target working condition, so that it can be clear what the normal impact originally included in this working condition is.

[0086] It is understandable that when working under normal conditions, the number of excitations generated by the engine when completing a full working cycle is basically fixed, that is, the number of impact segments included in the full cycle vibration signal is basically fixed. Although the variable working conditions will affect the intensity of the excitation, it has little effect on the number of excitations. Therefore, the present application can detect whether there is an abnormal impact by determining the number of impacts. In other words, the aforementioned determined number threshold can be used as a reference for subsequent detection.

[0087] S104: Acquire a full-cycle second vibration signal generated during the operation of the target engine in the state to be detected under the target operating condition.

[0088] In the application stage, the full-cycle second vibration signal generated during the operation of the target engine in the state to be detected under the target working condition can be obtained. The state to be detected can be a state that needs to be detected, such as detecting the operation process of the target engine. And, the same target working condition as the aforementioned determination of the effective value of the noise signal is also targeted. Therefore, the full-cycle second vibration signal can indicate the situation of the vibration signal generated during the actual operation of the target engine under the target working condition, and whether there is an abnormal impact can be detected based on this in the subsequent detection.

[0089] In a specific implementation, a vibration acceleration sensor installed on the target engine may be used to collect vibration signals within a whole cycle to obtain the aforementioned whole cycle second vibration signal.

[0090] S105: Determine the number of actual impact segments included in the target engine under the target operating condition according to the effective value of the noise signal and the full-cycle second vibration signal.

[0091] Similar to the implementation of S104, for the state to be detected, the number of actual impact segments included in the target engine under the target working condition is still determined based on the effective value of the noise signal and the second vibration signal of the entire cycle. The number of actual impact segments can be used to indicate the actual impact conditions included in the entire cycle of the target engine running under the same target working condition in the state to be detected.

[0092] S106: If the number of actual impact segments exceeds the number threshold, it is determined that the target engine has abnormal impacts within the entire cycle.

[0093] The number threshold can be used to indicate the normal impact conditions included in the whole cycle of the target engine under normal state and under target working conditions, and the number of actual impact fragments can be used to indicate the actual impact conditions included in the whole cycle of the target engine under the state to be detected and under the same target working conditions. Therefore, if the number of actual impact fragments exceeds the number threshold, it indicates that in the actual operation process, impacts other than normal impacts are generated, such as additional impacts caused by knocking failures, so it can be determined that the target engine has abnormal impacts in the whole cycle.

[0094] In order to detect more accurately and promptly prompt users (such as ship owners and car owners) to perform maintenance, in one possible implementation, the proportion of abnormalities can be detected within a detection period to determine whether maintenance is needed.

[0095] In specific implementation, within a detection period including multiple whole cycles, if it is determined that the proportion of the number of whole cycles with abnormal impacts is greater than the proportion threshold, it indicates that many abnormal impacts have occurred within the detection period, which can be considered to be a knocking fault in the target engine. Therefore, an early warning prompt information is generated at this time. The early warning prompt information is used to indicate that the target engine has a knocking fault, wherein the number proportion is the ratio between the number of whole cycles with abnormal impacts and the total number of whole cycles in the detection period.

[0096] Based on this, when many abnormalities are detected, early warnings and prompts can be issued to notify relevant personnel (such as ship owners, vehicle owners, operators, etc.) to inspect and repair the engine and its parts to avoid possible safety risks in continuing to work. Alternatively, the vehicle or ship's automatic execution safety procedures can be notified for further testing.

[0097] For example, the detection period may be a period of 10 to 60 minutes, and the percentage threshold may be 80%. If the percentage of samples with abnormal impact exceeds 80%, an early warning message (such as a knocking fault warning) is given. Otherwise, a normal status prompt message may be given.

[0098] In addition, if there are abnormal impacts for a continuous period of time during the detection period, an early warning can be issued. For example, if there are abnormal impacts for 5 minutes in a row, an early warning can be issued.

[0099] It can be seen from the above technical solution that since the vibration signal generated during the operation of the engine includes a normal impact signal and a certain noise signal, and when running in a stable working condition, the noise signal does not change much in the whole cycle. Therefore, for the target engine, firstly, the first vibration signal of the whole cycle generated during the operation under the target working condition in a normal state, and the correlation between the first vibration signal of the whole cycle, the normal impact signal, and the noise signal are used to determine the effective value of the noise signal. Among them, the correlation is used to indicate that the signal value of the first vibration signal of the whole cycle at the tth moment includes the normal impact signal value at the tth moment and the noise signal value at the tth moment, and the tth moment is any moment in the time domain of the whole cycle. And, the effective value of the noise signal is used to indicate the signal strength characteristics of the noise signal in the time domain, that is, the effective value of the noise signal can indicate the size of the noise signal of the target engine in the target working condition in the whole cycle. Then, according to the effective value of the noise signal and the first vibration signal of the whole cycle, the threshold value of the number of normal impact segments included in the target engine under the target working condition is determined. And, the whole cycle second vibration signal generated during the operation of the target engine in the state to be detected under the target working condition is obtained, and the number of actual impact segments included in the target engine under the target working condition is determined according to the effective value of the noise signal and the whole cycle second vibration signal. It can be seen that the number threshold is used to indicate the normal impact situation included in the whole cycle of the target engine under the normal state and the target working condition, and the number of actual impact segments is used to indicate the actual impact situation included in the whole cycle of the target engine under the state to be detected and the same target working condition. Therefore, if the number of actual impact segments exceeds the number threshold, it indicates that in the actual operation process, impacts other than normal impacts are generated, such as additional impacts caused by knocking failures, so it is determined that the target engine has abnormal impacts in the whole cycle.

[0100] Compared with related technologies, it is no longer necessary to manually set the impact signal threshold, which can avoid human subjective bias and improve detection accuracy. In addition, since the effective value of the noise signal is determined under the target working condition and in the normal state, and in the application stage, the detection is also performed on the state to be detected under the same target working condition, because the working conditions before and after are consistent, it is possible to take into account the impact of the changing working conditions on the generated impact, so that it can be applied to the detection of various working conditions and has higher applicability.

[0101] That is to say, for the same target engine, the corresponding effective value of the noise signal is different under different target conditions. For example, the effective value of the noise signal under target condition A with high speed and high load may be higher than the effective value of the noise signal under target condition B with low speed and low load. In contrast, in practical applications, what is the target condition corresponding to the state to be detected? It is also necessary to use the corresponding effective value of the noise signal under the same condition to ensure the accuracy of detection. Based on this, the engine impact detection method provided in the present application can be considered as a detection method that is adaptive to variable working conditions. It can adapt to various working conditions and calculate the corresponding effective value of the noise signal, and then perform detection. It can take into account the impact of variable working conditions on the impact. Compared with related technologies, it can more accurately perform detection under various working conditions.

[0102] Through the above embodiments, the engine impact detection method provided by the present application is illustrated. Among them, the present application does not make any limitation on the specific implementation of how to determine the quantity threshold and other related contents. For ease of understanding, the present application embodiment provides the following example description:

[0103] Since the effective value of the noise signal is used to indicate the signal strength characteristics of the noise signal in the time domain, in a possible implementation, the effective value of the noise signal can be used to indicate the mean value of the signal strength of the noise signal in the time domain, such as the root mean square value of the signal strength shown in the aforementioned embodiment. It can be understood that such an effective value of the noise signal is usually less than the maximum signal strength of the noise signal. It can be seen from the above embodiments that the role of the effective value of the noise signal is to determine whether it belongs to an impact segment or a noise segment, thereby determining the number of impact segments. Therefore, in order to be able to more accurately determine whether it belongs to an impact or noise, in a possible implementation, first determine the impact signal threshold value based on the effective value of the noise signal, and the impact signal threshold value is used to indicate the maximum signal strength of the noise signal in the time domain, and then determine the aforementioned quantity threshold value and subsequent detection based on this.

[0104] In a specific implementation, if the effective value of the noise signal is used to indicate the mean signal strength of the noise signal in the time domain, first, the noise signal correction coefficient corresponding to the target operating condition can be determined, the noise signal correction coefficient is greater than 1, and the product of the noise signal correction coefficient and the effective value of the noise signal is determined as the impact signal threshold, wherein the impact signal threshold is used to indicate the maximum signal strength of the noise signal in the time domain.

[0105] Accordingly, when the aforementioned S103 is implemented, the quantity threshold can be determined according to the impact signal threshold and the first vibration signal of the entire cycle. Also, when the aforementioned S105 is implemented, the number of actual impact segments can be determined according to the impact signal threshold and the second vibration signal of the entire cycle.

[0106] Based on this, a larger shock signal threshold is determined based on the effective value of the noise signal, which is used as a reference for judging whether it is shock or noise, which is conducive to judging noise as noise and avoiding misjudging it as shock because the size of a certain section of the noise signal is higher than the mean. This is conducive to improving detection accuracy.

[0107] In specific implementation, the impact signal threshold can be determined by the following formula:

[0108] T=α*M0

[0109] In the above formula, T is used to represent the impulse signal threshold, M0 is used to represent the effective value of the noise signal, and α is used to represent the noise signal correction coefficient.

[0110] It should be noted that the present application does not impose any limitation on how to determine the noise signal correction coefficient. For ease of understanding, the present application provides the following example description:

[0111] In order to identify weak shocks (i.e. shocks with a vibration signal value that is not too large), the noise signal correction coefficient can be smaller than the coefficient threshold, so that the determined shock signal threshold is greater than the effective value of the noise signal, closer to the maximum signal strength of the noise signal, but at the same time greater than the maximum signal strength of the noise segment in the time domain and does not exceed the strength threshold. Based on this, when judging whether it is shock or noise based on the shock signal threshold, it is avoided to misjudge weak shocks as noise, so that weak shocks can be identified, which is conducive to ensuring detection accuracy.

[0112] In one example, the noise signal correction coefficient can be set according to whether or not the whole cycle vibration signal is subjected to noise reduction processing. For example, if the first vibration signal of the whole cycle and the second vibration signal of the whole cycle are both original signals that are actually collected, that is, no noise reduction processing is performed, then a larger noise signal correction coefficient can be set, such as the value range of α can be determined to be (3, 4). On the contrary, if the first vibration signal of the whole cycle and the second vibration signal of the whole cycle are both subjected to noise reduction processing (such as filtering processing, etc.) on the basis of the original signals that are actually collected, then a smaller noise signal correction coefficient can be set, such as the value range of α can be determined to be (2, 3).

[0113] In another example, the noise type can also be combined with the noise to evaluate the range of noise that the effective value of the noise signal can cover, to set the noise signal correction coefficient so that the impact signal threshold can basically cover all noise. For example, if the noise type is Gaussian noise, the effective value M0 of the noise signal is close to the standard deviation of the noise. At this time, it can be seen from the Gaussian distribution that (-3M0, 3M0) covers approximately 99.74% of the amplitude range of the noise. Therefore, when α>3, the impact signal threshold T value can be considered to cover almost all noise fragments. At the same time, in order to identify weak impacts, α<4 can be set, that is, the value range of α is determined to be (3, 4). For another example, if the noise type is white noise, the effective value M0 of the noise signal is close to the mean of the noise. At this time, when α>2, the impact signal threshold T value can be considered to cover almost all noise fragments. At the same time, in order to identify weak impacts, α<3 can be set, that is, the value range of α is determined to be (2, 3).

[0114] Furthermore, it should be noted that the present application does not impose any limitation on how to determine the number threshold and the number of actual impact segments according to the impact signal threshold. For ease of understanding, the present application provides the following example description:

[0115] In a possible implementation, the first vibration signal of the entire cycle can be divided into a plurality of first signal segments according to the impact signal threshold. If the signal amplitudes in the first signal segments are all greater than or equal to the impact signal threshold, the first signal segments are determined as normal impact segments, and the quantity threshold is obtained. Otherwise, it is determined as a noise segment. Accordingly, the second vibration signal of the entire cycle can be divided into a plurality of second signal segments according to the impact signal threshold, and if the signal amplitudes in the second signal segments are all greater than or equal to the impact signal threshold, the second signal segments are determined as actual impact segments, and the number of actual impact segments is obtained.

[0116] That is, the number threshold may refer to the number of impact segments in a full cycle vibration signal (in a normal state), and the number of actual impact segments refers to the number of impact segments in a full cycle vibration signal (in a state to be detected).

[0117] In specific implementation, the impact signal threshold is a value between the minimum and maximum values ​​of the whole cycle vibration signal. By comparing the amplitude of each sequence point of the whole cycle vibration signal with the impact threshold, the whole cycle vibration signal will be divided into several continuous signal segments, and then the corresponding number is determined. For ease of understanding, the embodiment of the present application provides the following Figure 2 A schematic diagram of determining the number of impact fragments is shown, specifically:

[0118] exist Figure 2 In this example, the horizontal axis is time, which can be normalized to the duration of a whole cycle in the time domain. Figure 2 In the example, within 0 to 1s, the vertical axis is the full-cycle vibration signal (which can be a vibration acceleration signal, in m / s 2 ). And, in Figure 2 In the example, the waveform is formed by the vibration signal within the entire cycle, and the horizontal line indicates the shock signal threshold. And, Figure 2 An example of an impact fragment is shown. Based on this, it can be determined that Figure 2 The full cycle vibration signal in includes 9 impact segments.

[0119] In another possible implementation, the envelope corresponding to the full-cycle vibration signal may be determined first, and then compared with the impact signal threshold to determine the aforementioned quantity threshold.

[0120] In a specific implementation, the first envelope of the entire cycle corresponding to the first vibration signal of the entire cycle is determined, and then the number of intersections between the first envelope of the entire cycle and the reference line identified by the impact signal threshold is determined as the first number. The moment corresponding to the intersection is the impact moment, that is, the intersection can indicate the moment when the impact starts and ends, so the intersection can indicate the impact segment. Therefore, the quantity threshold can be determined based on the first number. Correspondingly, the second envelope of the entire cycle corresponding to the second vibration signal of the entire cycle is determined, and then the number of intersections between the second envelope of the entire cycle and the reference line identified by the impact signal threshold is determined as the second number, and the number of actual impact segments is determined based on the second number.

[0121] For ease of understanding, the present application provides the following examples: Figure 3 A schematic diagram of a first envelope of a whole cycle as shown, and Figure 4 A schematic diagram of a second envelope of a full cycle is shown, specifically:

[0122] exist Figure 3 In the example, the blue waveform line represents the first vibration signal S(t) of the entire cycle, and the orange line is the first envelope H(t) of the entire cycle. Figure 4 In the example, the blue waveform line represents the second vibration signal S(t) of the entire period, and the orange line is the second envelope H(t) of the entire period.

[0123] In specific implementation, the vibration acceleration signal envelope can be calculated based on the vibration signal based on the Hilbert envelope method, and the corresponding full-cycle envelope can be obtained accordingly. The full-cycle envelope can represent the envelope signal of the full-cycle vibration signal.

[0124] For further understanding, Figure 3 Based on the first envelope curve of the full cycle of the example, the following is provided: Figure 5 A schematic diagram of determining the threshold value of the number of normal impact fragments is shown, and Figure 4Based on the second envelope curve of the full cycle of the example, the following is provided: Figure 6 A schematic diagram of determining the number of actual impact fragments is shown, specifically:

[0125] exist Figure 5 In the example, the dashed box takes the first impact segment as an example to show two intersections between the first envelope of the entire cycle and the reference line identified by the impact signal threshold. The first intersection indicates the moment when the impact starts, and the second intersection indicates the moment when the impact ends. Therefore, two intersections can determine an impact segment. Similarly, after determining the first number of intersections (i.e., the total number of intersections), the number threshold of normal impact segments can be determined. Figure 5 In the example, only two intersection points are shown. The other intersection points are similar and will not be described here. For an impact, the end points of the start and end of the impact can be used as the impact identification time of the impact, that is, Figure 5 As shown in the red dot mark, the number of red dot marks is the number of impact fragments included, which is the aforementioned number threshold. Figure 5 In this example, the quantity threshold is 9.

[0126] Similarly, in Figure 6 In the example, 10 impact recognition moments can be determined, that is, the number of the aforementioned actual impact segments is 10.

[0127] Accordingly, it can be determined Figure 6 In the state to be tested, there is an abnormal impact, which can be the impact marked at the fourth impact identification moment. If it is not repaired in time, the target engine will continue to work for a period of time, and then continue to test to get the following Figure 7 The impact diagram of a knock fault in the late stage is shown in FIG. 1 , specifically:

[0128] Compared to Figure 6 For the abnormal impact marked by the fourth impact moment, the signal strength of the impact signal increases, that is, compared with Figure 6 The detection, in Figure 7 During the test, the knocking fault of the target engine became more serious. Figure 6 The detected knock fault is called an early knock fault. Figure 7 The knocking fault detected is called a late knocking fault. It can be seen that after adopting the present application, the knocking fault can be detected during the operation of the engine, and the early knocking fault can be discovered in time, and the user can be reminded to repair it in time, so as to avoid more serious knocking faults and other risks.

[0129] For the above example, in order to ensure the accuracy of recognition during specific implementation, when identifying two adjacent impacts, the interval duration can be set to be greater than or equal to the duration threshold, wherein the interval duration can refer to the time interval between the end impact moment of the first impact and the start impact moment of the second impact of the two adjacent impacts, and the duration threshold can be determined based on the duration of the time domain, for example, it can be 0.5 thousandths of the normalized duration of the time domain. Based on this, misidentification caused by abnormal sensor acquisition is avoided. If the time interval between the end impact moment of the first impact and the start impact moment of the second impact is less than the duration threshold, the two impacts are identified as one impact.

[0130] It is precisely because the excitation generated by the knocking fault is relatively weak, and the corresponding vibration signal is relatively small. Therefore, it is difficult to identify this abnormal impact that appears as a weak impact by using the method of setting the impact signal threshold artificially in the related art. In addition, the related art also detects whether there is a knocking fault by statistically analyzing the excitation characteristics of the entire cycle vibration signal (such as peak value, effective value, kurtosis, skewness, impact factor, waveform factor, etc.), but under variable working conditions, the working conditions will affect the intensity of the excitation, that is, the size of the vibration signal is different, making it more difficult to set an accurate impact signal threshold, and the excitation generated by the knocking fault is relatively weak, so the detection sensitivity of this method in the related art to the excitation generated by the knocking fault is very low.

[0131] After adopting the present application, it is possible to adaptively determine the appropriate shock signal threshold according to the current target working condition to ensure accuracy, and based on this statistical shock number, detect whether there is an abnormal shock. Due to the existence of knocking faults, additional and new shocks are generated in the entire cycle vibration signal impulse, which will directly affect the number of shocks, and the correlation between the number of shocks and the working conditions is weak. Based on this, the present application's method of using the number of statistical shocks to detect whether there are abnormal shocks can better be compatible with detection under variable working conditions and improve the detection sensitivity of potential knocking fault problems of the engine. In this way, manual settings are no longer required, human subjective bias is avoided, and it is no longer dependent on expert experience, etc., so the demand for expert experience is lower.

[0132] In addition, if the detection determines that the number of actual impact fragments is less than the number threshold, it means that the number of actual impacts is less than the number of impacts that should be generated. In this case, it can also be determined that there is an abnormality and generate a warning prompt information, which can be used to indicate that the actual number of impacts in the entire cycle is lower than the number threshold. In actual applications, this may be because the engine itself has a malfunction (such as a fire, etc.), resulting in the impact that should have been generated not being generated. Therefore, a warning is issued for this situation so that timely maintenance can be carried out to avoid further damage to the engine.

[0133] In practical applications, some control strategies can also be set to shield special situations. For example, during the engine startup and shutdown process, the engine has not yet been ignited and the number of impacts is normally reduced. Therefore, no detection is performed during these processes, thereby avoiding the alarm that the actual number of impacts is less than the number threshold and saving computer resources.

[0134] And, in practical applications, the detection method provided by the present application can be embedded in a real-time monitoring system (such as a real-time monitoring system for the operation of a ship's engine) so that the engine vibration signal collected by the sensor can be continuously acquired, and the number of shocks in the current operation process can be calculated, and then it can be determined whether it exceeds the normal shock number threshold. Once exceeded, an early warning can be triggered to remind the ship's old track. It can be understood that this example implementation method can also be used when detecting the engines of other means of transportation.

[0135] Figure 8 This is a structural diagram of an engine impact detection device provided in an embodiment of the present application, wherein the device includes an acquisition unit 801 and a determination unit 802:

[0136] The acquisition unit 801 is used to acquire a first vibration signal of a whole period generated during the operation of the target engine in a normal state under the target working condition;

[0137] The determining unit 802 is used to determine the effective value of the noise signal according to the first vibration signal of the entire cycle, and the correlation relationship between the first vibration signal of the entire cycle, the normal impact signal, and the noise signal, wherein the effective value of the noise signal is used to indicate the signal strength characteristics of the noise signal in the time domain, and the correlation relationship is used to indicate that the signal value of the first vibration signal of the entire cycle at the tth moment includes the normal impact signal value at the tth moment and the noise signal value at the tth moment, and the tth moment is any moment in the time domain of the entire cycle;

[0138] The determination unit 802 is further configured to determine a threshold value of the number of normal impact segments included in the target engine under the target operating condition according to the effective value of the noise signal and the first vibration signal of the entire period;

[0139] The acquisition unit 801 is further used to acquire a full-cycle second vibration signal generated during the operation of the target engine in the to-be-detected state under the target operating condition;

[0140] The determining unit 802 is further configured to determine the number of actual impact segments included in the target engine under the target operating condition according to the effective value of the noise signal and the full-cycle second vibration signal;

[0141] The determining unit 802 is further configured to determine that the target engine has abnormal impacts within the entire cycle if the number of the actual impact fragments exceeds the number threshold.

[0142] In a possible implementation manner, if the noise signal effective value is used to indicate a signal strength mean value of the noise signal in the time domain, the determining unit is further used to:

[0143] Determining a noise signal correction coefficient corresponding to the target operating condition, wherein the noise signal correction coefficient is greater than 1;

[0144] The product of the noise signal correction coefficient and the effective value of the noise signal is determined as an impulse signal threshold, where the impulse signal threshold is used to indicate the maximum signal strength of the noise signal in the time domain;

[0145] Determining the quantity threshold according to the impact signal threshold and the first vibration signal of the entire period;

[0146] The number of the actual impact segments is determined according to the impact signal threshold and the full-cycle second vibration signal.

[0147] In a possible implementation, the noise signal correction coefficient is less than a coefficient threshold, the impulse signal threshold is greater than the maximum signal strength of the noise segment in the time domain and does not exceed a strength threshold.

[0148] In a possible implementation manner, the determining unit is further configured to:

[0149] According to the impact signal threshold, dividing the first vibration signal of the entire period into a plurality of first signal segments;

[0150] If the signal amplitudes in the first signal segment are all greater than or equal to the impact signal threshold, the first signal segment is determined as the normal impact segment, and the quantity threshold is obtained;

[0151] According to the impact signal threshold, dividing the full-cycle second vibration signal into a plurality of second signal segments;

[0152] If the signal amplitudes in the second signal segment are all greater than or equal to the impact signal threshold, the second signal segment is determined as the actual impact segment, and the number of the actual impact segments is obtained.

[0153] In a possible implementation manner, the determining unit is further configured to:

[0154] Determine a first envelope of a whole period corresponding to the first vibration signal of the whole period;

[0155] Determine the number of intersections between the first envelope of the entire period and the reference line identified by the impact signal threshold as a first number, and the moment corresponding to the intersection is the impact moment;

[0156] determining the quantity threshold according to the first quantity;

[0157] Determine a second envelope of an entire period corresponding to the second vibration signal of the entire period;

[0158] Determine the number of intersections between the second envelope of the entire period and the reference line identified by the impulse signal threshold as a second number;

[0159] The number of actual impact fragments is determined based on the second number.

[0160] In a possible implementation manner, the determining unit is further configured to:

[0161] S(t)=A(t)+N(t)

[0162] In the above formula, S(t) is used to represent the signal value of the first vibration signal of the entire period at the t-th moment, A(t) is used to represent the normal impact signal value at the t-th moment, N(t) is used to represent the noise signal value at the t-th moment, t∈[0, t1], t1 is used to represent the time length of the time domain;

[0163] According to the first vibration signal of the entire period and the association relationship, the following formula is constructed in the time domain:

[0164]

[0165] Let the effective value of the noise signal M0 = N(t), and construct the following formula in the time domain:

[0166]

[0167] In summary, construct the following expression:

[0168]

[0169] The zero point of R(M) is solved, and the value of M corresponding to the zero point of R(M) is determined as the effective value M0 of the noise signal.

[0170] In a possible implementation manner, the noise signal effective value is used to indicate a root mean square value of a signal strength of the noise signal in the time domain.

[0171] In a possible implementation manner, the determining unit is further configured to:

[0172] In a detection period including multiple whole cycles, if it is determined that the proportion of the whole cycles in which the abnormal impact exists is greater than the proportion threshold, an early warning prompt information is generated, and the early warning prompt information is used to indicate that the target engine has a knocking fault, and the number proportion is the ratio between the number of whole cycles in which the abnormal impact exists and the total number of whole cycles in the detection period.

[0173] It can be seen from the above technical solution that since the vibration signal generated during the operation of the engine includes a normal impact signal and a certain noise signal, and when running in a stable working condition, the noise signal does not change much in the whole cycle. Therefore, for the target engine, firstly, the first vibration signal of the whole cycle generated during the operation under the target working condition in a normal state, and the correlation between the first vibration signal of the whole cycle, the normal impact signal, and the noise signal are used to determine the effective value of the noise signal. Among them, the correlation is used to indicate that the signal value of the first vibration signal of the whole cycle at the tth moment includes the normal impact signal value at the tth moment and the noise signal value at the tth moment, and the tth moment is any moment in the time domain of the whole cycle. And, the effective value of the noise signal is used to indicate the signal strength characteristics of the noise signal in the time domain, that is, the effective value of the noise signal can indicate the size of the noise signal of the target engine in the target working condition in the whole cycle. Then, according to the effective value of the noise signal and the first vibration signal of the whole cycle, the threshold value of the number of normal impact segments included in the target engine under the target working condition is determined. And, the whole cycle second vibration signal generated during the operation of the target engine in the state to be detected under the target working condition is obtained, and the number of actual impact segments included in the target engine under the target working condition is determined according to the effective value of the noise signal and the whole cycle second vibration signal. It can be seen that the number threshold is used to indicate the normal impact situation included in the whole cycle of the target engine under the normal state and the target working condition, and the number of actual impact segments is used to indicate the actual impact situation included in the whole cycle of the target engine under the state to be detected and the same target working condition. Therefore, if the number of actual impact segments exceeds the number threshold, it indicates that in the actual operation process, impacts other than normal impacts are generated, such as additional impacts caused by knocking failures, so it is determined that the target engine has abnormal impacts in the whole cycle.

[0174] Compared with related technologies, it is no longer necessary to manually set the impact signal threshold, which can avoid human subjective bias and improve detection accuracy. In addition, since the effective value of the noise signal is determined under the target working condition and in the normal state, and in the application stage, the detection is also performed on the state to be detected under the same target working condition, because the working conditions before and after are consistent, it is possible to take into account the impact of the changing working conditions on the generated impact, so that it can be applied to the detection of various working conditions and has higher applicability.

[0175] In another aspect, an embodiment of the present application provides a computer device, the computer device comprising a processor and a memory:

[0176] The memory is used to store program code and transmit the program code to the processor;

[0177] The processor is used to execute the engine impact detection method provided by the above embodiment according to the instructions in the program code.

[0178] The computer device may include a terminal device or a server, and the aforementioned engine impact detection device may be configured in the computer device.

[0179] On the other hand, an embodiment of the present application further provides a storage medium, wherein the storage medium is used to store a computer program, and the computer program is used to execute the engine impact detection method provided in the above embodiment.

[0180] In addition, an embodiment of the present application also provides a computer program product including instructions, which, when executed on a computer, enables the computer to execute the engine impact detection method provided in the above embodiment.

[0181] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, the above program can be stored in a computer-readable storage medium, and when the program is executed, it executes the steps of the above method embodiment; and the above storage medium can be at least one of the following media: read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc., various media that can store program codes.

[0182] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative work.

[0183] It should be noted that, in this article, relational terms such as "first" and "second", etc. (if any) are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0184] The above is a detailed introduction to an engine impact detection method and related devices provided by the embodiment of the present application. The principles and implementation methods of the present application are described in detail using specific examples. The description of the above embodiments is only used to help understand the method of the present application. At the same time, for those skilled in the art, according to the method of the present application, there will be changes in the specific implementation method and application scope.

[0185] In summary, the content of this specification should not be understood as limiting the present application, and any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in this application should be included in the protection scope of this application. Moreover, based on the implementations provided in the above aspects, this application can also be further combined to provide more implementations.

Claims

1. An engine impact detection method, characterized in that: The method comprises: Acquire a first vibration signal of a whole period generated during the operation of the target engine in a normal state under the target working condition; Determine the effective value of the noise signal according to the first vibration signal of the entire cycle, and the correlation relationship between the first vibration signal of the entire cycle, the normal impact signal, and the noise signal, wherein the effective value of the noise signal is used to indicate the signal strength characteristics of the noise signal in the time domain, and the correlation relationship is used to indicate that the signal value of the first vibration signal of the entire cycle at the tth moment includes the normal impact signal value at the tth moment and the noise signal value at the tth moment, and the tth moment is any moment in the time domain of the entire cycle; Determining a threshold value of the number of normal impact segments included in the target engine under the target operating condition according to the effective value of the noise signal and the first vibration signal of the entire period; Acquire a full-cycle second vibration signal generated during the operation of the target engine in the to-be-detected state under the target operating condition; Determining the number of actual impact fragments included in the target engine under the target operating condition according to the effective value of the noise signal and the full-cycle second vibration signal; If the number of the actual impact segments exceeds the number threshold, it is determined that the target engine has abnormal impacts within the entire cycle.

2. The method according to claim 1, characterized in that If the noise signal effective value is used to indicate the signal strength mean value of the noise signal in the time domain, the method further includes: Determining a noise signal correction coefficient corresponding to the target operating condition, wherein the noise signal correction coefficient is greater than 1; The product of the noise signal correction coefficient and the effective value of the noise signal is determined as an impulse signal threshold, where the impulse signal threshold is used to indicate the maximum signal strength of the noise signal in the time domain; The step of determining the number threshold of normal impact segments included in the target engine under the target operating condition according to the effective value of the noise signal and the first vibration signal of the entire period includes: Determining the quantity threshold according to the impact signal threshold and the first vibration signal of the entire period; The step of determining the number of actual impact segments included in the target engine under the target operating condition according to the effective value of the noise signal and the full-cycle second vibration signal comprises: The number of the actual impact segments is determined according to the impact signal threshold and the full-cycle second vibration signal.

3. The method according to claim 2, characterized in that The noise signal correction coefficient is less than the coefficient threshold, the impulse signal threshold is greater than the maximum signal strength of the noise segment in the time domain and does not exceed the strength threshold.

4. The method according to claim 2, characterized in that: The determining the quantity threshold according to the impact signal threshold and the first vibration signal of the entire period includes: According to the impact signal threshold, dividing the first vibration signal of the entire period into a plurality of first signal segments; If the signal amplitudes in the first signal segment are all greater than or equal to the impact signal threshold, the first signal segment is determined as the normal impact segment, and the quantity threshold is obtained; The determining the number of the actual impact segments according to the impact signal threshold and the full-cycle second vibration signal comprises: According to the impact signal threshold, dividing the full-cycle second vibration signal into a plurality of second signal segments; If the signal amplitudes in the second signal segment are all greater than or equal to the impact signal threshold, the second signal segment is determined as the actual impact segment, and the number of the actual impact segments is obtained.

5. The method according to claim 2, characterized in that: The determining the quantity threshold according to the impact signal threshold and the first vibration signal of the entire period includes: Determine a first envelope of an entire period corresponding to the first vibration signal of the entire period; Determine the number of intersections between the first envelope of the entire period and the reference line identified by the impact signal threshold as a first number, and the moment corresponding to the intersection is the impact moment; determining the quantity threshold according to the first quantity; The determining the number of the actual impact segments according to the impact signal threshold and the full-cycle second vibration signal comprises: Determine a second envelope of an entire period corresponding to the second vibration signal of the entire period; Determine the number of intersections between the second envelope of the entire period and the reference line identified by the impulse signal threshold as a second number; The number of actual impact fragments is determined based on the second number.

6. The method according to claim 1, characterized in that The association relationship is expressed by the following formula: S(t)=A(t)+N(t) In the above formula, S(t) is used to represent the signal value of the first vibration signal of the entire period at the t-th moment, A(t) is used to represent the normal impact signal value at the t-th moment, N(t) is used to represent the noise signal value at the t-th moment, t∈[0, t1], t1 is used to represent the time length of the time domain; The determining the effective value of the noise signal according to the first vibration signal of the entire period, and the correlation relationship between the first vibration signal of the entire period, the normal impact signal, and the noise signal, comprises: According to the first vibration signal of the entire period and the association relationship, the following formula is constructed in the time domain: Let the effective value of the noise signal M0 = N(t), and construct the following formula in the time domain: In summary, construct the following expression: The zero point of R(M) is solved, and the value of M corresponding to the zero point of R(M) is determined as the effective value M0 of the noise signal.

7. The method according to claim 6, characterized in that The noise signal effective value is used to indicate the root mean square value of the signal strength of the noise signal in the time domain.

8. The method according to any one of claims 1 to 7, characterized in that: The method further comprises: In a detection period including multiple whole cycles, if it is determined that the proportion of the whole cycles in which the abnormal impact exists is greater than the proportion threshold, an early warning prompt information is generated, and the early warning prompt information is used to indicate that the target engine has a knocking fault, and the number proportion is the ratio between the number of whole cycles in which the abnormal impact exists and the total number of whole cycles in the detection period.

9. An engine impact detection device, characterized in that: The device comprises an acquisition unit and a determination unit: The acquisition unit is used to acquire a first vibration signal of a whole period generated during the operation of the target engine in a normal state under the target working condition; The determining unit is used to determine the effective value of the noise signal according to the first vibration signal of the entire cycle, and the correlation relationship between the first vibration signal of the entire cycle, the normal impact signal, and the noise signal, wherein the effective value of the noise signal is used to indicate the signal strength characteristics of the noise signal in the time domain, and the correlation relationship is used to indicate that the signal value of the first vibration signal of the entire cycle at the tth moment includes the normal impact signal value at the tth moment and the noise signal value at the tth moment, and the tth moment is any moment in the time domain of the entire cycle; The determination unit is further used to determine a threshold value of the number of normal impact segments included in the target engine under the target operating condition according to the effective value of the noise signal and the first vibration signal of the entire period; The acquisition unit is further used to acquire a full-cycle second vibration signal generated during the operation of the target engine in the to-be-detected state under the target operating condition; The determination unit is further used to determine the number of actual impact fragments included in the target engine under the target operating condition according to the effective value of the noise signal and the full-cycle second vibration signal; The determining unit is further configured to determine that the target engine has abnormal impacts within the entire cycle if the number of the actual impact fragments exceeds the number threshold.

10. A computer device, characterized in that: The computer device comprises a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is configured to execute the method according to any one of claims 1 to 8 according to the instructions in the program code.

Images