A method, system, and medium for temperature monitoring of a heat dissipating ignition coil

Abstract

The present application relates to the field of temperature measurement, in particular to a temperature monitoring method and system of a heat dissipation ignition coil and a medium, comprising: reducing the coil temperature and the radiator temperature into a monitoring vector, and obtaining the equilibrium temperature of the monitoring vector; when the coil temperature is greater than a preset early warning temperature, reducing the coil temperature and the radiator temperature into an abnormal vector, and recording the latest obtained monitoring vector as a normal vector; recording several equilibrium temperatures that present an increasing trend in time sequence among the equilibrium temperatures of all monitoring vectors as reference equilibrium temperatures, obtaining a first dimension and a second dimension according to the continuity and discreteness of the value of each dimension of all monitoring vectors when the reference equilibrium temperature changes; and performing temperature early warning by using the difference between the abnormal vector and the normal vector in the first dimension and the second dimension. The present application avoids the problem of false early warning caused by the carbon and dust accumulation of the radiator.

Description

Technical Field

[0001] The present invention relates to the technical field of temperature measurement, and in particular to a temperature monitoring method, system and medium for a heat dissipation ignition coil. Background Art

[0002] As a critical vehicle component, the cooling ignition coil requires timely and accurate temperature monitoring and temperature warnings, crucial for ensuring normal operation. Conventional real-time temperature monitoring measures the internal coil temperature. When the coil temperature exceeds a set warning temperature, a warning is issued, alerting the driver that the vehicle cannot operate normally and requiring prompt parking or repairs to prevent coil failure (such as a short circuit) from causing the cooling ignition coil to burn out.

[0003] However, unlike standard ignition coils, heat-dissipating ignition coils contain a heat sink. Carbon and dust deposits can prevent the heat generated by the coil from being dissipated quickly enough, leading to internal heat accumulation and potentially triggering a warning, potentially even damaging the coil. Both radiator failure (carbon and dust buildup) and coil failure can cause the coil temperature to exceed the set warning temperature, triggering a warning. Warnings triggered by radiator failure are false alarms and can affect the vehicle's normal operation. Summary of the Invention

[0004] In order to solve the problem of false alarm caused by carbon and dust accumulation in the radiator, the present invention provides a temperature monitoring method, system and medium for a heat dissipation ignition coil.

[0005] The temperature monitoring method, system and medium of a heat dissipation ignition coil of the present invention adopt the following technical solutions:

[0006] An embodiment of the present invention provides a method for monitoring the temperature of a heat dissipation ignition coil, the method comprising the following steps:

[0007] The temperature sensors installed inside and outside the heat dissipation ignition coil are used to collect the coil temperature and radiator temperature respectively;

[0008] During the operation of the heat dissipation ignition coil, whenever the coil temperature and the radiator temperature both reach their maximum values, the coil temperature and the radiator temperature collected in the same time period are reduced to monitoring vectors of several dimensions. The coil temperature when the temperature reaches the maximum value is recorded as the equilibrium temperature of the monitoring vector;

[0009] When the coil temperature is higher than the preset warning temperature, the coil temperature and radiator temperature collected in the same time period are reduced to an abnormal vector, and the most recently obtained monitoring vector is recorded as a normal vector.

[0010] Among the equilibrium temperatures of all monitoring vectors, several equilibrium temperatures that show an increasing trend in time sequence are recorded as reference equilibrium temperatures. The continuity and discreteness of the values ​​of each dimension in all monitoring vectors as they change with the reference equilibrium temperature are obtained, and the dimensions with the largest difference in continuity and discreteness are recorded as target dimensions. Among the target dimensions, the dimension with greater continuity than discreteness is recorded as the first dimension, and the dimension with greater discreteness than or equal to continuity is recorded as the second dimension.

[0011] When the first difference between the abnormal vector and the normal vector in the first dimension is greater than or equal to the second difference between the abnormal vector and the normal vector in the second dimension, a radiator fault warning is issued; when the first difference is less than the second difference, a coil fault warning is issued.

[0012] Preferably, the method for determining when both the coil temperature and the radiator temperature have reached maximum values ​​includes:

[0013] For any one of the coil temperature and the radiator temperature, and for any moment, the standard deviation of the temperature at several moments before the moment is obtained. When the standard deviation is less than or equal to a first preset threshold, it is determined that the maximum temperature is reached at the moment.

[0014] Preferably, the step of reducing the dimension of the coil temperature and the radiator temperature collected in the same time period includes the following specific steps:

[0015] The sequences of coil temperature and radiator temperature obtained in the same time period are recorded as coil temperature sequence and radiator temperature sequence respectively, and the coil temperature sequence and radiator temperature sequence are spliced ​​into a spliced ​​sequence; all spliced ​​sequences obtained during the operation of the heat dissipation ignition coil are subjected to PCA dimensionality reduction, and the dimensionality reduction result obtained for each spliced ​​sequence is used as a monitoring vector or anomaly vector.

[0016] Preferably, the equilibrium temperatures of all the monitoring vectors that show an increasing trend in time sequence are recorded as reference equilibrium temperatures, which includes the following specific steps:

[0017] The equilibrium temperatures are arranged in the order in which they are obtained to obtain an equilibrium temperature curve, wherein the horizontal coordinate of each coordinate point in the equilibrium temperature curve is the arrangement sequence, and the vertical coordinate is the equilibrium temperature;

[0018] Obtain any minimum point in the equilibrium temperature curve and record it as the first reference point. Obtain all coordinate points whose ordinate difference with the ordinate of the first reference point is greater than a preset difference and record them as a candidate point set. In the candidate point set, the coordinate point whose abscissa is greater than the abscissa of the first reference point and whose abscissa difference with the first reference point is the smallest is recorded as the second reference point.

[0019] Obtain all coordinate points whose ordinate difference with the ordinate of the second reference point is greater than a preset difference, and record them as a candidate point set again. The coordinate point in the candidate point set whose abscissa is greater than the abscissa of the second reference point and whose abscissa difference with the second reference point is the smallest is recorded as the third reference point;

[0020] And so on, until the candidate point set is empty, all equilibrium temperatures corresponding to all reference points are recorded as temporary reference equilibrium temperatures of any minimum point; for the temporary reference equilibrium temperatures of all minimum points in the equilibrium temperature curve, the temporary reference equilibrium temperature containing the most equilibrium temperatures is recorded as the reference equilibrium temperature.

[0021] Preferably, the step of obtaining the continuity and discreteness of the values ​​of each dimension in all monitoring vectors as they change with the reference equilibrium temperature includes the following specific steps:

[0022] For two monitoring vectors corresponding to any two adjacent reference equilibrium temperatures, the average difference in the values ​​of the two monitoring vectors in each dimension is recorded as the equilibrium temperature change interference;

[0023] The continuity is negatively correlated with the equilibrium temperature change interference amount, and the discreteness is positively correlated with the equilibrium temperature change interference amount.

[0024] Preferably, the specific steps of obtaining the first difference between the abnormal vector and the normal vector in the first dimension, and the second difference between the abnormal vector and the normal vector in the second dimension are as follows:

[0025] For any first dimension, obtain the absolute value of the difference between the abnormal vector and the normal vector in the first dimension, and record it as the abnormal value of any first dimension;

[0026] Normalizing the continuity of all first dimensions to obtain the abnormal attention of each first dimension, and using the abnormal attention of each first dimension to perform weighted summation on the abnormal values ​​of all first dimensions to obtain the first difference;

[0027] For any second dimension, obtain the absolute value of the difference between the abnormal vector and the normal vector in the second dimension, and record it as the abnormal value of any second dimension;

[0028] The discreteness of all second dimensions is normalized to obtain the abnormal attention degree of each second dimension, and the abnormal attention degree of each second dimension is used to perform weighted summation on the abnormal values ​​of all second dimensions to obtain the second difference.

[0029] Preferably, the continuity is negatively correlated with the equilibrium temperature change interference amount, and the discreteness is positively correlated with the equilibrium temperature change interference amount, and the specific steps included are as follows:

[0030] The result of arranging the reference equilibrium temperatures from small to large is recorded as the first sequence. The sequence consisting of the values ​​of the same dimension in all monitoring vectors corresponding to all reference equilibrium temperatures in the first sequence is recorded as the second sequence. The absolute value of the Pearson correlation coefficient between the first sequence and the second sequence is recorded as f.

[0031] Let (1-f)×B be the discreteness and f×(1-B) be the continuity, where B is the equilibrium temperature change disturbance.

[0032] Preferably, the specific steps for obtaining the equilibrium temperature change interference amount are as follows:

[0033] For any two adjacent reference equilibrium temperatures, the difference in the values ​​of the two monitoring vectors corresponding to the two reference equilibrium temperatures in each dimension is recorded as the interference amplitude;

[0034] For all adjacent reference equilibrium temperatures, the interference amplitudes corresponding to all reference equilibrium temperatures are linearly normalized, and the mean value of the normalized interference amplitudes is recorded as the equilibrium temperature change interference amount;

[0035] Any two adjacent reference equilibrium temperatures are: two adjacent reference equilibrium temperatures in the results of arranging the reference equilibrium temperatures from small to large.

[0036] Another embodiment of the present invention provides a temperature monitoring system for a heat dissipation ignition coil. The system includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. The system also includes a heat dissipation ignition coil. The heat dissipation ignition coil includes a radiator. Temperature sensors are respectively installed inside and on the outer surface of the heat dissipation ignition coil. When the processor runs the computer program, all steps of the above-mentioned method for monitoring the temperature of a heat dissipation ignition coil are executed.

[0037] Another embodiment of the present invention provides a temperature monitoring medium for a heat dissipation ignition coil. The medium is a storage medium for storing temperature data collected by a temperature sensor in the temperature monitoring system for the heat dissipation ignition coil.

[0038] The beneficial effects of the technical solution of the present invention are:

[0039] The present invention incorporates temperature sensors mounted inside and outside the heat-dissipating ignition coil to collect coil and radiator temperatures, respectively. During operation, whenever the coil and radiator temperatures reach their maximum values, the collected coil and radiator temperatures are recorded. This combines both coil and radiator temperature considerations, rather than solely considering coil temperature, enabling more detailed and reliable temperature warnings.

[0040] Furthermore, the present invention records several equilibrium temperatures among all monitoring vectors that exhibit a time-sequential increasing trend as reference equilibrium temperatures. The continuity and discreteness of the values ​​of each dimension of all monitoring vectors as they change with the reference equilibrium temperatures are then determined. The dimensions with the largest difference in continuity and discreteness are designated as target dimensions. Dimensions within the target dimensions where continuity is greater than discreteness are designated as first dimensions, and those where discreteness is greater than or equal to continuity are designated as second dimensions. This process analyzes the relative changes in coil and radiator temperatures when the internal and external surfaces of the heat-dissipating ignition coil reach thermal equilibrium, as well as the evolution of these relative changes when the internal and external surfaces of the heat-dissipating ignition coil reach thermal equilibrium, to determine the first and second dimensions associated with radiator faults. This allows for distinguishing between coil and radiator faults in the heat-dissipating ignition coil.

[0041] Based on this, the present invention issues a radiator fault warning when the first difference between the abnormal vector and the normal vector in the first dimension is greater than or equal to the second difference between the abnormal vector and the normal vector in the second dimension. When the first difference is less than the second difference, a coil fault warning is issued. This process uses the changes in the abnormal vector (temperature data when the temperature is abnormal) and the normal vector (temperature data when the temperature is normal) in the first dimension related to the radiator fault and the second dimension related to the coil fault to provide different warnings when the coil temperature exceeds the preset warning temperature. This avoids the potential for false warnings that can occur when relying solely on coil temperature warnings. Such false warnings can alert the driver (or even force the driver to make an emergency stop) when the radiator is experiencing carbon and dust accumulation but the coil is functioning normally. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 A flow chart of the steps of a temperature monitoring method for a heat dissipation ignition coil provided by one embodiment of the present invention;

[0044] Figure 2 This is the curve of the coil temperature and radiator temperature when the ignition coil is not faulty;

[0045] Figure 3 It is the normalized change curve of coil temperature and radiator temperature when the ignition coil has no faults;

[0046] Figure 4 The curves of coil temperature and radiator temperature change when carbon deposition occurs on the radiator;

[0047] Figure 5 It is the normalized change curve of coil temperature and radiator temperature when carbon deposition occurs on the radiator;

[0048] Figure 6 The curves of coil temperature and radiator temperature change when a warning is issued due to simultaneous coil fault and radiator fault;

[0049] Figure 7 It is the normalized change curve of coil temperature and radiator temperature when the early warning is issued due to the simultaneous existence of coil fault and radiator fault;

[0050] Figure 8 The distribution diagrams of the first dimension and the second dimension at different reference equilibrium temperatures for multiple vehicles;

[0051] Figure 9 The distribution diagram of the first dimension and the second dimension at different reference equilibrium temperatures for one of the vehicles. DETAILED DESCRIPTION

[0052] To further illustrate the technical means and effectiveness of the present invention in achieving its intended objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, provides a detailed description of the specific implementation, structure, features, and effectiveness of a method, system, and medium for monitoring the temperature of a heat-dissipating ignition coil according to the present invention. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0053] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0054] The following describes in detail a temperature monitoring method, system, and medium for a heat dissipation ignition coil provided by the present invention with reference to the accompanying drawings.

[0055] Example 1:

[0056] One embodiment of the present invention provides a temperature monitoring system for a heat dissipation ignition coil. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The system also includes a heat dissipation ignition coil for applying a high voltage to a vehicle's spark plug, thereby igniting the vehicle's engine. The heat dissipation ignition coil comprises a primary coil, a secondary coil, and an iron core mounted within a housing. The primary coil and the secondary coil have a large turns ratio. When the primary coil is powered, a strong magnetic field is generated around it as the current increases, and the iron core stores the magnetic field energy. When a switching device disconnects the primary coil circuit, the magnetic field in the primary coil rapidly decays, and a high voltage is induced in the secondary coil. The faster the primary coil's magnetic field disappears, the greater the current at the moment of disconnection, and the greater the turns ratio between the two coils, the higher the voltage induced in the secondary coil.

[0057] In this embodiment, the heat-dissipating ignition coil also includes a heat sink for dissipating heat from the coil (including the primary and secondary coils) and other components within the housing. In this embodiment, the heat sink is constructed from a metal fin embedded in the housing, with the fin exposed externally. The heat sink in the ignition coil housing disclosed in CN204117819U includes the heat sink described in this embodiment.

[0058] The specific structure of the heat dissipation ignition coil is a well-known technology, and its specific structure and principle will not be described in detail here.

[0059] In addition, temperature sensors are installed inside and on the outer shell surface of the heat dissipation ignition coil of this embodiment, and the collected temperatures are recorded as the coil temperature and the radiator temperature respectively.

[0060] When the processor runs the computer program, all steps of a temperature monitoring method for a heat dissipation ignition coil are executed.

[0061] Example 2:

[0062] This embodiment provides a temperature monitoring medium for a heat dissipating ignition coil, which is specifically a storage medium for storing the coil temperature and the radiator temperature collected by the temperature sensor in the first embodiment.

[0063] As an example, the storage medium is a solid-state hard drive installed in the vehicle.

[0064] Example 3:

[0065] See also Figure 1 , which shows a flow chart of a method for monitoring the temperature of a heat dissipation ignition coil according to an embodiment of the present invention, the method comprising the following steps:

[0066] Step S301 : During the operation of the heat dissipation ignition coil, whenever the coil temperature and the radiator temperature both reach maximum values, a coil temperature sequence and a radiator temperature sequence consisting of the coil temperature and the radiator temperature collected in the same time period are obtained.

[0067] When the vehicle is ignited, the heat dissipation ignition coil begins operating, and the temperature sensor reads the coil and radiator temperatures in real time. In this embodiment, the coil and radiator temperatures are collected simultaneously at one-second intervals. In other embodiments, the temperature sensor can also collect data at one-minute intervals to reduce the amount of data collected.

[0068] Since the heat dissipation ignition coil continuously generates heat during operation and dissipates it through the heat dissipation ignition coil radiator, the coil temperature and the heat sink temperature gradually increase during this process. In this embodiment, under normal circumstances, the coil temperature generally rises to 140-150°C and the radiator temperature generally rises to 60-70°C.

[0069] As a comparative example, the temperature monitoring method of the heat dissipation ignition coil is as follows:

[0070] Set a warning temperature (for example, 180°C). When the coil temperature exceeds the warning temperature, an alarm will sound to remind the driver that the vehicle cannot drive normally and the cooling ignition coil needs to be replaced (or the driver needs to stop for maintenance) to prevent the cooling ignition coil from continuing to rise in temperature and burning out, or causing other vehicle failures and traffic accidents.

[0071] However, this comparative example has the following problems:

[0072] With the long-term operation or use of the heat dissipation ignition coil, the heat dissipation ignition coil may malfunction. The malfunctions described in this embodiment include coil malfunction and radiator malfunction; the coil malfunction is specifically a coil short circuit malfunction, which will cause the coil temperature to rise to a higher temperature, or even continue to rise until it burns out; the radiator malfunction is specifically carbon or dust accumulation on the radiator, which leads to a decrease in heat dissipation capacity, and in turn, the heat dissipation ignition coil cannot dissipate heat, causing the heat accumulation to cause the coil temperature to rise to a higher temperature or even burn out.

[0073] Since the failure of the heat dissipation ignition coil includes both coil failure and radiator failure, the above comparative embodiment only considers the coil temperature. When the coil temperature is greater than the warning temperature, it is not necessarily caused by coil short circuit and overheating, but is caused by radiator failure, or both. At this time, the coil temperature being greater than the warning temperature does not mean that the heat dissipation ignition coil will continue to heat up or burn out, or cause other vehicle failures. At this time, replacing the heat dissipation ignition coil or reminding the driver to stop will reduce the driving experience or cause unnecessary economic losses.

[0074] In order to solve the problems of the above comparative embodiment, this embodiment first monitors the coil temperature and radiator temperature of the heat dissipation ignition coil at the same time when the coil is working. The specific process is:

[0075] Each time the vehicle is started, whenever the coil and radiator temperatures reach their maximum values, a coil temperature sequence and a radiator temperature sequence are acquired during the same time period. These temperature sequences are then recorded as a temperature sequence group. This means that each time the coil and radiator temperatures reach their maximum values, a temperature sequence group is generated. Over the long-term use of the vehicle, multiple temperature sequence groups can be generated. Subsequent temperature monitoring using these multiple temperature sequence groups can avoid the aforementioned issues associated with considering only coil temperature magnitude in the comparative embodiment.

[0076] A temperature sequence group is obtained when the coil temperature and the radiator temperature reach their maximum values. The purpose of this is to take into account that when the coil temperature and the radiator temperature reach their maximum values, it indicates that the coil and the radiator have reached thermal equilibrium within a local time period, or that the coil has completed a process of generating heat, transferring heat to the radiator for heat dissipation, and finally the temperature no longer changes. The temperature sequence groups obtained for two consecutive times represent the temperature changes when the thermal equilibrium state of the coil and the radiator changes. This situation is likely caused by a failure of the heat dissipation ignition coil. Based on this, this embodiment can further provide reliable early warnings for heat dissipation ignition coil failures by subsequently analyzing the same temperature sequence group and different temperature sequence groups, distinguishing whether the heat dissipation ignition coil failure is caused by a radiator failure or a coil failure; or when both failures exist at the same time, which failure to give a warning for is preferred; thereby assisting the vehicle driving and maintenance process and avoiding unnecessary losses.

[0077] It should be noted that a temperature sequence group is obtained each time the coil temperature and the radiator temperature reach a maximum value. This means that when both the coil temperature and the radiator temperature reach a certain maximum value, if the maximum value continues to remain unchanged, then there is no need to repeatedly obtain the temperature sequence group. Only when one of the coil temperature and the radiator temperature is no longer at the maximum value and both reach the maximum value again at the same time, will the temperature sequence group continue to be obtained.

[0078] As an example, the method for determining whether a maximum value is reached is:

[0079] For any one of the coil temperature and the radiator temperature, at any moment, the standard deviation of the temperature at several moments before that moment (for example, 10 moments, including the aforementioned arbitrary moment) is obtained. When the standard deviation is less than or equal to the first preset threshold th1, it indicates that the temperature has reached a maximum value at that moment; in other words, the coil temperature or the radiator temperature has reached a maximum value at that moment.

[0080] Additionally, for the moment when the coil temperature reaches the maximum value, the average of the coil temperature at several moments before the moment is taken as the equilibrium temperature.

[0081] The embodiment takes th1=2 Celsius degrees as an example for description, and th1 can be set to other values in other embodiments, which are not limited in the embodiment.

[0082] As an example, the coil temperature sequence and the radiator temperature sequence collected in the same time period are obtained, including the method that:

[0083] After each start of the vehicle, when the coil temperature and the radiator temperature both reach the maximum value, the coil temperature sequence and the radiator temperature sequence are obtained, including the method that:

[0084] Step S302, dimension reduction of the coil temperature sequence and the radiator temperature sequence collected in the same time period into a plurality of dimensions of monitoring vectors.

[0085] First of all, it needs to be pointed out that when the heat dissipation ignition coil is faulty (including coil fault and radiator fault), the fault will not only change the change rule between the coil temperature sequence and the radiator temperature sequence collected in the same time period, but also change the change rule between the coil temperature sequence and the radiator temperature sequence before and after the fault occurs.

[0086] For example, as shown in Figure 2 , which shows the coil temperature and radiator temperature change during the ignition start process of the heat dissipation ignition coil, when the heat dissipation ignition coil is not faulty, the coil temperature will be relatively slowly increased to the temperature maximum value (for example, 150°C), and there is a clear temperature difference between the surface of the radiator fin and the internal coil, which can dissipate the heat generated by the coil and avoid the heat generated by the coil from accumulating to cause high temperature and burn the coil. At this time, the change rule between the coil temperature sequence and the radiator temperature sequence is that both the coil temperature sequence and the radiator temperature sequence gradually increase to the temperature maximum value, and the radiator temperature sequence has obvious lag characteristics compared with the coil temperature sequence, as shown in Figure 3 , Figure 3 is the linear normalization result of Figure 2 .

[0087] For another example, when there is only a coil short circuit in the heat dissipation ignition coil, the change pattern between the coil temperature sequence and the radiator temperature sequence is: the coil temperature sequence will change to a higher temperature value faster, and the radiator temperature sequence will also change to a higher temperature value faster. The radiator temperature sequence has a significant lag compared to the coil temperature sequence. When there is only carbon deposit on the radiator in the heat dissipation ignition coil, the change pattern between the coil temperature sequence and the radiator temperature sequence is: in the process of gradually increasing to the maximum temperature value, the radiator temperature sequence will have the characteristic of approaching the coil temperature sequence. At the same time, due to the heat accumulation caused by untimely heat dissipation, the coil temperature sequence and the radiator temperature sequence will rise to a higher temperature, such as Figure 4 、 5 As shown, Figure 5 Yes Figure 4 The linear normalization result of .

[0088] When the heat dissipation ignition coil has both coil fault and radiator fault, the variation pattern between the coil temperature sequence and the radiator temperature sequence is a combination of the above variation patterns. At this time, when the coil temperature exceeds the warning temperature (180°C) (that is, when the temperature is monitored using the above comparative embodiment), as shown in FIG. Figure 6 、 7 As shown, Figure 7 Yes Figure 6 The linear normalization result cannot be determined whether it is caused by a coil failure or a radiator failure, which leads to the problems existing in the above comparative embodiment.

[0089] In this regard, this embodiment further reduces the dimensions of the coil temperature sequence and the radiator temperature sequence based on the coil temperature sequence and the radiator temperature sequence in the same time period, and records the reduced vectors as detection vectors.

[0090] The purpose of dimensionality reduction is to extract and describe the variation patterns between the coil temperature series and the radiator temperature series.

[0091] As an example, the dimensionality reduction of the coil temperature series and the radiator temperature series includes the following methods:

[0092] The coil temperature series and radiator temperature series obtained during the same time period are concatenated into a single sequence, referred to as a concatenated sequence. All concatenated sequences obtained during the operation of the heat dissipation ignition coil are subjected to PCA dimensionality reduction, and the dimensionality reduction result of each concatenated sequence is used as a monitoring vector. In this embodiment, each concatenated sequence is reduced to a 20-dimensional monitoring vector.

[0093] In other embodiments, before concatenating the coil and radiator temperature sequences, they can each be downsampled to reduce the computational complexity of PCA dimensionality reduction. This embodiment uses a Gaussian pyramid algorithm for downsampling (e.g., downsampling 6 times). The downsampled coil and radiator temperature sequences are then concatenated into a concatenated sequence, which is then dimensionality-reduced into a monitoring vector.

[0094] As another example, the dimensionality reduction of the coil temperature series and the heat sink temperature series includes the following methods:

[0095] For the coil temperature sequence and radiator temperature sequence obtained in the same time period, the coil temperature sequence and the radiator temperature sequence are divided into several subsequences of the same length (for example, divided into 5 subsequences of the same length); for the i-th subsequence in the coil temperature sequence , and the ith subsequence in the heat sink temperature sequence , get the subsequence and subsequence Pearson correlation coefficient; then get the subsequence The average value of the subsequence The difference between the average values ​​of Relative to subsequence lag time.

[0096] As an example, get the subsequence Relative to subsequence Latency, including:

[0097] The coil temperature sequence and the radiator temperature sequence are linearly normalized respectively, and the normalized coil temperature sequence and the radiator temperature sequence are divided into several subsequences of the same length.

[0098] For subsequence and The two temperatures T1 and T2 have the same value, where T1 is In the In the example, the difference between the acquisition time of T2 and the acquisition time of T1 is obtained and recorded as and The time difference between two temperatures with the same value; and The average of the time differences for all temperatures with the same value is recorded as the lag time.

[0099] In particular, if the coil temperature sequence or the radiator temperature sequence cannot be equally divided into several subsequences of the same length, zeros are added after the coil temperature sequence or the radiator temperature sequence so that the coil temperature sequence or the radiator temperature sequence can be exactly equally divided.

[0100] The Pearson correlation coefficient is used to describe whether the coil temperature sequence and the radiator temperature sequence have an approaching trend in the local time, and the smaller the absolute value of the Pearson correlation coefficient is, the more likely the radiator temperature sequence approaches the coil temperature sequence, and the larger the absolute value of the Pearson correlation coefficient is, the more likely the approaching trend does not exist. The difference between the average values describes whether the coil and the radiator have a large temperature difference in the local time. The lag time describes the length of time that the radiator temperature lags behind the coil temperature in the local time. In this embodiment, the Pearson correlation coefficient, the difference between the average values, and the lag time obtained by each sub-sequence are taken as a vector of each sub-sequence, and the vectors of all sub-sequences are spliced together in order as a feature vector, and the feature vector is taken as a monitoring vector obtained after dimension reduction.

[0101] In some other embodiments, for each feature vector obtained during the operation of the ignition coil, all the feature vectors are subjected to PCA dimension reduction, and each dimension reduction result is taken as a monitoring vector.

[0102] In this embodiment, all the obtained monitoring vectors are subjected to ZCA whitening processing and then unitization, so as to eliminate the order of magnitude difference of different dimensions.

[0103] In addition, as described in step S301, each time the coil temperature and the radiator temperature reach a maximum value, a balance temperature is obtained, and therefore each monitoring vector corresponds to a balance temperature, which is the maximum value of the coil temperature and represents the temperature of the coil when the ignition coil is in a certain fault (or no fault) during continuous operation.

[0104] In step S303, when the coil temperature is greater than the preset warning temperature, a reference balance temperature is obtained, and a target dimension is obtained according to the continuity and discreteness of the value of each dimension of all monitoring vectors when the reference balance temperature changes.

[0105] When the coil temperature is greater than the warning temperature, the embodiment performs temperature monitoring according to all the obtained monitoring vectors. Each monitoring vector contains the change rule between the coil temperature and the radiator temperature, and the difference between different monitoring vectors contains the change or evolution of the change rule between the coil temperature and the radiator temperature.

[0106] When the coil temperature is greater than the warning temperature, for all the obtained monitoring vectors, a plurality of balance temperatures that show an increasing trend in time sequence among the balance temperatures of all the monitoring vectors are recorded as reference balance temperatures. The reference balance temperature describes a balance temperature whose value gradually increases with the change of time in the use process of the vehicle (or in the long-term operation process of the ignition coil).

[0107] It's important to note that the equilibrium temperature represents the temperature at which the coil and radiator temperatures reach thermal equilibrium. A change in the equilibrium temperature indicates that the two temperatures have lost and then regained thermal equilibrium. This loss of thermal equilibrium is caused by a failure in the heat dissipation ignition coil, typically a significant coil failure (e.g., a short circuit) or a radiator failure (e.g., increased carbon deposits). Therefore, subsequent analysis will be based on the reference equilibrium temperature.

[0108] Furthermore, the continuity and discreteness of the values ​​of each dimension of all monitoring vectors as they change with the reference equilibrium temperature are obtained.

[0109] The variation between the coil and heat sink temperatures, as captured by the monitoring vectors, is specifically reflected in some of the dimensions within the monitoring vectors. The changes in the values ​​of these dimensions across different monitoring vectors describe the evolution of the variation between the coil and heat sink temperatures. Therefore, analyzing how each dimension of all monitoring vectors changes with the reference equilibrium temperature is crucial for temperature monitoring of heat-dissipating ignition coils.

[0110] In this embodiment, the variation of each dimension of all monitoring vectors with the reference equilibrium temperature specifically includes the continuity of the value of each dimension when the reference equilibrium temperature changes, and the discreteness of the value of each dimension when the reference equilibrium temperature changes.

[0111] The continuity of each dimension's value as the reference equilibrium temperature changes (referred to as the continuity of each dimension) describes whether the evolution of the change pattern between the coil temperature and the heat sink temperature is continuous as the reference equilibrium temperature gradually increases; the larger the value, the more continuous the evolution of the change pattern. The discreteness of each dimension's value as the reference equilibrium temperature changes (referred to as the discreteness of each dimension) describes whether the evolution of the change pattern between the coil temperature and the heat sink temperature is abrupt and without a transition process as the reference equilibrium temperature gradually increases; the larger the value, the more abrupt the change pattern is or whether it is a transitional change.

[0112] Furthermore, ash and carbon accumulation on radiators evolve gradually and continuously, increasing slowly and steadily. Therefore, the values ​​for the dimension encompassing ash and carbon accumulation are continuous. In contrast, coil short-circuit failures are sudden, with no gradual transition from the onset to the onset of the short-circuit. Therefore, the values ​​for the dimension encompassing coil short-circuit failures are abrupt.

[0113] The N1 dimensions with the largest difference between continuity and discreteness (i.e., the absolute value of the difference between continuity and discreteness) are recorded as target dimensions. The larger the difference between continuity and discreteness, the more the dimension can be used to clearly distinguish between a coil fault and a radiator fault. In other words, the larger the dimension, the more it represents either one fault or the other. Therefore, the obtained target dimension can clearly distinguish the two faults when the two faults exist at the same time.

[0114] In this embodiment, N1 is equal to half of the detection vector dimension (rounded up). In other embodiments, N1 may be set to other values, which is not specifically limited in this embodiment.

[0115] As an example, among the equilibrium temperatures of all monitoring vectors, several equilibrium temperatures that show an increasing trend in time sequence are recorded as reference equilibrium temperatures, including the following methods:

[0116] The equilibrium temperatures are arranged in the order in which they are obtained to obtain an equilibrium temperature curve, wherein the horizontal coordinate of each coordinate point in the curve is the arrangement sequence, and the vertical coordinate is the equilibrium temperature;

[0117] Obtain any minimum point in the equilibrium temperature curve and record it as the first reference point. Obtain all coordinate points whose ordinate difference with the ordinate of the first reference point is greater than the preset difference t1 and record them as the candidate point set. The coordinate point in the candidate point set whose abscissa is greater than the abscissa of the first reference point and whose abscissa difference with the first reference point is the smallest is recorded as the second reference point.

[0118] Next, all coordinate points whose ordinate difference with the ordinate of the second reference point is greater than t1 are obtained and recorded as the candidate point set. The coordinate point in the candidate point set whose abscissa is greater than the abscissa of the second reference point and whose abscissa difference with the second reference point is the smallest is recorded as the third reference point.

[0119] This process is repeated in this way until the candidate point set is empty, and all equilibrium temperatures corresponding to all reference points are recorded as temporary reference equilibrium temperatures.

[0120] Then, for all the minimum points in the equilibrium temperature curve, a temporary reference equilibrium temperature can be obtained using the above method, and the temporary reference equilibrium temperature containing the most equilibrium temperatures is recorded as the reference equilibrium temperature.

[0121] This embodiment is described by taking t1 = 5 degrees Celsius as an example. In other embodiments, t1 may be set to other values, which is not specifically limited in this embodiment.

[0122] As an example, the continuity of the value of each dimension when it changes with the reference equilibrium temperature, and the discreteness of the value of each dimension when it changes with the reference equilibrium temperature are obtained as follows:

[0123] For any two adjacent reference equilibrium temperatures (that is, two adjacent reference equilibrium temperatures in the result of arranging the reference equilibrium temperatures from small to large), the two reference equilibrium temperatures correspond to two monitoring vectors respectively, and the difference in the values ​​of the two monitoring vectors in each dimension (that is, the absolute value of the difference) is obtained. The difference describes the change in each dimension of the monitoring vector each time the reference equilibrium temperature changes (that is, when the thermal balance of the heat dissipation ignition coil changes due to a fault). That is, for any two adjacent reference equilibrium temperatures, there is a corresponding difference as described above in each dimension (recorded as the interference amplitude of each dimension). For all adjacent reference equilibrium temperatures, there are multiple interference amplitudes in each dimension. These interference amplitudes are linearly normalized, and the average value of the normalized interference amplitude is recorded as the equilibrium temperature change interference amount, which describes the average change (or average difference) of each dimension in the monitoring vector each time the reference equilibrium temperature changes (that is, when the thermal balance of the heat dissipation ignition coil changes due to a fault). The smaller the value, the gradual and continuous change of each dimension with the change of the reference equilibrium temperature; the larger the value, the obvious and significant change of each dimension with the change of the reference equilibrium temperature (that is, each dimension presents a discrete distribution characteristic with the change of the reference equilibrium temperature).

[0124] As an optional example, the equilibrium temperature change disturbance is recorded as B, B is used as the discreteness, and 1-B is used as the continuity.

[0125] As another example, the result of arranging the reference equilibrium temperatures from small to large is recorded as the first sequence, and the sequence consisting of the values ​​of the same dimension in all monitoring vectors corresponding to all reference equilibrium temperatures in the first sequence is recorded as the second sequence. The absolute value of the Pearson correlation coefficient between the first sequence and the second sequence is recorded as f. The larger the f, the more obvious the trend of change in the value of each dimension as the reference equilibrium temperature increases. At this time, the value of each dimension changes gradually and continuously with the reference equilibrium temperature. The smaller the f, the less the value of each dimension does not change with the change of the reference equilibrium temperature as the reference equilibrium temperature increases, or the change in the value of each dimension does not change with the gradual increase of the reference equilibrium temperature. For example, the sudden change in the value of each dimension leads to a discrete distribution of values.

[0126] Furthermore, (1-f)×B is regarded as the discreteness, and f×(1-B) is regarded as the continuity.

[0127] This preferred example further accurately obtains the discreteness and continuity.

[0128] So far, this embodiment has obtained the discreteness and continuity of each dimension.

[0129] Step S304: Obtain the abnormal vector and normal vector when the coil temperature is greater than the preset warning temperature. The dimension in the target dimension where continuity is greater than discreteness is recorded as the first dimension, and the dimension where discreteness is greater than or equal to continuity is recorded as the second dimension. The difference between the abnormal vector and the normal vector in the first dimension and the difference between the abnormal vector and the normal vector in the second dimension are used to perform temperature warning.

[0130] The dimension in which continuity is greater than discreteness in the target dimension is recorded as the first dimension, and the dimension in which discreteness is greater than or equal to continuity is recorded as the second dimension.

[0131] Since the first dimension is relatively continuous, it mainly contains radiator fault information of the heat dissipation ignition coil. Since the second dimension is relatively discrete, it mainly contains coil fault information of the heat dissipation ignition coil.

[0132] like Figure 8 、 9 As shown, it shows the distribution of the first dimension and the second dimension in the monitoring vector at different reference equilibrium temperatures (ie, 145, 150, 155, 160, 170, 175°C).

[0133] Specifically, Figure 8 shows the distribution of monitoring vectors obtained from the startup process of multiple vehicles in a first dimension and a second dimension; Figure 9 The distribution of the first dimension and the second dimension in the monitoring vector of the heat dissipation ignition coil of one vehicle in this embodiment is shown.

[0134] Depend on Figure 8 、 9 It can be seen that the value of the monitoring vector in the first dimension changes continuously with the increase of the reference equilibrium temperature, and changes significantly in the second dimension with the increase of the reference equilibrium temperature.

[0135] When the coil temperature is greater than the warning temperature and reaches the maximum value most recently, the monitoring vector obtained is recorded as the normal vector.

[0136] When the coil temperature is greater than the warning temperature, the coil temperature and radiator temperature collected in the same time period are reduced in dimension into abnormal vectors of several dimensions (the specific method is the same as step S302 ).

[0137] A first difference between the abnormal vector and the normal vector in a first dimension is obtained, and a second difference between the abnormal vector and the normal vector in a second dimension is obtained. When the first difference is greater than or equal to the second difference, a radiator fault warning is issued. When the first difference is less than the second difference, a coil fault warning is issued.

[0138] In other embodiments, a radiator fault warning is issued when the ratio of the first difference to the second difference is greater than or equal to the threshold th2, and a coil fault warning is issued when the ratio of the first difference to the second difference is less than the threshold th2; this embodiment is described using th2=1 as an example, and in other embodiments, th2 can be set to other values, which is not specifically limited in this embodiment.

[0139] In particular, when the vehicle is started, the coil temperature does not reach the maximum value, but directly rises to the warning temperature. If a coil fault warning was issued when the vehicle was last used and the coil was not replaced, a coil fault warning will be issued at this time, otherwise a radiator fault warning will be issued directly.

[0140] Among them, the first difference is greater than or equal to the second difference, indicating that the value of the normal vector in the first dimension has undergone a relatively large change, indicating that the reason why the coil temperature is greater than the preset warning temperature is more likely due to a radiator failure; the first difference is less than the second difference, indicating that the value of the normal vector in the second dimension has undergone a relatively large change, indicating that the reason why the coil temperature is greater than the preset warning temperature is more likely due to a collar failure.

[0141] When a radiator fault warning is issued, it means that the coil temperature is too high due to carbon deposits on the radiator of the heat dissipation ignition coil. However, the coil can still work normally. At this time, the driver should first clear the carbon deposits on the radiator to eliminate the fault. When a coil fault warning is issued, it means that the coil temperature is too high due to a short circuit fault in the heat dissipation ignition coil. In this case, the heat dissipation ignition coil needs to be replaced in time. It is not recommended to continue driving before the heat dissipation ignition coil is replaced in time.

[0142] In other embodiments, other factors may contribute to coil temperatures exceeding the preset warning temperature, such as a substandard heat dissipation ignition coil housing (e.g., low-quality, easily deformed materials) or excessive voltage in the heat dissipation ignition coil's power supply circuit. Therefore, in other embodiments, when a heat dissipation ignition coil exhibits a fault such as deformation or discoloration, replacement of the heat dissipation ignition coil is prioritized. When a radiator fault warning is issued, the vehicle should be operated only after correcting other faults.

[0143] It should be noted that after the fault is eliminated by clearing the carbon deposits and dust deposits on the radiator, when the vehicle is driven again, if the coil temperature is again greater than the preset warning temperature, a warning is still issued using the above method of this embodiment.

[0144] As an optional example, obtaining the first difference between the abnormal vector and the normal vector in the first dimension includes the following methods:

[0145] For any first dimension, the absolute value of the difference between the abnormal vector and the normal vector in the first dimension is obtained, and the average of the absolute values ​​of the differences between all abnormal vectors and the normal vectors in all first dimensions is recorded as the first difference.

[0146] As a preferred example, obtaining the first difference between the abnormal vector and the normal vector in the first dimension includes the following method:

[0147] For any first dimension, obtain the absolute value of the difference between the abnormal vector and the normal vector in the first dimension, and record it as the abnormal value of any first dimension;

[0148] The continuity of all first dimensions is normalized using the softmax formula to obtain the abnormal attention of each first dimension. The abnormal attention of each first dimension is used to perform weighted summation on the abnormal values ​​of all first dimensions, and the result is used as the first difference.

[0149] As an optional example, obtaining the second difference between the abnormal vector and the normal vector in the second dimension includes the following methods:

[0150] For any second dimension, the absolute value of the difference between the abnormal vector and the normal vector in the second dimension is obtained, and the average of the absolute values ​​of the differences between all abnormal vectors and the normal vectors in all second dimensions is recorded as the second difference.

[0151] As a preferred example, obtaining the second difference between the abnormal vector and the normal vector in the second dimension includes the following method:

[0152] For any second dimension, obtain the absolute value of the difference between the abnormal vector and the normal vector in the second dimension, and record it as the abnormal value of any second dimension;

[0153] The discreteness of all second dimensions is normalized using the softmax formula to obtain the abnormal attention of each second dimension. The abnormal attention of each second dimension is used to perform weighted summation on the abnormal values ​​of all second dimensions, and the result is used as the second difference.

[0154] As an example, the method for obtaining the first difference and the second difference described above further includes:

[0155] For all monitoring vectors (including abnormal vectors and normal vectors), obtain the absolute value of the difference between any two monitoring vectors in any first dimension; and record the average of the absolute values ​​of the differences of all monitoring vectors in all first dimensions as A1.

[0156] For all monitoring vectors (including abnormal vectors and normal vectors), obtain the absolute value of the difference between any two monitoring vectors in any second dimension; and record the average of the absolute values ​​of the differences of all monitoring vectors in all second dimensions as A2.

[0157] The ratio of the first difference obtained in the above optional example or preferred example to A1 is taken as the first difference obtained in this example; the ratio of the first difference obtained in the above optional example or preferred example to A2 is taken as the second difference obtained in this example.

[0158] The purpose of this example is to eliminate the difference in magnitude between the first difference and the second difference.

[0159] Example 4:

[0160] The coil temperature series and the radiator temperature series are reduced in dimension, including the following methods:

[0161] Multiple temperature sequence groups are acquired from a heat dissipation ignition coil that is operating without any faults. The method for acquiring the temperature sequence groups is described in Example 3. For example, a heat dissipation ignition coil that is operating without any faults is installed on different vehicles. After these vehicles are started, several temperature sequence group samples (e.g., 10 temperature sequence groups) are acquired.

[0162] Next, multiple temperature sequence groups were collected from operating cooling ignition coils with varying degrees of soot or carbon deposits. Several fault-free cooling ignition coils were artificially treated with varying degrees of soot or carbon deposits and installed in different vehicles. Several temperature sequence group samples (e.g., 10) were also obtained after these vehicles were started.

[0163] Similarly, multiple temperature sequence groups (for example, 10 temperature sequence groups) collected when the heat dissipation ignition coil with coil short circuit is in operation are obtained.

[0164] All the temperature sequence groups obtained above are used as sample sets, and PCA dimensionality reduction is performed on the coil temperature sequences and radiator temperature sequences contained in all the temperature sequence groups in these sample sets. The specific method is described in step S302 of the third embodiment.

[0165] The difference from step S302 of the third embodiment is that this embodiment does not directly obtain the dimensionality reduction result, but obtains the feature vector in the dimensionality reduction process.

[0166] The following explains the eigenvectors used in the dimensionality reduction process: According to the PCA dimensionality reduction algorithm, the algorithm first obtains several mutually orthogonal eigenvectors for the coil and radiator temperature sequences in the sample set. The projections of the coil and radiator temperature sequences onto each eigenvector are then taken. The projections of the coil and radiator temperature sequences contained in each temperature sequence group onto all eigenvectors form the dimensionality reduction results for the coil and radiator temperature sequences in each temperature sequence group. For example, when the dimensionality is reduced to a 20-dimensional vector, the PCA dimensionality reduction process yields 20 mutually orthogonal eigenvectors.

[0167] This process is well known and will not be demonstrated in detail in this embodiment. In this embodiment, instead of obtaining a dimensionality reduction result, several mutually orthogonal eigenvectors obtained in the above PCA dimensionality reduction process are obtained.

[0168] This embodiment stores these feature vectors in the solid-state hard disk of the vehicle.

[0169] During vehicle operation, whenever the coil temperature and the radiator temperature both reach maximum values, the coil temperatures collected in the same time period constitute a coil temperature sequence, and the radiator temperatures constitute a radiator temperature sequence; the coil temperature sequence and the radiator temperature sequence are spliced ​​end to end into a spliced ​​sequence (the spliced ​​sequence is regarded as a vector), the stored eigenvectors are read out, and the projection of the spliced ​​sequence on each eigenvector is obtained. The projection of the spliced ​​sequence on all eigenvectors is used as the dimensionality reduction result of the coil temperature sequence and the radiator temperature sequence (i.e., the monitoring vector).

[0170] After the monitoring vector is obtained, only the monitoring vector is stored in the solid state drive, and the coil temperature sequence and the radiator temperature sequence are deleted and do not need to be stored in the solid state drive.

[0171] On the one hand, the dimensionality reduction method of this embodiment saves data storage space (that is, there is no need to store the coil temperature sequence and the radiator temperature sequence for a long time). On the other hand, since the temperature sequence group samples for obtaining the feature vector include the coil temperature sequences and the radiator temperature sequences collected under multiple faults, it avoids the situation where there is a large amount of noise interference in the monitoring vector. The noise interference refers to the presence of too many dimensions in the monitoring vector that cannot describe the coil fault and the radiator fault.

[0172] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for monitoring the temperature of a heat dissipation ignition coil, characterized in that: The method comprises the following steps: The temperature sensors installed inside and outside the heat dissipation ignition coil are used to collect the coil temperature and radiator temperature respectively; During the operation of the heat dissipation ignition coil, whenever the coil temperature and the radiator temperature both reach their maximum values, the coil temperature and the radiator temperature collected in the same time period are reduced to monitoring vectors of several dimensions. The coil temperature when the temperature reaches the maximum value is recorded as the equilibrium temperature of the monitoring vector; When the coil temperature is higher than the preset warning temperature, the coil temperature and radiator temperature collected in the same time period are reduced to an abnormal vector, and the most recently obtained monitoring vector is recorded as a normal vector. Among the equilibrium temperatures of all monitoring vectors, several equilibrium temperatures that show an increasing trend in time sequence are recorded as reference equilibrium temperatures. The continuity and discreteness of the values ​​of each dimension in all monitoring vectors as they change with the reference equilibrium temperature are obtained, and the dimensions with the largest difference in continuity and discreteness are recorded as target dimensions. Among the target dimensions, the dimension with greater continuity than discreteness is recorded as the first dimension, and the dimension with greater discreteness than or equal to continuity is recorded as the second dimension. When the first difference between the abnormal vector and the normal vector in the first dimension is greater than or equal to the second difference between the abnormal vector and the normal vector in the second dimension, a radiator fault warning is issued; when the first difference is less than the second difference, a coil fault warning is issued.

2. The method for monitoring the temperature of a heat dissipation ignition coil according to claim 1, characterized in that: The method for determining when both the coil temperature and the radiator temperature have reached maximum values ​​includes: For any one of the coil temperature and the radiator temperature, and for any moment, the standard deviation of the temperature at several moments before the moment is obtained. When the standard deviation is less than or equal to a first preset threshold, it is determined that the maximum temperature is reached at the moment.

3. The method for monitoring the temperature of a heat dissipation ignition coil according to claim 1, characterized in that: The specific steps of reducing the dimension of the coil temperature and the radiator temperature collected in the same time period are as follows: The sequences of coil temperature and radiator temperature obtained in the same time period are recorded as coil temperature sequence and radiator temperature sequence respectively, and the coil temperature sequence and radiator temperature sequence are spliced ​​into a spliced ​​sequence; all spliced ​​sequences obtained during the operation of the heat dissipation ignition coil are subjected to PCA dimensionality reduction, and the dimensionality reduction result obtained for each spliced ​​sequence is used as a monitoring vector or anomaly vector.

4. The method for monitoring the temperature of a heat dissipation ignition coil according to claim 1, characterized in that: The equilibrium temperatures of all the monitoring vectors that show an increasing trend in time sequence are recorded as reference equilibrium temperatures, and the specific steps are as follows: The equilibrium temperatures are arranged in the order in which they are obtained to obtain an equilibrium temperature curve, wherein the horizontal coordinate of each coordinate point in the equilibrium temperature curve is the arrangement sequence, and the vertical coordinate is the equilibrium temperature; Obtain any minimum point in the equilibrium temperature curve and record it as the first reference point. Obtain all coordinate points whose ordinate difference with the ordinate of the first reference point is greater than a preset difference and record them as a candidate point set. In the candidate point set, the coordinate point whose abscissa is greater than the abscissa of the first reference point and whose abscissa difference with the first reference point is the smallest is recorded as the second reference point. Obtain all coordinate points whose ordinate difference with the ordinate of the second reference point is greater than a preset difference, and record them as a candidate point set again. The coordinate point in the candidate point set whose abscissa is greater than the abscissa of the second reference point and whose abscissa difference with the second reference point is the smallest is recorded as the third reference point; And so on, until the candidate point set is empty, all equilibrium temperatures corresponding to all reference points are recorded as temporary reference equilibrium temperatures of any minimum point; for the temporary reference equilibrium temperatures of all minimum points in the equilibrium temperature curve, the temporary reference equilibrium temperature containing the most equilibrium temperatures is recorded as the reference equilibrium temperature.

5. The method for monitoring the temperature of a heat dissipation ignition coil according to claim 1, characterized in that: The specific steps of obtaining the continuity and discreteness of the values ​​of each dimension in all monitoring vectors as the reference equilibrium temperature changes are as follows: For two monitoring vectors corresponding to any two adjacent reference equilibrium temperatures, the average difference in the values ​​of the two monitoring vectors in each dimension is recorded as the equilibrium temperature change interference; The continuity is negatively correlated with the equilibrium temperature change interference amount, and the discreteness is positively correlated with the equilibrium temperature change interference amount.

6. The method for monitoring the temperature of a heat dissipation ignition coil according to claim 1, characterized in that: The specific steps of obtaining the first difference between the abnormal vector and the normal vector in the first dimension, and the second difference between the abnormal vector and the normal vector in the second dimension are as follows: For any first dimension, obtain the absolute value of the difference between the abnormal vector and the normal vector in the first dimension, and record it as the abnormal value of any first dimension; Normalizing the continuity of all first dimensions to obtain the abnormal attention of each first dimension, and using the abnormal attention of each first dimension to perform weighted summation on the abnormal values ​​of all first dimensions to obtain the first difference; For any second dimension, obtain the absolute value of the difference between the abnormal vector and the normal vector in the second dimension, and record it as the abnormal value of any second dimension; The discreteness of all second dimensions is normalized to obtain the abnormal attention degree of each second dimension, and the abnormal attention degree of each second dimension is used to perform weighted summation on the abnormal values ​​of all second dimensions to obtain the second difference.

7. The method for monitoring the temperature of a heat dissipation ignition coil according to claim 5, characterized in that: The continuity is negatively correlated with the equilibrium temperature change interference amount, and the discreteness is positively correlated with the equilibrium temperature change interference amount. The specific steps included are as follows: The result of arranging the reference equilibrium temperatures from small to large is recorded as the first sequence. The sequence consisting of the values ​​of the same dimension in all monitoring vectors corresponding to all reference equilibrium temperatures in the first sequence is recorded as the second sequence. The absolute value of the Pearson correlation coefficient between the first sequence and the second sequence is recorded as f. Let (1-f)×B be the discreteness and f×(1-B) be the continuity, where B is the equilibrium temperature change disturbance.

8. The method for monitoring the temperature of a heat dissipation ignition coil according to any one of claims 5 or 7, characterized in that: The specific steps for obtaining the equilibrium temperature change interference amount are as follows: For any two adjacent reference equilibrium temperatures, the difference in the values ​​of the two monitoring vectors corresponding to the two reference equilibrium temperatures in each dimension is recorded as the interference amplitude; For all adjacent reference equilibrium temperatures, the interference amplitudes corresponding to all reference equilibrium temperatures are linearly normalized, and the mean value of the normalized interference amplitudes is recorded as the equilibrium temperature change interference amount; Any two adjacent reference equilibrium temperatures are: two adjacent reference equilibrium temperatures in the results of arranging the reference equilibrium temperatures from small to large.

9. A temperature monitoring system for a heat dissipation ignition coil, the system comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor. The system further comprises a heat dissipation ignition coil, which comprises a radiator. The heat dissipation ignition coil is characterized in that temperature sensors are respectively installed on the interior and outer surface of the heat dissipation ignition coil. When the processor runs the computer program, all steps of a temperature monitoring method for a heat dissipation ignition coil according to any one of claims 1 to 8 are executed.

10. A temperature monitoring medium for a heat dissipation ignition coil, characterized in that: The medium is a storage medium for storing temperature data collected by a temperature sensor in a temperature monitoring system for a heat dissipation ignition coil as described in claim 9.

Images